Add files using upload-large-folder tool

output/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE48801.csv

@@ -0,0 +1,2 @@
+,GSM1184717,GSM1184718,GSM1184719,GSM1184720,GSM1184721,GSM1184722,GSM1184723,GSM1184724,GSM1184725,GSM1184726,GSM1184727,GSM1184728,GSM1184729,GSM1184730,GSM1184731,GSM1184732,GSM1184733,GSM1184734,GSM1184735,GSM1184736,GSM1184737,GSM1184738,GSM1184739,GSM1184740,GSM1184741,GSM1184742,GSM1184743,GSM1184744,GSM1184745,GSM1184746,GSM1184747,GSM1184748,GSM1184749,GSM1184750,GSM1184751,GSM1184752,GSM1184753,GSM1184754,GSM1184755,GSM1184756,GSM1184757,GSM1184758,GSM1184759,GSM1184760,GSM1184761,GSM1184762,GSM1184763,GSM1184764,GSM1184765,GSM1184766,GSM1184767,GSM1184768,GSM1184769,GSM1184770,GSM1184771,GSM1184772,GSM1184773,GSM1184774,GSM1184775,GSM1184776,GSM1184777,GSM1184778,GSM1184779,GSM1184780,GSM1184781,GSM1184782,GSM1184783,GSM1184784,GSM1184785,GSM1184786,GSM1184787,GSM1184788,GSM1184789,GSM1184790,GSM1184791,GSM1184792,GSM1184793,GSM1184794,GSM1184795,GSM1184796,GSM1184797,GSM1184798,GSM1184799,GSM1184800,GSM1184801,GSM1184802,GSM1184803,GSM1184804,GSM1184805,GSM1184806,GSM1184807,GSM1184808,GSM1184809,GSM1184810,GSM1184811,GSM1184812,GSM1184813,GSM1184814,GSM1184815,GSM1184816,GSM1184817,GSM1184818,GSM1184819,GSM1184820,GSM1184821,GSM1184822,GSM1184823,GSM1184824,GSM1184825,GSM1184826,GSM1184827,GSM1184828,GSM1184829,GSM1184830,GSM1184831,GSM1184832,GSM1184833,GSM1184834,GSM1184835,GSM1184836,GSM1184837,GSM1184838,GSM1184839,GSM1184840,GSM1184841,GSM1184842,GSM1184843,GSM1184844,GSM1184845,GSM1184846,GSM1184847,GSM1184848,GSM1184849,GSM1184850,GSM1184851,GSM1184852,GSM1184853,GSM1184854,GSM1184855,GSM1184856,GSM1184857,GSM1184858,GSM1184859,GSM1184860,GSM1184861,GSM1184862,GSM1184863,GSM1184864,GSM1184865,GSM1184866,GSM1184867,GSM1184868,GSM1184869,GSM1184870,GSM1184871,GSM1184872,GSM1184873,GSM1184874,GSM1184875,GSM1184876,GSM1184877,GSM1184878,GSM1184879,GSM1184880,GSM1184881,GSM1184882,GSM1184883,GSM1184884,GSM1184885,GSM1184886,GSM1184887,GSM1184888,GSM1184889,GSM1184890,GSM1184891,GSM1184892,GSM1184893,GSM1184894,GSM1184895
+Glucocorticoid_Sensitivity,90.2096916857165,90.2096916857165,92.0660852718675,92.0660852718675,85.8770390662799,85.8770390662799,87.4945143923344,87.4945143923344,85.1993812425936,85.1993812425936,84.9616236229156,84.9616236229156,83.9341340611542,83.9341340611542,88.7663927292959,88.7663927292959,88.4126127755346,88.4126127755346,90.1302355511097,90.1302355511097,86.3038207243861,86.3038207243861,97.9389927348314,97.9389927348314,85.6565800452145,85.6565800452145,72.080026977723,72.080026977723,95.7902581814721,95.7902581814721,84.7169700775247,84.7169700775247,97.2440363125325,97.2440363125325,98.6965291984436,98.6965291984436,96.3897437049292,96.3897437049292,93.7864779279733,93.7864779279733,88.9409584548941,88.9409584548941,95.2180128029044,95.2180128029044,80.3262384967705,80.3262384967705,98.9664822965928,98.9664822965928,86.7141270837215,86.7141270837215,94.1342236284511,94.1342236284511,76.5646360533747,76.5646360533747,94.4880035822124,94.4880035822124,84.2040871593034,84.2040871593034,81.2524330708547,81.2524330708547,75.0377332194718,75.0377332194718,103.111196853422,103.111196853422,93.7264007046898,93.7264007046898,98.4358920138007,98.4358920138007,91.1219245341963,91.1219245341963,89.7952307882158,89.7952307882158,100.164196369324,100.164196369324,92.2726878044167,92.2726878044167,83.653786832453,83.653786832453,85.4308536742686,85.4308536742686,95.9867474842918,95.9867474842918,97.4697626834784,97.4697626834784,87.1103581762748,87.1103581762748,106.335980304372,106.335980304372,95.0323274416373,95.0323274416373,93.2741255092367,93.2741255092367,88.0517452462257,88.0517452462257,92.7703808066373,92.7703808066373,90.2966860598886,90.2966860598886,90.2966860598886,87.6826035548426,87.6826035548426,87.6826035548426,110.820589380024,110.820589380024,110.820589380024,91.2861567746556,91.2861567746556,90.9575303422268,90.9575303422268,99.844023580098,99.844023580098,92.4380615886291,92.4380615886291,90.6279285533303,90.6279285533303,90.6279285533303,95.4061019654126,95.4061019654126,95.4061019654126,102.574377860977,102.574377860977,102.574377860977,83.3617762565561,83.3617762565561,92.9375020882722,92.9375020882722,83.056592777649,83.056592777649,101.239617979237,101.239617979237,86.5108726528178,86.5108726528178,87.8682889161097,87.8682889161097,89.9631142694748,89.9631142694748,95.5967828443558,95.5967828443558,94.3102962675651,94.3102962675651,97.0235772914671,97.0235772914671,94.6674807225407,94.6674807225407,86.0926581231368,86.0926581231368,107.862883138275,107.862883138275,82.7364199884225,82.7364199884225,88.2331356352063,88.2331356352063,93.614568433444,93.614568433444,103.717527073529,103.717527073529,103.717527073529,91.4503081788735,91.4503081788735,91.4503081788735,89.4567969526773,89.4567969526773,89.4567969526773,91.9430860155202,91.9430860155202,92.1077238348243,92.1077238348243,100.50157318001,100.50157318001,77.6379974012028,77.6379974012028,99.246829525294,99.246829525294,93.4438194050697,93.4438194050697,80.8101665274758,80.8101665274758,93.1053855695312,93.1053855695312

output/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE50012.csv

@@ -1,4 +1,3 @@
 ,GSM832137,GSM832138,GSM832139,GSM832140,GSM832141,GSM832142,GSM832143,GSM832144,GSM832145,GSM832146,GSM832147,GSM832148,GSM832149,GSM832150,GSM832151,GSM832152,GSM832153,GSM832154,GSM832155,GSM832156,GSM832157,GSM832158,GSM832159,GSM832160,GSM832161,GSM832162,GSM832163,GSM832164,GSM832165,GSM832166,GSM832167,GSM832168,GSM832169,GSM832170,GSM832171,GSM832172,GSM832173,GSM832174,GSM832175,GSM832176,GSM832177,GSM832178,GSM832179,GSM832180,GSM832181,GSM832182,GSM832183,GSM832184,GSM1212354,GSM1212355,GSM1212356,GSM1212357,GSM1212358,GSM1212359,GSM1212360,GSM1212361,GSM1212362,GSM1212363,GSM1212364,GSM1212365,GSM1212366,GSM1212367,GSM1212368,GSM1212369,GSM1212370,GSM1212371,GSM1212372,GSM1212373,GSM1212374,GSM1212375,GSM1212376,GSM1212377
-Glucocorticoid_Sensitivity,89.43486,89.43486,89.43486,89.43486,95.88507,95.88507,95.88507,95.88507,95.22036,95.22036,95.22036,95.22036,92.86704,92.86704,92.86704,92.86704,93.71633,93.71633,93.71633,93.71633,96.76962,96.76962,96.76962,96.76962,88.55031,88.55031,88.55031,88.55031,90.09957,90.09957,90.09957,90.09957,94.17097,94.17097,94.17097,94.17097,86.97089,86.97089,86.97089,86.97089,98.34904,98.34904,98.34904,98.34904,91.14896,91.14896,91.14896,91.14896,8.0,24.0,8.0,24.0,8.0,24.0,8.0,24.0,8.0,24.0,8.0,24.0,24.0,8.0,24.0,8.0,24.0,8.0,24.0,8.0,24.0,8.0,24.0,8.0
-Age,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,44.15,44.15,24.72,24.72,32.38,32.38,20.38,20.38,21.24,21.24,22.54,22.54,26.14,26.14,21.56,21.56,21.99,21.99,26.77,26.77,23.59,23.59,23.48,23.48
-Gender,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0
+Glucocorticoid_Sensitivity,89.43486,89.43486,89.43486,89.43486,95.88507,95.88507,95.88507,95.88507,95.22036,95.22036,95.22036,95.22036,92.86704,92.86704,92.86704,92.86704,93.71633,93.71633,93.71633,93.71633,96.76962,96.76962,96.76962,96.76962,88.55031,88.55031,88.55031,88.55031,90.09957,90.09957,90.09957,90.09957,94.17097,94.17097,94.17097,94.17097,86.97089,86.97089,86.97089,86.97089,98.34904,98.34904,98.34904,98.34904,91.14896,91.14896,91.14896,91.14896,,,,,,,,,,,,,,,,,,,,,,,,
+Age,44.15342,44.15342,44.15342,44.15342,24.72329,24.72329,24.72329,24.72329,32.37808,32.37808,32.37808,32.37808,20.38082,20.38082,20.38082,20.38082,21.2411,21.2411,21.2411,21.2411,22.54247,22.54247,22.54247,22.54247,26.13973,26.13973,26.13973,26.13973,21.5616,21.5616,21.5616,21.5616,21.9863,21.9863,21.9863,21.9863,26.76712,26.76712,26.76712,26.76712,23.59452,23.59452,23.59452,23.59452,23.47945,23.47945,23.47945,23.47945,,,,,,,,,,,,,,,,,,,,,,,,

output/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE57795.csv

@@ -1,2 +1,2 @@
-Sample_0_Sensitive patient,Sample_0_Resistant patient,Sample_1_Sensitive patient,Sample_1_Resistant patient
-1.0,0.0,1.0,0.0
+,GSM1388640,GSM1388641,GSM1388642,GSM1388643,GSM1388644,GSM1388645,GSM1388646,GSM1388647,GSM1388648,GSM1388649,GSM1388650,GSM1388651,GSM1388652,GSM1388653,GSM1388654,GSM1388655,GSM1388656,GSM1388657,GSM1388658,GSM1388659,GSM1388660,GSM1388661,GSM1388662,GSM1388663,GSM1388664,GSM1388665,GSM1388666,GSM1388667,GSM1388668,GSM1388669,GSM1388670,GSM1388671,GSM1388672,GSM1388673,GSM1388674,GSM1388675,GSM1388676,GSM1388677,GSM1388678,GSM1388679,GSM1388680,GSM1388681,GSM1388682,GSM1388683,GSM1388684,GSM1388685,GSM1388686,GSM1388687,GSM1388688,GSM1388689,GSM1388690,GSM1388691,GSM1388692,GSM1388693,GSM1388694,GSM1388695,GSM1388696,GSM1388697
+Glucocorticoid_Sensitivity,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0

output/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE58715.csv

@@ -1,2 +1,2 @@
-Sample1,Sample2
-1.0,0.0
+,GSM1417252,GSM1417253,GSM1417254,GSM1417255,GSM1417256,GSM1417257,GSM1417258,GSM1417259,GSM1417260,GSM1417261,GSM1417262,GSM1417263,GSM1417264,GSM1417265,GSM1417266,GSM1417267,GSM1417268,GSM1417269,GSM1417270,GSM1417271,GSM1417272,GSM1417273,GSM1417274,GSM1417275,GSM1417276,GSM1417277,GSM1417278,GSM1417279,GSM1417280,GSM1417281,GSM1417282,GSM1417283,GSM1417284,GSM1417285,GSM1417286,GSM1417287,GSM1417288,GSM1417289,GSM1417290,GSM1417291
+Glucocorticoid_Sensitivity,1.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,1.0,1.0,1.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,0.0,1.0,0.0,1.0,0.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,1.0,1.0,0.0,1.0,1.0,0.0

output/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE66705.csv

@@ -1,2 +1,2 @@
 ,GSM1629982,GSM1629983,GSM1629984,GSM1629985,GSM1629986,GSM1629987,GSM1629988,GSM1629989,GSM1629990,GSM1629991,GSM1629992,GSM1629993,GSM1629994,GSM1629995,GSM1629996,GSM1629997,GSM1629998,GSM1629999,GSM1630000,GSM1630001,GSM1630002,GSM1630003,GSM1630004,GSM1630005,GSM1630006,GSM1630007,GSM1630008,GSM1630009,GSM1630010,GSM1630011,GSM1630012,GSM1630013,GSM1630014,GSM1630015,GSM1630016,GSM1630017,GSM1630018,GSM1630019,GSM1630020,GSM1630021,GSM1630022,GSM1630023,GSM1630024,GSM1630025,GSM1630026,GSM1630027,GSM1630028,GSM1630029,GSM1630030,GSM1630031,GSM1630032,GSM1630033,GSM1630034,GSM1630035,GSM1630036,GSM1630037,GSM1630038,GSM1630039,GSM1630040,GSM1630041,GSM1630042,GSM1630043,GSM1630044,GSM1630045,GSM1630046,GSM1630047,GSM1630048,GSM1630049,GSM1630050,GSM1630051,GSM1630052,GSM1630053,GSM1630054,GSM1630055,GSM1630056,GSM1630057,GSM1630058,GSM1630059,GSM1630060,GSM1630061,GSM1630062,GSM1630063,GSM1630064,GSM1630065,GSM1630066,GSM1630067,GSM1630068,GSM1630069,GSM1630070,GSM1630071,GSM1630072,GSM1630073,GSM1630074,GSM1630075,GSM1630076,GSM1630077,GSM1630078,GSM1630079,GSM1630080,GSM1630081,GSM1630082,GSM1630083,GSM1630084,GSM1630085,GSM1630086,GSM1630087,GSM1630088,GSM1630089,GSM1630090,GSM1630091,GSM1630092,GSM1630093,GSM1630094,GSM1630095,GSM1630096,GSM1630097,GSM1630098,GSM1630099,GSM1630100,GSM1630101,GSM1630102,GSM1630103,GSM1630104,GSM1630105,GSM1630106,GSM1630107,GSM1630108,GSM1630109,GSM1630110,GSM1630111,GSM1630112,GSM1630113,GSM1630114,GSM1630115,GSM1630116,GSM1630117,GSM1630118,GSM1630119,GSM1630120,GSM1630121,GSM1630122,GSM1630123,GSM1630124,GSM1630125,GSM1630126,GSM1630127,GSM1630128,GSM1630129,GSM1630130,GSM1630131,GSM1630132,GSM1630133,GSM1630135,GSM1630137,GSM1630139,GSM1630142,GSM1630144,GSM1630146,GSM1630149,GSM1630151,GSM1630154,GSM1630156,GSM1630158,GSM1630160,GSM1630162,GSM1630163,GSM1630164,GSM1630165,GSM1630166,GSM1630167,GSM1630168
-Glucocorticoid_Sensitivity,,,,,1.0,,1.0,1.0,,,,,,0.0,,,,,,0.0,1.0,,,,,,,,0.0,0.0,,,1.0,,0.0,,,,,0.0,,,1.0,0.0,1.0,0.0,,0.0,0.0,,,1.0,,,,,0.0,,,,,,,,,0.0,1.0,0.0,,,,,1.0,,,,0.0,,,,0.0,0.0,1.0,,,,0.0,0.0,0.0,1.0,,0.0,,,,0.0,,,,1.0,,0.0,,1.0,1.0,,,,,0.0,,,0.0,0.0,,,1.0,1.0,,0.0,,0.0,0.0,1.0,1.0,0.0,0.0,1.0,0.0,0.0,,0.0,0.0,1.0,1.0,0.0,0.0,0.0,,0.0,0.0,1.0,0.0,,0.0,1.0,1.0,0.0,1.0,0.0,,1.0,0.0,1.0,0.0,1.0,0.0,0.0,,0.0,0.0,0.0,,,0.0,1.0,1.0,1.0,,0.0,
+Glucocorticoid_Sensitivity,,,,,0.0,,0.0,0.0,,,,,,1.0,,0.0,,,0.0,1.0,0.0,,,,,,,,1.0,1.0,,,0.0,,1.0,,,0.0,,1.0,,,0.0,1.0,0.0,1.0,,1.0,1.0,,,0.0,,,,,1.0,0.0,,,,,,0.0,,1.0,0.0,1.0,,,,0.0,0.0,0.0,,,1.0,,0.0,,1.0,1.0,0.0,,0.0,,1.0,1.0,1.0,0.0,,1.0,0.0,,,1.0,,,,0.0,,1.0,,0.0,0.0,,,,,1.0,,,1.0,1.0,0.0,0.0,0.0,0.0,,1.0,0.0,1.0,1.0,0.0,0.0,1.0,1.0,0.0,1.0,1.0,,1.0,1.0,0.0,0.0,1.0,1.0,1.0,,1.0,1.0,0.0,1.0,0.0,1.0,0.0,0.0,1.0,0.0,1.0,0.0,0.0,1.0,0.0,1.0,0.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,1.0,0.0

output/preprocess/Glucocorticoid_Sensitivity/code/GSE15820.py

@@ -0,0 +1,154 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE15820"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE15820"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE15820.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE15820.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE15820.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+# Determine data availability based on the provided background and sample characteristics.
+# This is a cell-line experiment with Affymetrix Exon 1.0 microarrays (gene expression data present),
+# but no human clinical traits (age/gender) and no measurable human glucocorticoid sensitivity.
+is_gene_available = True
+
+# No human-level trait, age, or gender in this dataset
+trait_row = None
+age_row = None
+gender_row = None
+
+# Define converters (not used here since trait_row/age_row/gender_row are None)
+def _extract_after_colon(x):
+    if x is None:
+        return None
+    if isinstance(x, str):
+        parts = x.split(":", 1)
+        val = parts[1].strip() if len(parts) > 1 else x.strip()
+        return val if val not in {"", "NA", "N/A", "na", "n/a", "null", "None"} else None
+    return None
+
+def convert_trait(x):
+    # Not applicable here; placeholder for interface compatibility
+    val = _extract_after_colon(x)
+    return None
+
+def convert_age(x):
+    val = _extract_after_colon(x)
+    if val is None:
+        return None
+    try:
+        return float(val)
+    except Exception:
+        return None
+
+def convert_gender(x):
+    val = _extract_after_colon(x)
+    if val is None:
+        return None
+    v = val.strip().lower()
+    if v in {"female", "f", "girl", "woman", "women"}:
+        return 0
+    if v in {"male", "m", "boy", "man", "men"}:
+        return 1
+    return None
+
+# Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# Clinical feature extraction is skipped because trait_row is None (no human clinical data available).
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Decide mapping columns based on annotation preview:
+# - Expression data index: IDs like '52_36n...'
+# - Annotation: 'ID' column matches these probe/transcript IDs; 'symbol' contains human gene symbols.
+
+id_col = 'ID'
+gene_symbol_col = 'symbol'
+
+# 2. Build mapping dataframe: probe/transcript ID -> gene symbol
+mapping_df = get_gene_mapping(gene_annotation, prob_col=id_col, gene_col=gene_symbol_col)
+
+# 3. Apply mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(gene_data, mapping_df)
+
+# Step 7: Data Normalization and Linking
+# Normalize and save gene expression data; skip linking since no clinical trait is available
+
+import os
+
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. No clinical data available to link
+linked_data = None
+
+# 5. Final quality validation and save cohort info (trait unavailable)
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=True,
+    is_trait_available=False,
+    is_biased=False,
+    df=normalized_gene_data.T,  # pass a non-empty df to avoid abnormality override
+    note="INFO: Cell-line series without human trait/age/gender; no linked dataset created."
+)
+
+# 6. No saving of linked data since dataset is unusable for association analysis (no trait)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE32962.py

@@ -0,0 +1,176 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE32962"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE32962"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE32962.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE32962.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE32962.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import pandas as pd
+
+# 1) Gene expression data availability (expression profiles; not miRNA/methylation)
+is_gene_available = True
+
+# 2) Variable availability from the sample characteristics dictionary
+# trait: prednisolone sensitivity (glucocorticoid sensitivity) at key 4
+trait_row = 4
+
+# age: "< 1 year of age" for all samples -> constant, treat as unavailable
+age_row = None
+
+# gender: not present
+gender_row = None
+
+# 2.2) Data type conversions
+def convert_trait(v):
+    if v is None or (isinstance(v, float) and pd.isna(v)):
+        return None
+    s = str(v)
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    s = s.strip().lower()
+    # Map glucocorticoid (prednisolone) sensitivity to binary: sensitive=1, resistant=0
+    if s in {'sensitive', 'responder', 'responsive', 'good responder'}:
+        return 1
+    if s in {'resistant', 'nonresponder', 'non-responder', 'non responder', 'poor responder'}:
+        return 0
+    return None
+
+def convert_age(v):
+    # Not used because age_row is None
+    return None
+
+def convert_gender(v):
+    # Not used because gender_row is None
+    return None
+
+# 3) Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (only if trait is available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=None,
+        gender_row=gender_row,
+        convert_gender=None
+    )
+    clinical_preview = preview_df(selected_clinical_df)
+    print(clinical_preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+# Affymetrix probe set IDs detected (e.g., '1007_s_at', '1053_at'), not human gene symbols
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# 1. Identify the appropriate columns for mapping: probe IDs ('ID') and gene symbols ('Gene Symbol')
+probe_col = 'ID'
+gene_symbol_col = 'Gene Symbol'
+
+# 2. Create the mapping dataframe
+gene_mapping = get_gene_mapping(annotation=gene_annotation, prob_col=probe_col, gene_col=gene_symbol_col)
+
+# 3. Apply the mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=gene_mapping)
+
+# Step 7: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1. Normalize the obtained gene data and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# Ensure clinical data is available in scope; reload if necessary
+if 'selected_clinical_df' not in locals() or selected_clinical_df is None:
+    selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+
+# 2. Link the clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Determine bias and remove biased demographic features if needed
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final quality validation and save cohort info
+note = "INFO: Age not available (all infants <1 year); Gender not provided. Affymetrix probe mapping applied and gene symbols normalized."
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=True,
+    is_trait_available=True,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE33649.py

@@ -0,0 +1,184 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE33649"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE33649"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE33649.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE33649.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE33649.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+
+# 1) Availability
+is_gene_available = True  # Transcriptome-wide response indicates gene expression data is available.
+
+# 2) Variable availability (from the provided Sample Characteristics Dictionary)
+trait_row = 3    # "in vitro lymphocyte gc sensitivity (lgs - %inhibition by dex): <float>"
+age_row = 6      # "age (years): <float>"
+gender_row = 5   # "gender: female/male"
+
+# 2.2) Converters
+def _after_colon(val):
+    if val is None:
+        return None
+    if isinstance(val, str):
+        parts = val.split(":", 1)
+        return parts[1].strip() if len(parts) > 1 else val.strip()
+    return str(val).strip()
+
+def _to_float(val):
+    if val is None:
+        return None
+    s = _after_colon(val)
+    if s is None:
+        return None
+    s = s.strip()
+    # Extract a numeric token robustly
+    m = re.search(r'[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?', s)
+    if not m:
+        return None
+    try:
+        return float(m.group(0))
+    except Exception:
+        return None
+
+def convert_trait(x):
+    # Glucocorticoid sensitivity (LGS - % inhibition by dex); continuous
+    return _to_float(x)
+
+def convert_age(x):
+    # Age in years; continuous
+    return _to_float(x)
+
+def convert_gender(x):
+    # Gender; binary: female -> 0, male -> 1
+    s = _after_colon(x)
+    if s is None:
+        return None
+    s = s.strip().lower()
+    if s in {"female", "f", "woman", "women"}:
+        return 0
+    if s in {"male", "m", "man", "men"}:
+        return 1
+    if s in {"na", "n/a", "unknown", "not available", "null", ""}:
+        return None
+    return None
+
+# 3) Initial filtering metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+
+    clinical_preview = preview_df(selected_clinical_df, n=5)
+    print("Clinical data preview:", clinical_preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Determine appropriate columns for probe IDs and gene symbols based on annotation preview
+probe_col = 'ID'
+symbol_candidates = ['Symbol', 'ILMN_Gene', 'Gene Symbol', 'GeneSymbol', 'GENE_SYMBOL']
+gene_col = next((c for c in symbol_candidates if c in gene_annotation.columns), None)
+if gene_col is None:
+    raise ValueError(f"No suitable gene symbol column found in gene_annotation. Available columns: {list(gene_annotation.columns)}")
+
+# 2) Build mapping dataframe (probe -> gene symbol)
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_col)
+
+# 3) Apply mapping to convert probe-level data to gene-level data
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Bias assessment and removal of biased demographics
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort info
+note = "INFO: ILMN probes mapped to gene symbols via 'Symbol'; LGS used as continuous trait; Age and Gender included."
+is_usable = validate_and_save_cohort_info(
+    True, cohort, json_path, True, True, is_trait_biased, unbiased_linked_data, note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE42002.py

@@ -0,0 +1,185 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE42002"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE42002"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE42002.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE42002.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE42002.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+# Determine data availability based on provided background and sample characteristics
+is_gene_available = True  # mRNA expression arrays are mentioned in the summary
+
+# No explicit or inferable per-subject glucocorticoid sensitivity variable in sample characteristics
+trait_row = None  # Glucocorticoid_Sensitivity not explicitly available
+age_row = None    # No age information in sample characteristics
+gender_row = None # No gender information in sample characteristics
+
+# Conversion functions
+def _extract_value(cell):
+    if cell is None:
+        return None
+    s = str(cell).strip()
+    # Typical format "key: value"
+    if ':' in s:
+        s = s.split(':', 1)[1].strip()
+    return s if s != '' else None
+
+def convert_trait(cell):
+    # Trait not available for this dataset; return None for any input
+    return None
+
+def convert_age(cell):
+    s = _extract_value(cell)
+    if s is None:
+        return None
+    # Extract the first number (integer or float) from the string
+    import re
+    m = re.search(r'(-?\d+(\.\d+)?)', s)
+    if m:
+        try:
+            return float(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def convert_gender(cell):
+    s = _extract_value(cell)
+    if s is None:
+        return None
+    s_low = s.lower()
+    if s_low in {'male', 'm', 'man', 'boy'}:
+        return 1
+    if s_low in {'female', 'f', 'woman', 'girl'}:
+        return 0
+    return None
+
+# Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# Clinical feature extraction: only if trait_row is available (skip here since trait_row is None)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    _ = preview_df(selected_clinical_df, n=5)
+    # Ensure directory exists and save
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Identify the appropriate columns for probe IDs and gene symbols from the annotation preview
+probe_col = 'ID'       # e.g., 'ILMN_1343291'
+gene_col = 'Symbol'    # human gene symbols
+
+# Build mapping dataframe (ID -> Gene)
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_col)
+
+# Map probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize the obtained gene data and save to file
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2-6. Proceed only if clinical features (including trait) were extracted earlier
+if 'selected_clinical_data' in globals():
+    # Link clinical and genetic data
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
+
+    # Handle missing values
+    linked_data = handle_missing_values(linked_data, trait)
+
+    # Bias evaluation
+    is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+    # Final validation and cohort info saving
+    is_usable = validate_and_save_cohort_info(
+        is_final=True,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=True,
+        is_trait_available=True,
+        is_biased=is_trait_biased,
+        df=unbiased_linked_data,
+        note="INFO: Proceeded with full pipeline as clinical trait data was available."
+    )
+
+    # Save linked data only if usable
+    if is_usable:
+        os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+        unbiased_linked_data.to_csv(out_data_file)
+else:
+    # Trait/clinical data unavailable; record (or reaffirm) initial metadata only
+    _ = validate_and_save_cohort_info(
+        is_final=False,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=True,
+        is_trait_available=False
+    )

output/preprocess/Glucocorticoid_Sensitivity/code/GSE48801.py

@@ -0,0 +1,220 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE48801"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE48801"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE48801.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE48801.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE48801.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+
+# 1) Gene expression data availability
+is_gene_available = True  # Transcriptome-wide response to GCs indicates gene expression data (not miRNA/methylation)
+
+# 2) Variable availability and conversion functions
+
+# From provided characteristics:
+# 0 -> treatment condition (not used as a clinical covariate here)
+# 1 -> in vitro lymphocyte GC sensitivity (LGS - % inhibition by dex) -> our trait
+trait_row = 1
+age_row = None
+gender_row = None
+
+def _get_value_after_colon(x):
+    if x is None:
+        return None
+    try:
+        # Take the substring after the last colon, common in GEO annotations
+        val = str(x).split(':')[-1].strip()
+        return val if val != '' else None
+    except Exception:
+        return None
+
+def convert_trait(x):
+    # LGS is continuous (% inhibition by dex)
+    val = _get_value_after_colon(x)
+    if val is None:
+        return None
+    # Extract the first float-like number
+    m = re.search(r'-?\d+(?:\.\d+)?', val)
+    if m:
+        try:
+            return float(m.group(0))
+        except Exception:
+            return None
+    return None
+
+def convert_age(x):
+    # Not available for this dataset, but define for completeness
+    val = _get_value_after_colon(x)
+    if val is None:
+        return None
+    # Extract age number if present
+    m = re.search(r'\d+(?:\.\d+)?', val)
+    if m:
+        try:
+            return float(m.group(0))
+        except Exception:
+            return None
+    return None
+
+def convert_gender(x):
+    # Not available for this dataset, but define for completeness
+    val = _get_value_after_colon(x)
+    if val is None:
+        return None
+    s = val.strip().lower()
+    if s in {'female', 'f', 'woman', 'girl'}:
+        return 0
+    if s in {'male', 'm', 'man', 'boy'}:
+        return 1
+    if s == '0':
+        return 0
+    if s == '1':
+        return 1
+    if s in {'na', 'n/a', 'unknown', 'undet', 'undetermined', '-', ''}:
+        return None
+    return None
+
+# 3) Save metadata - initial filtering
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical Feature Extraction (only if trait_row is available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age if age_row is not None else None,
+        gender_row=gender_row,
+        convert_gender=convert_gender if gender_row is not None else None
+    )
+    # Preview
+    print(preview_df(selected_clinical_df, n=5))
+
+    # Ensure directory exists and save
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Determine appropriate columns for mapping:
+# - Probe identifiers in expression data: "ILMN_..." -> corresponds to 'ID' in gene_annotation
+# - Gene symbols in annotation: corresponds to 'Symbol'
+
+# In case gene_annotation is not in scope (robustness), reload it
+try:
+    gene_annotation
+except NameError:
+    gene_annotation = get_gene_annotation(soft_file)
+
+# 2) Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Symbol')
+
+# 3) Apply mapping: convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+# Ensure clinical dataframe is available; if not, load it from file
+try:
+    selected_clinical_df
+except NameError:
+    selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# Determine availability flags before missing-value handling
+pre_mv_is_gene_available = (normalized_gene_data.shape[0] > 0 and normalized_gene_data.shape[1] > 0)
+pre_mv_is_trait_available = trait in linked_data.columns
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Bias evaluation and removal of biased demographic features
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort info
+note = ("INFO: LGS is continuous; each donor appears to contribute multiple GSMs under different treatments, "
+        "sharing identical LGS values across those replicates.")
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=pre_mv_is_gene_available,
+    is_trait_available=pre_mv_is_trait_available,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked dataset if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE50012.py

@@ -0,0 +1,206 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE50012"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE50012"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE50012.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE50012.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE50012.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import pandas as pd
+
+# 1) Gene expression availability (based on background: expression profiling of PBMCs with steroids)
+is_gene_available = True
+
+# 2) Variable availability inferred from Sample Characteristics Dictionary:
+# Trait (Glucocorticoid_Sensitivity): row 3 contains "in vitro lymphocyte gc sensitivity (lgs - %inhibition by dex): <value>"
+trait_row = 3
+
+# Age: use the row that is purely ages to avoid mixing with gender; row 6 fits this
+age_row = 6
+
+# Gender: gender entries appear to align with a different GSM subset than the trait.
+# To avoid adding a mostly-missing feature for the trait samples, set to None.
+gender_row = None
+
+# 2.2) Converters
+def _after_colon(val: str) -> str:
+    if val is None:
+        return ""
+    parts = val.rsplit(":", 1)
+    return parts[1].strip() if len(parts) == 2 else val.strip()
+
+def convert_trait(x):
+    # Expect strings like "in vitro lymphocyte gc sensitivity (lgs - %inhibition by dex): 95.22036"
+    if not isinstance(x, str):
+        return None
+    s = x.lower()
+    if "gc sensitivity" in s or "glucocorticoid" in s:
+        v = _after_colon(x)
+        try:
+            return float(v)
+        except:
+            m = re.search(r"[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?", v)
+            return float(m.group(0)) if m else None
+    return None
+
+def convert_age(x):
+    # Expect strings like "age (years): 24.72"
+    if not isinstance(x, str):
+        return None
+    if "age" in x.lower():
+        v = _after_colon(x)
+        try:
+            return float(v)
+        except:
+            m = re.search(r"[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?", v)
+            return float(m.group(0)) if m else None
+    return None
+
+def convert_gender(x):
+    # Expect strings like "gender: female" or "gender: male"
+    if not isinstance(x, str):
+        return None
+    s = x.lower()
+    if "gender" in s:
+        v = _after_colon(s)
+        if "female" in v:
+            return 0
+        if "male" in v:
+            return 1
+        if v in {"na", "n/a", "unknown", ""}:
+            return None
+    return None
+
+# 3) Save metadata with initial filtering
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction if trait is available
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    preview = preview_df(selected_clinical_df, n=5)
+    print("Clinical features preview:", preview)
+
+    # Save clinical data
+    out_dir = os.path.dirname(out_clinical_data_file)
+    os.makedirs(out_dir, exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Decide the columns for probe IDs and gene symbols based on the annotation preview
+probe_col = 'ID'
+gene_symbol_candidates = ['Symbol', 'ILMN_Gene']
+gene_col = next((c for c in gene_symbol_candidates if c in gene_annotation.columns), None)
+if gene_col is None:
+    raise ValueError("No suitable gene symbol column found in gene annotation.")
+
+# Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_col)
+
+# Apply mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Determine bias and remove biased demographic features
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# Compute availability flags for final validation as Python bools
+is_gene_available_final = bool((normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0))
+is_trait_available_final = bool((trait in linked_data.columns) and (len(linked_data) > 0) and (linked_data[trait].notna().sum() > 0))
+
+# 5. Final validation and save cohort info
+note = ("INFO: Trait values (GC sensitivity) are available only for a subset of samples (dexamethasone-treated PBMCs). "
+        "Gender was unavailable/aligned for trait subset and was omitted.")
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available_final,
+    is_trait_available=is_trait_available_final,
+    is_biased=bool(is_trait_biased),
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Conditionally save linked data
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE57795.py

@@ -0,0 +1,215 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE57795"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE57795"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE57795.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE57795.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE57795.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import pandas as pd
+
+# 1) Gene expression availability
+is_gene_available = True  # Illumina HumanWG-6 v3 Expression BeadChips (gene expression microarray)
+
+# 2) Determine availability rows from Sample Characteristics Dictionary:
+# trait -> 'dexamethasone response' (Sensitive vs Resistant) at row 5
+trait_row = 5
+
+# Age and Gender (human) are not available in this PDX mouse dataset
+age_row = None
+gender_row = None
+
+# 2.2) Converters
+def _extract_value(x):
+    if x is None or (isinstance(x, float) and pd.isna(x)):
+        return None
+    s = str(x)
+    # Extract substring after the first colon if present
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    return s.strip()
+
+def convert_trait(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    s = val.lower()
+    # Map sensitivity/resistance; accept synonyms 'good/poor responder'
+    if ('sensitive' in s) or ('good' in s and 'respond' in s):
+        return 1
+    if ('resistant' in s) or ('poor' in s and 'respond' in s):
+        return 0
+    return None
+
+def convert_age(x):
+    # Not used (no human age available). Generic parser that converts to years if possible.
+    val = _extract_value(x)
+    if val is None:
+        return None
+    s = val.lower()
+    # Find first number (can be a range like 6-10; take mean)
+    # and unit among years/months/weeks/days
+    num = None
+    # Handle ranges like "6-10 weeks"
+    m_range = re.search(r'(\d+(?:\.\d+)?)\s*[-–]\s*(\d+(?:\.\d+)?)', s)
+    if m_range:
+        a = float(m_range.group(1))
+        b = float(m_range.group(2))
+        num = 0.5 * (a + b)
+    else:
+        m_num = re.search(r'(\d+(?:\.\d+)?)', s)
+        if m_num:
+            num = float(m_num.group(1))
+    if num is None:
+        return None
+    # Detect unit
+    if 'year' in s or 'yr' in s or 'y/o' in s or 'y ' in s:
+        years = num
+    elif 'month' in s or 'mo' in s:
+        years = num / 12.0
+    elif 'week' in s or 'wk' in s or 'w ' in s:
+        years = num / 52.0
+    elif 'day' in s or 'd ' in s:
+        years = num / 365.0
+    else:
+        # Unknown unit; return as-is
+        years = num
+    # Typically we want continuous numeric
+    return years
+
+def convert_gender(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    s = val.strip().lower()
+    # Map to 0/1 (female=0, male=1)
+    if s in ['male', 'm', 'man', 'boy']:
+        return 1
+    if s in ['female', 'f', 'woman', 'girl']:
+        return 0
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical Feature Extraction (only if trait is available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    preview_selected = preview_df(selected_clinical_df, n=5)
+    # Ensure output directory exists and save
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+    preview_selected
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+# ILMN_* identifiers are Illumina probe IDs, not human gene symbols.
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Identify the appropriate columns for mapping: probe IDs and gene symbols
+probe_col = 'ID'       # Matches ILMN_* probe identifiers in the expression matrix
+gene_symbol_col = 'Symbol'  # Gene symbols
+
+# 2. Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_symbol_col)
+
+# 3. Apply mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save normalized gene data
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link the clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Bias checks and remove biased demographic features if any
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort metadata
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=True,
+    is_trait_available=True,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note="INFO: PDX dataset; human Age/Gender unavailable."
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE58715.py

@@ -0,0 +1,169 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE58715"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE58715"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE58715.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE58715.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE58715.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import pandas as pd
+
+# 1) Gene expression data availability
+is_gene_available = True  # mRNA gene expression study with hormone treatment per design
+
+# 2) Variable availability and converters
+# Trait is inferred from hormone treatment status (dexamethasone vs ethanol)
+trait_row = 2  # 'hormone' field
+age_row = None  # not human subjects
+gender_row = None  # not human subjects
+
+def convert_trait(x):
+    if pd.isna(x):
+        return None
+    s = str(x)
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    s = s.strip().lower()
+    # Map dexamethasone exposure to 1, ethanol control to 0
+    if s.startswith('dexamethasone') or s.startswith('dex'):
+        return 1
+    if s.startswith('ethanol') or s in {'etoh', 'control'}:
+        return 0
+    return None
+
+def convert_age(x):
+    # No human age data available in this cell line study
+    return None
+
+def convert_gender(x):
+    # No human gender data available in this cell line study
+    return None
+
+# 3) Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (only if trait is available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    preview = preview_df(selected_clinical_df)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# 1-2. Decide columns for probe IDs and gene symbols, and extract the mapping dataframe
+probe_col = 'ID'       # Matches probe identifiers like 'ILMN_1343291'
+gene_symbol_col = 'Symbol'  # Contains gene symbols
+
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_symbol_col)
+
+# 3. Apply the mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+# Ensure clinical data is available in this step
+if 'selected_clinical_df' not in globals():
+    selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Assess bias and remove biased covariates
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort info
+is_gene_available = normalized_gene_data.shape[0] > 0
+is_trait_available = trait in linked_data.columns
+
+note = "INFO: Cell line study with hormone treatment; Age/Gender not applicable."
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE65645.py

@@ -0,0 +1,188 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE65645"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE65645"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE65645.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE65645.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE65645.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import re
+import pandas as pd
+
+# 1) Gene expression availability (lncRNA expression on Agilent arrays is acceptable transcriptomic data)
+is_gene_available = True
+
+# 2) Variable availability based on provided Sample Characteristics Dictionary:
+# {0: ['sample_type: bone marrow'], 1: ['translocation: E2A_PBX1', 'translocation: MLL', 'translocation: TEL_AML1']}
+# No explicit or inferable Glucocorticoid_Sensitivity, Age, or Gender fields are present or variable.
+trait_row = None
+age_row = None
+gender_row = None
+
+# 2.2) Converters
+def _extract_after_colon(x):
+    if x is None:
+        return None
+    s = str(x)
+    parts = s.split(":", 1)
+    val = parts[1] if len(parts) == 2 else parts[0]
+    return val.strip()
+
+def convert_trait(x):
+    # Binary: 1 = sensitive/responder, 0 = resistant/non-responder; unknown -> None
+    val = _extract_after_colon(x)
+    if val is None or val == "":
+        return None
+    s = val.lower()
+    # Positive patterns
+    pos_patterns = [
+        "sensitive", "responder", "responsive", "good response", "good-responder", "good_responder",
+        "prednisone sensitive", "prednisolone sensitive", "high sensitivity", "sensitive to prednisone",
+        "r"  # common shorthand, resolve later if ambiguous
+    ]
+    # Negative patterns
+    neg_patterns = [
+        "resistant", "non-responder", "nonresponder", "non responsive", "non-responsive",
+        "poor response", "refractory", "prednisone resistant", "prednisolone resistant",
+        "low sensitivity", "nr"  # common shorthand
+    ]
+    # Heuristic resolution: check explicit negations before generic substrings to avoid collisions
+    for pat in neg_patterns:
+        if pat in s and not re.search(r'\b(r)\b', s):  # avoid misclassifying isolated 'r' that could mean responder
+            return 0
+    for pat in pos_patterns:
+        if pat in s:
+            # If 'r' matched, ensure it's not 'nr'
+            if pat == "r" and ("nr" in s or "non-responder" in s or "nonresponder" in s):
+                continue
+            return 1
+    # Short codes exactly equal to labels
+    if s.strip() in {"r", "responder"}:
+        return 1
+    if s.strip() in {"nr", "non-responder", "nonresponder"}:
+        return 0
+    return None
+
+def convert_age(x):
+    val = _extract_after_colon(x)
+    if val is None or val == "":
+        return None
+    s = val.lower()
+    # Extract first floating number as age in years
+    m = re.search(r'[-+]?\d*\.?\d+', s)
+    if m:
+        try:
+            return float(m.group())
+        except Exception:
+            return None
+    return None
+
+def convert_gender(x):
+    val = _extract_after_colon(x)
+    if val is None or val == "":
+        return None
+    s = val.strip().lower()
+    if s in {"female", "f", "woman", "girl"}:
+        return 0
+    if s in {"male", "m", "man", "boy"}:
+        return 1
+    # Numeric coding already
+    if s in {"0", "1"}:
+        return int(s)
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (skip since trait_row is None)
+if is_trait_available and 'clinical_data' in locals():
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    _ = preview_df(selected_clinical_df)
+    # Save
+    out_dir = os.path.dirname(out_clinical_data_file)
+    os.makedirs(out_dir, exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Identify appropriate columns for probe IDs and gene symbols
+probe_col = 'ID'
+candidate_gene_cols = ['GENE_SYMBOL', 'GENE', 'GENE_NAME']
+available_gene_cols = [c for c in candidate_gene_cols if c in gene_annotation.columns]
+
+# Select the gene symbol column with the most non-null entries
+if available_gene_cols:
+    non_null_counts = {c: gene_annotation[c].notna().sum() for c in available_gene_cols}
+    gene_col = max(non_null_counts, key=non_null_counts.get)
+else:
+    raise ValueError("No suitable gene symbol column found in gene annotation.")
+
+# Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_col)
+
+# Apply mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)

output/preprocess/Glucocorticoid_Sensitivity/code/GSE66705.py

@@ -0,0 +1,196 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+cohort = "GSE66705"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Glucocorticoid_Sensitivity"
+in_cohort_dir = "../DATA/GEO/Glucocorticoid_Sensitivity/GSE66705"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/GSE66705.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/GSE66705.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/GSE66705.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import pandas as pd
+
+# 1) Gene expression availability based on background ("Gene expression profiling" on HG-U133_Plus_2 platform)
+is_gene_available = True
+
+# 2) Identify rows in the sample characteristics dictionary
+# Provided dictionary:
+# {0: ['predlc50group: #N/A', 'predlc50group: SEN', 'predlc50group: RES', 'predlc50group: INT'],
+#  1: ['lin: B', 'lin: T']}
+trait_row = 0     # 'predlc50group' reflects glucocorticoid sensitivity categories
+age_row = None    # Not available
+gender_row = None # Not available
+
+# 2.2) Converters
+def convert_trait(x):
+    if x is None or (isinstance(x, float) and pd.isna(x)):
+        return None
+    v = str(x).split(':', 1)[-1].strip().lower()
+    if v in {'#n/a', 'na', 'n/a', 'nan', ''}:
+        return None
+    # Map to binary sensitivity: SEN -> 1 (sensitive); RES/INT -> 0 (non-sensitive)
+    if v in {'sen', 'sensitive'}:
+        return 1
+    if v in {'res', 'resistant', 'int', 'intermediate'}:
+        return 0
+    return None
+
+def convert_age(x):
+    if x is None or (isinstance(x, float) and pd.isna(x)):
+        return None
+    v = str(x).split(':', 1)[-1].strip().lower()
+    if v in {'#n/a', 'na', 'n/a', 'nan', ''}:
+        return None
+    m = re.search(r'[-+]?\d*\.?\d+', v)
+    if not m:
+        return None
+    try:
+        return float(m.group())
+    except Exception:
+        return None
+
+def convert_gender(x):
+    if x is None or (isinstance(x, float) and pd.isna(x)):
+        return None
+    v = str(x).split(':', 1)[-1].strip().lower()
+    if v in {'#n/a', 'na', 'n/a', 'nan', ''}:
+        return None
+    if v in {'female', 'f', 'girl', 'woman', 'women'}:
+        return 0
+    if v in {'male', 'm', 'boy', 'man', 'men'}:
+        return 1
+    return None
+
+# 3) Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (only if trait data available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=None,
+        gender_row=gender_row,
+        convert_gender=None
+    )
+    preview = preview_df(selected_clinical_df, n=5)
+    print(preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Decide the appropriate columns for probe IDs and gene symbols based on the annotation preview
+probe_col = 'ID'            # Matches probe IDs like '1007_s_at'
+gene_symbol_col = 'Gene Symbol'  # Contains gene symbols (may include multiple symbols per probe)
+
+# 2) Build mapping dataframe from annotation
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_symbol_col)
+
+# 3) Apply mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save normalized gene expression data
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link the clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# Derive availability indicators from actual data before downstream processing (ensure pure Python bools)
+is_gene_available = bool((normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0))
+is_trait_available = bool((trait in linked_data.columns) and bool(linked_data[trait].notna().any()))
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Assess bias and remove biased demographic features
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+is_trait_biased = bool(is_trait_biased)  # ensure pure Python bool
+
+# 5. Final validation and save cohort info
+# Pre-create JSON file to avoid creation-branch quirks
+os.makedirs(os.path.dirname(json_path), exist_ok=True)
+if not os.path.exists(json_path):
+    with open(json_path, "w") as f:
+        f.write("{}")
+
+note = "INFO: Trait derived from 'predlc50group' (SEN=1, RES/INT=0). No age or gender available."
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=bool(is_gene_available),
+    is_trait_available=bool(is_trait_available),
+    is_biased=bool(is_trait_biased),
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Glucocorticoid_Sensitivity/code/TCGA.py

@@ -0,0 +1,52 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Glucocorticoid_Sensitivity"
+
+# Input paths
+tcga_root_dir = "../DATA/TCGA"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/TCGA.csv"
+out_gene_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/gene_data/TCGA.csv"
+out_clinical_data_file = "./output/z3/preprocess/Glucocorticoid_Sensitivity/clinical_data/TCGA.csv"
+json_path = "./output/z3/preprocess/Glucocorticoid_Sensitivity/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+import os
+import pandas as pd
+
+# List subdirectories in TCGA root
+subdirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+
+# Identify directories related to the trait "Glucocorticoid_Sensitivity"
+trait_keywords = [
+    "glucocorticoid", "corticosteroid", "cortisol", "dexamethasone", "prednis", "methylpred", "hydrocortisone"
+]
+matches = [d for d in subdirs if any(k in d.lower() for k in trait_keywords)]
+
+if len(matches) == 0:
+    # No suitable directory found; mark as skipped and complete
+    validate_and_save_cohort_info(
+        is_final=False,
+        cohort="TCGA",
+        info_path=json_path,
+        is_gene_available=False,
+        is_trait_available=False
+    )
+else:
+    # Choose the most specific match (shortest name as a proxy)
+    selected_dir = sorted(matches, key=len)[0]
+    cohort_dir = os.path.join(tcga_root_dir, selected_dir)
+
+    # Get file paths
+    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)
+
+    # Load data
+    clinical_df = pd.read_csv(clinical_file_path, sep='\t', index_col=0, low_memory=False)
+    genetic_df = pd.read_csv(genetic_file_path, sep='\t', index_col=0, low_memory=False)
+
+    # Print clinical column names
+    print(list(clinical_df.columns))

output/preprocess/Head_and_Neck_Cancer/clinical_data/GSE151181.csv

@@ -1,2 +1,2 @@
 ,GSM4567912,GSM4567913,GSM4567914,GSM4567915,GSM4567916,GSM4567917,GSM4567918,GSM4567919,GSM4567920,GSM4567921,GSM4567922,GSM4567923,GSM4567924,GSM4567925,GSM4567926,GSM4567927,GSM4567928,GSM4567929,GSM4567930,GSM4567931,GSM4567932,GSM4567933,GSM4567934,GSM4567935,GSM4567936,GSM4567937,GSM4567938,GSM4567939,GSM4567940,GSM4567941,GSM4567942,GSM4567943,GSM4567944,GSM4567945,GSM4567946,GSM4567947,GSM4567948,GSM4567949,GSM4567950,GSM4567951,GSM4567952,GSM4567953,GSM4567954,GSM4567955,GSM4567956,GSM4567957,GSM4567958,GSM4567959,GSM4567960,GSM4567961,GSM4567962,GSM4567963
-Head_and_Neck_Cancer,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+Head_and_Neck_Cancer,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0

output/preprocess/Head_and_Neck_Cancer/clinical_data/GSE201777.csv

@@ -1,2 +1,2 @@
-GSM6071161,GSM6071162,GSM6071163,GSM6071164,GSM6071165,GSM6071166,GSM6071167,GSM6071168,GSM6071169,GSM6071170,GSM6071171,GSM6071172,GSM6071173,GSM6071174,GSM6071175,GSM6071176,GSM6071177,GSM6071178,GSM6071179,GSM6071180,GSM6071181,GSM6071182,GSM6071183,GSM6071184,GSM6071185,GSM6071186,GSM6071187,GSM6071188,GSM6071189,GSM6071190,GSM6071191,GSM6071192,GSM6071193,GSM6071194,GSM6071195,GSM6071196,GSM6071197,GSM6071198,GSM6071199,GSM6071200,GSM6071201,GSM6071202,GSM6071203,GSM6071204,GSM6071205,GSM6071206,GSM6071207
-0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0
+,GSM6071161,GSM6071162,GSM6071163,GSM6071164,GSM6071165,GSM6071166,GSM6071167,GSM6071168,GSM6071169,GSM6071170,GSM6071171,GSM6071172,GSM6071173,GSM6071174,GSM6071175,GSM6071176,GSM6071177,GSM6071178,GSM6071179,GSM6071180,GSM6071181,GSM6071182,GSM6071183,GSM6071184,GSM6071185,GSM6071186,GSM6071187,GSM6071188,GSM6071189,GSM6071190,GSM6071191,GSM6071192,GSM6071193,GSM6071194,GSM6071195,GSM6071196,GSM6071197,GSM6071198,GSM6071199,GSM6071200,GSM6071201,GSM6071202,GSM6071203,GSM6071204,GSM6071205,GSM6071206,GSM6071207
+Head_and_Neck_Cancer,,1.0,0.0,0.0,1.0,,1.0,0.0,,1.0,0.0,,1.0,,0.0,,0.0,1.0,1.0,0.0,,,1.0,0.0,1.0,,0.0,0.0,,0.0,1.0,,0.0,1.0,,1.0,,0.0,1.0,0.0,,,0.0,1.0,1.0,0.0,

output/preprocess/Head_and_Neck_Cancer/clinical_data/GSE244580.csv

@@ -1,2 +1,2 @@
-GSM7820913,GSM7820914,GSM7820915,GSM7820916,GSM7820917,GSM7820918,GSM7820919,GSM7820920,GSM7820921,GSM7820922,GSM7820923,GSM7820924,GSM7820925,GSM7820926,GSM7820927,GSM7820928,GSM7820929,GSM7820930,GSM7820931,GSM7820932,GSM7820933,GSM7820934,GSM7820935,GSM7820936,GSM7820937,GSM7820938,GSM7820939,GSM7820940,GSM7820941,GSM7820942
-0.0,0.0,0.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0
+,GSM7820913,GSM7820914,GSM7820915,GSM7820916,GSM7820917,GSM7820918,GSM7820919,GSM7820920,GSM7820921,GSM7820922,GSM7820923,GSM7820924,GSM7820925,GSM7820926,GSM7820927,GSM7820928,GSM7820929,GSM7820930,GSM7820931,GSM7820932,GSM7820933,GSM7820934,GSM7820935,GSM7820936,GSM7820937,GSM7820938,GSM7820939,GSM7820940,GSM7820941,GSM7820942
+Head_and_Neck_Cancer,0.0,0.0,0.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0

output/preprocess/Head_and_Neck_Cancer/code/GSE104006.py

@@ -0,0 +1,133 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE104006"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE104006"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE104006.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE104006.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE104006.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+
+# 1. Gene Expression Data Availability
+# Title indicates both miRNA and gene expression profiling; assume gene expression is available for this cohort.
+is_gene_available = True
+
+# 2. Variable Availability and Data Type Conversion
+# Sample Characteristics Dictionary keys inferred from the provided preview:
+trait_row = 0     # disease: Thyroid_carcinoma vs Non-neoplastic_thyroid
+age_row = 2       # age: numeric values
+gender_row = 3    # Sex: F/M
+
+def convert_trait(v):
+    if v is None:
+        return None
+    val = str(v).split(":", 1)[-1].strip().lower().replace(" ", "_").replace("-", "_")
+    # Map cancer vs non-cancer
+    if val in {"thyroid_carcinoma", "carcinoma", "cancer", "tumor"}:
+        return 1
+    if val in {"non_neoplastic_thyroid", "non_neoplastic", "normal", "control", "benign"}:
+        return 0
+    return None
+
+def convert_age(v):
+    if v is None:
+        return None
+    s = str(v).split(":", 1)[-1].strip()
+    s_lower = s.lower()
+    if s_lower in {"", "na", "n/a", "nan", "none", "unknown"}:
+        return None
+    m = re.search(r"-?\d+\.?\d*", s)
+    if not m:
+        return None
+    try:
+        x = float(m.group())
+        return x
+    except Exception:
+        return None
+
+def convert_gender(v):
+    if v is None:
+        return None
+    val = str(v).split(":", 1)[-1].strip().lower()
+    if val in {"female", "f", "woman", "w"}:
+        return 0
+    if val in {"male", "m", "man"}:
+        return 1
+    return None
+
+# 3. Save Metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4. Clinical Feature Extraction (only if clinical data is available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    clinical_preview = preview_df(selected_clinical_df)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))

output/preprocess/Head_and_Neck_Cancer/code/GSE148320.py

@@ -0,0 +1,139 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE148320"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE148320"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE148320.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE148320.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE148320.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+# Determine availability based on provided background and sample characteristics
+is_gene_available = True  # Gene expression profiling is likely (xenograft/cell line expression), not miRNA/methylation.
+trait_row = None          # No human trait variability; all are tumor xenograft/cell line samples (constant for our trait).
+age_row = None            # No human age information present.
+gender_row = None         # No human gender information present.
+
+# Conversion helpers (defined for completeness; they won't be used since rows are None)
+def _extract_value(x):
+    if x is None:
+        return None
+    s = str(x)
+    parts = s.split(":", 1)
+    return parts[1].strip() if len(parts) == 2 else s.strip()
+
+def convert_trait(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    v_low = v.lower()
+    # Generic mapping if ever applicable
+    if any(k in v_low for k in ["normal", "control", "healthy", "adjacent normal", "benign"]):
+        return 0
+    if any(k in v_low for k in ["cancer", "tumor", "tumour", "carcinoma", "malignant", "hnscc", "head and neck"]):
+        return 1
+    return None
+
+def convert_age(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    # Extract first numeric (years assumed)
+    import re
+    m = re.search(r"(\d+(\.\d+)?)", v)
+    return float(m.group(1)) if m else None
+
+def convert_gender(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    v_low = v.lower()
+    if v_low in ["male", "m"]:
+        return 1
+    if v_low in ["female", "f"]:
+        return 0
+    return None
+
+# Initial filtering metadata save
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# Clinical feature extraction is skipped because trait_row is None (no usable human clinical data).
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    clinical_preview = preview_df(selected_clinical_df)
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Identify the appropriate columns in the annotation: probe IDs and gene symbols
+probe_col = 'ID'
+gene_symbol_col = 'Gene Symbol'
+
+# 1-2. Build mapping dataframe from annotation
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_symbol_col)
+
+# 3. Apply mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)

output/preprocess/Head_and_Neck_Cancer/code/GSE151179.py

@@ -0,0 +1,1518 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE151179"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE151179"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE151179.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE151179.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE151179.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import pandas as pd
+
+# 1. Gene Expression Data Availability
+is_gene_available = True  # Gene expression profiling with Thermo Fisher Human Clariom S Assay
+
+# 2. Variable Availability and Data Type Conversion
+
+# Trait: Use 'tissue type' to distinguish cancer vs non-neoplastic thyroid
+trait_row = 1  # 'tissue type' field varies across samples and can indicate cancer presence
+age_row = None
+gender_row = None
+
+def _extract_value(cell):
+    if cell is None or (isinstance(cell, float) and pd.isna(cell)):
+        return None
+    s = str(cell)
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    return s.strip()
+
+def convert_trait(cell):
+    val = _extract_value(cell)
+    if val is None:
+        return None
+    v = val.lower()
+    # Controls
+    if 'non-neoplastic' in v:
+        return 0
+    # Cancer (tumor/metastasis)
+    if ('tumor' in v) or ('metastasis' in v):
+        return 1
+    return None
+
+# 3. Save Metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4. Clinical Feature Extraction (only if clinical data is available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        gender_row=gender_row
+    )
+    preview = preview_df(selected_clinical_df)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+import re
+import pandas as pd
+
+# Keep a copy of the raw expression data
+expr_df = gene_data.copy()
+exp_ids = set(expr_df.index.astype(str))
+
+def normalize_id_values(series: pd.Series) -> set:
+    # Build a set of candidate IDs from a column using multiple normalization heuristics
+    vals = series.astype(str).fillna("").str.strip()
+    id_set = set()
+    # 1) raw and strip trailing .0
+    stripped = vals.str.replace(r'\.0$', '', regex=True)
+    id_set.update(stripped.tolist())
+    # 2) only keep pure digits
+    id_set.update([s for s in stripped.tolist() if s.isdigit()])
+    # 3) extract long digit groups (length >= 7) from strings
+    for s in stripped:
+        for g in re.findall(r'\d+', s):
+            if len(g) >= 7:
+                id_set.add(g)
+    return id_set
+
+# 1) Decide which annotation column matches the expression IDs
+overlaps = []
+for col in gene_annotation.columns:
+    try:
+        ann_id_set = normalize_id_values(gene_annotation[col])
+        overlap = len(exp_ids & ann_id_set)
+        overlaps.append((col, overlap))
+    except Exception:
+        continue
+
+# Sort columns by overlap
+overlaps.sort(key=lambda x: x[1], reverse=True)
+
+# Pick the best ID column with the largest overlap
+best_id_col, best_overlap = (overlaps[0] if overlaps else (None, 0))
+
+# If the best overlap is zero, try a few common candidate columns explicitly in case they exist
+if best_overlap == 0:
+    explicit_candidates = [
+        'transcript_cluster_id', 'TRANSCRIPT_CLUSTER_ID', 'cluster_id',
+        'probeset_id', 'ProbeSetID', 'PROBESET_ID', 'ID'
+    ]
+    for cand in explicit_candidates:
+        if cand in gene_annotation.columns:
+            ann_id_set = normalize_id_values(gene_annotation[cand])
+            overlap = len(exp_ids & ann_id_set)
+            if overlap > best_overlap:
+                best_id_col, best_overlap = cand, overlap
+
+# 2) Choose gene symbol column: prefer rich annotation field containing symbols
+if 'SPOT_ID.1' in gene_annotation.columns:
+    best_gene_col = 'SPOT_ID.1'
+elif 'SPOT_ID' in gene_annotation.columns:
+    best_gene_col = 'SPOT_ID'
+else:
+    # Fallback: pick the column with most rows yielding at least one extractable human gene symbol
+    best_gene_col = None
+    best_symbol_hits = -1
+    sample_annotation = gene_annotation.sample(min(500, len(gene_annotation)), random_state=42)
+    for col in gene_annotation.columns:
+        try:
+            texts = sample_annotation[col].astype(str).fillna("")
+            hits = texts.map(lambda x: len(extract_human_gene_symbols(x)) > 0).sum()
+            if hits > best_symbol_hits:
+                best_symbol_hits = hits
+                best_gene_col = col
+        except Exception:
+            continue
+
+# Guard against identical selections
+if best_gene_col == best_id_col:
+    # Try switching to a preferred gene annotation column if available
+    if 'SPOT_ID.1' in gene_annotation.columns and best_id_col != 'SPOT_ID.1':
+        best_gene_col = 'SPOT_ID.1'
+    elif 'SPOT_ID' in gene_annotation.columns and best_id_col != 'SPOT_ID':
+        best_gene_col = 'SPOT_ID'
+    else:
+        # Pick the next best gene-like column
+        for col, _ in overlaps[1:]:
+            if col != best_id_col:
+                best_gene_col = col
+                break
+
+# Validate selections
+if best_id_col is None:
+    raise ValueError("Could not determine an annotation ID column matching the expression index.")
+
+# 3) Build mapping dataframe and validate overlap with expression IDs
+mapping_df = get_gene_mapping(gene_annotation, prob_col=best_id_col, gene_col=best_gene_col)
+
+# Normalize ID in mapping to improve matching with expression index
+mapping_df['ID'] = mapping_df['ID'].astype(str).str.strip().str.replace(r'\.0$', '', regex=True)
+# Also try to recover long digit tokens (>=7) when necessary
+def recover_long_digits(s):
+    if s.isdigit():
+        return s
+    gs = re.findall(r'\d{7,}', s)
+    return gs[0] if gs else s
+mapping_df['ID'] = mapping_df['ID'].map(recover_long_digits)
+
+overlap_final = mapping_df['ID'].isin(expr_df.index.astype(str)).sum()
+print(f"Selected ID column: {best_id_col}, gene column: {best_gene_col}, overlapping probes: {overlap_final}")
+
+# If no overlap, attempt alternative ID columns (next best by overlap)
+if overlap_final == 0:
+    tried = {best_id_col}
+    for col, ov in overlaps[1:]:
+        if col in tried:
+            continue
+        try:
+            alt_map = get_gene_mapping(gene_annotation, prob_col=col, gene_col=best_gene_col)
+            alt_map['ID'] = alt_map['ID'].astype(str).str.strip().str.replace(r'\.0$', '', regex=True)
+            alt_map['ID'] = alt_map['ID'].map(recover_long_digits)
+            ov2 = alt_map['ID'].isin(expr_df.index.astype(str)).sum()
+            if ov2 > 0:
+                mapping_df = alt_map
+                overlap_final = ov2
+                best_id_col = col
+                print(f"Switched to alternative ID column: {best_id_col} with overlap: {overlap_final}")
+                break
+            tried.add(col)
+        except Exception:
+            continue
+
+# Final check
+if overlap_final == 0:
+    # Provide diagnostics then stop clearly
+    cand_cols = [c for c, _ in overlaps[:10]]
+    raise ValueError(
+        f"No overlap between expression IDs (e.g., {list(list(exp_ids)[:5])}) "
+        f"and any annotation column. Top candidate columns tested: {cand_cols}"
+    )
+
+# 4) Apply mapping to convert probe-level data to gene-level data
+gene_data = apply_gene_mapping(expr_df, mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# Dynamic availability flags for final validation
+is_gene_available_dyn = (normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0)
+is_trait_available_dyn = trait in linked_data.columns
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Assess bias and remove biased demographic features
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort info
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available_dyn,
+    is_trait_available=is_trait_available_dyn,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)
+
+# Step 8: Data Normalization and Linking
+import os
+
+# 1. Normalize the obtained gene data and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link the clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# Dynamic availability flags based on generated data
+is_gene_available_dyn = (normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0)
+is_trait_available_dyn = trait in linked_data.columns
+
+# 3. Handle missing values in the linked data
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Determine whether the trait and some demographic features are severely biased, and remove biased features.
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Conduct quality check and save the cohort information.
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available_dyn,
+    is_trait_available=is_trait_available_dyn,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data
+)
+
+# 6. If the linked data is usable, save it as a CSV file to 'out_data_file'.
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)
+
+# Step 9: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import json
+import gzip
+import io
+import numpy as np
+import pandas as pd
+
+# Helper: extract value after colon and normalize
+def _extract_value(v):
+    if v is None:
+        return None
+    s = str(v).strip()
+    if s == "" or s.lower() in {"na", "n/a", "nan", "none", "null", "not available", "unknown"}:
+        return None
+    # Split at first colon to separate header from value
+    parts = s.split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    val = val.strip()
+    return val if val != "" else None
+
+def _lower(s):
+    return s.lower() if isinstance(s, str) else s
+
+# Heuristics to detect is_gene_available from series matrix header
+def detect_gene_availability_from_series_matrix(in_dir: str) -> bool:
+    try:
+        # Find a series matrix file
+        cand_files = []
+        for root, _, files in os.walk(in_dir):
+            for fn in files:
+                fn_low = fn.lower()
+                if "series_matrix" in fn_low and fn_low.endswith(".txt.gz"):
+                    cand_files.append(os.path.join(root, fn))
+        if not cand_files:
+            # Fall back to filename keywords if no series matrix
+            flags = {"methylation": False, "mirna": False, "expression": False}
+            for root, _, files in os.walk(in_dir):
+                for fn in files:
+                    fl = fn.lower()
+                    if "methyl" in fl or "bisulfite" in fl or "450k" in fl or "850k" in fl or "wgbs" in fl:
+                        flags["methylation"] = True
+                    if "mirna" in fl or "micro-rna" in fl or "microRNA".lower() in fl:
+                        flags["mirna"] = True
+                    if "expression" in fl or "rnaseq" in fl or "transcript" in fl or "counts" in fl or "series_matrix" in fl:
+                        flags["expression"] = True
+            if flags["methylation"] and not flags["expression"]:
+                return False
+            if flags["mirna"] and not flags["expression"]:
+                return False
+            # If expression uncertain, default to True (most GEO H&N cohorts are gene expression)
+            return True
+
+        path = cand_files[0]
+        with gzip.open(path, 'rt', encoding='utf-8', errors='ignore') as f:
+            header = []
+            for _ in range(400):
+                line = f.readline()
+                if not line:
+                    break
+                header.append(line.strip())
+
+        header_text = "\n".join(header).lower()
+        # Strong negative signals
+        if "methylation profiling" in header_text or "methylation" in header_text:
+            return False
+        if "mirna" in header_text and "mrna" not in header_text and "messenger rna" not in header_text:
+            return False
+        # Positive signals
+        if "expression profiling" in header_text or "rna sequencing" in header_text or "high throughput sequencing" in header_text:
+            return True
+        # Molecule channel hints
+        if "molecule_ch1" in header_text:
+            if "genomic dna" in header_text:
+                return False
+            if "total rna" in header_text or "poly a rna" in header_text or "poly(a) rna" in header_text or "mrna" in header_text:
+                return True
+        # Default lean positive
+        return True
+    except Exception:
+        # On any parsing issue, lean positive
+        return True
+
+# Attempt to access clinical_data provided by previous steps
+clinical_data_available = 'clinical_data' in globals() or 'clinical_data' in locals()
+if not clinical_data_available:
+    clinical_data = None
+
+trait_row = None
+age_row = None
+gender_row = None
+
+def _find_rows_from_clinical(clin_df: pd.DataFrame):
+    candidate_trait = None
+    candidate_gender = None
+    candidate_age = None
+
+    # Build per-row normalized unique values and headers
+    for ridx in list(clin_df.index):
+        try:
+            row_vals = clin_df.loc[ridx].tolist()
+        except Exception:
+            continue
+        extracted_vals = []
+        for v in row_vals:
+            val = _extract_value(v)
+            if val is None:
+                continue
+            extracted_vals.append(val)
+
+        if len(extracted_vals) == 0:
+            continue
+
+        # Lowercase uniques for matching
+        low_uniques = set([_lower(v) for v in extracted_vals])
+
+        # Gender detection
+        if any(re.search(r'\bmale\b', lv) for lv in low_uniques) or any(re.search(r'\bfemale\b', lv) for lv in low_uniques) or any(lv in {"m","f"} for lv in low_uniques):
+            # Ensure not constant
+            gset = set()
+            for lv in low_uniques:
+                if re.fullmatch(r'm|male|males', lv):
+                    gset.add("male")
+                elif re.fullmatch(r'f|female|females', lv):
+                    gset.add("female")
+            if len(gset) >= 2:
+                candidate_gender = ridx
+
+        # Age detection
+        numeric_vals = []
+        for lv in extracted_vals:
+            lv_low = lv.lower()
+            # Try to extract a float
+            m = re.search(r'(\d+(\.\d+)?)', lv_low)
+            if m:
+                try:
+                    num = float(m.group(1))
+                    # Exclude clearly non-age values (e.g., 0, 1 used for binary)
+                    numeric_vals.append(num)
+                except Exception:
+                    pass
+        if len(numeric_vals) >= max(3, int(0.6 * len(extracted_vals))) and (np.max(numeric_vals) - np.min(numeric_vals) >= 5):
+            # Also sanity: check common age hints in text
+            if any(k in " ".join([str(x).lower() for x in extracted_vals]) for k in ["age", "years", "y/o", "year-old", "age at", "age(year"]):
+                candidate_age = ridx
+
+        # Trait detection: look for tumor/normal/cancer/control labels
+        trait_positive_terms = ["tumor", "tumour", "cancer", "carcinoma", "scc", "hnscc", "hn", "metastasis", "metastatic", "primary", "recurrent"]
+        trait_negative_terms = ["normal", "adjacent", "healthy", "control", "benign", "non-tumor", "non tumour", "noncancer"]
+        pos = any(any(t in lv for t in trait_positive_terms) for lv in low_uniques)
+        neg = any(any(t in lv for t in trait_negative_terms) for lv in low_uniques)
+        if pos or neg:
+            # Ensure not constant after mapping to binary
+            mapped = set()
+            for lv in low_uniques:
+                if any(t in lv for t in trait_negative_terms):
+                    mapped.add(0)
+                elif any(t in lv for t in trait_positive_terms):
+                    mapped.add(1)
+            if len(mapped) >= 2:
+                candidate_trait = ridx
+
+    return candidate_trait, candidate_age, candidate_gender
+
+if clinical_data is not None and isinstance(clinical_data, pd.DataFrame) and len(clinical_data.index) > 0:
+    try:
+        trow, arow, grow = _find_rows_from_clinical(clinical_data)
+        trait_row = trow
+        age_row = arow
+        gender_row = grow
+    except Exception:
+        trait_row = None
+        age_row = None
+        gender_row = None
+else:
+    trait_row = None
+    age_row = None
+    gender_row = None
+
+# Conversion functions
+def convert_trait(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    s = v.strip().lower()
+    # Negative/control
+    neg_terms = ["normal", "adjacent", "healthy", "control", "benign", "non-tumor", "non tumour", "noncancer", "non-cancer"]
+    if any(t in s for t in neg_terms):
+        return 0
+    # Positive/case
+    pos_terms = ["tumor", "tumour", "cancer", "carcinoma", "scc", "hnscc", "metastasis", "metastatic", "primary", "recurrent"]
+    if any(t in s for t in pos_terms):
+        return 1
+    # Some projects encode as case/control directly
+    if s in {"case"}:
+        return 1
+    if s in {"control"}:
+        return 0
+    return None
+
+def convert_age(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    s = v.strip().lower()
+    m = re.search(r'(\d+(\.\d+)?)', s)
+    if not m:
+        return None
+    try:
+        age = float(m.group(1))
+        # clamp non-sense ages
+        if age < 0 or age > 120:
+            return None
+        return age
+    except Exception:
+        return None
+
+def convert_gender(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    s = v.strip().lower()
+    # Standard mapping: female -> 0, male -> 1
+    if s in {"female", "f", "woman", "women"}:
+        return 0
+    if s in {"male", "m", "man", "men"}:
+        return 1
+    return None
+
+# 1) Gene expression availability
+is_gene_available = detect_gene_availability_from_series_matrix(in_cohort_dir)
+
+# 2) Trait availability is based on trait_row detection
+is_trait_available = trait_row is not None
+
+# 3) Save metadata (initial filtering)
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction if trait_row available
+if is_trait_available and (clinical_data is not None) and isinstance(clinical_data, pd.DataFrame):
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age if age_row is not None else None,
+        gender_row=gender_row,
+        convert_gender=convert_gender if gender_row is not None else None
+    )
+    # Preview
+    clinical_preview = preview_df(selected_clinical_df, n=5)
+    print("Preview of selected clinical features:", clinical_preview)
+
+    # Save to CSV
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+else:
+    print("Clinical trait data not available; skipping clinical feature extraction.")
+
+# Step 10: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+from typing import Optional
+
+# 1) Detect gene expression data availability by scanning cohort directory
+def detect_gene_expression_availability(directory: str) -> bool:
+    if not os.path.isdir(directory):
+        return False
+
+    all_files = []
+    for root, _, files in os.walk(directory):
+        for f in files:
+            all_files.append(os.path.join(root, f).lower())
+
+    if not all_files:
+        return False
+
+    # Exclusion patterns
+    methylation_terms = ['methyl', '450k', 'epic', 'idat', 'cpg', 'betavalue', 'beta-value']
+    mirna_terms = ['mirna', 'micro_rna', 'microrna']
+
+    # Inclusion patterns suggestive of gene expression
+    series_matrix_terms = ['series_matrix.txt', 'series_matrix.txt.gz']
+    expression_terms = [
+        'expression', 'expr', 'rsem', 'htseq', 'featurecount', 'feature_counts',
+        'fpkm', 'tpm', 'counts', 'readcount', 'dge', 'salmon', 'kallisto', 'abundance'
+    ]
+
+    has_methylation = any(any(t in f for t in methylation_terms) for f in all_files)
+    if has_methylation:
+        return False
+
+    has_mirna = any(any(t in f for t in mirna_terms) for f in all_files)
+    has_series_matrix = any(any(t in f for t in series_matrix_terms) for f in all_files)
+    has_expression_like = any(any(t in f for t in expression_terms) for f in all_files)
+
+    # If only miRNA-related files are present without general expression indicators, treat as not suitable
+    if has_mirna and not (has_series_matrix or has_expression_like):
+        return False
+
+    return has_series_matrix or has_expression_like
+
+is_gene_available = detect_gene_expression_availability(in_cohort_dir)
+
+# 2) Variable availability and conversion functions
+# Based on previous step output: "Clinical trait data not available", set trait_row to None.
+trait_row: Optional[int] = None
+age_row: Optional[int] = None
+gender_row: Optional[int] = None
+
+def _after_colon(x):
+    if x is None:
+        return None
+    s = str(x)
+    parts = s.split(':', 1)
+    val = parts[1] if len(parts) == 2 else parts[0]
+    val = val.strip()
+    return val if val not in ['', 'na', 'n/a', 'none', 'null', 'nan'] else None
+
+# Heuristic converters (will only be used if corresponding rows are available)
+def convert_trait(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    v_low = v.lower()
+    # Map common case/control labels
+    case_tokens = ['case', 'tumor', 'tumour', 'cancer', 'hnscc', 'head and neck', 'scc', 'carcinoma']
+    control_tokens = ['control', 'normal', 'adjacent normal', 'non-tumor', 'non tumour', 'healthy']
+    if any(t in v_low for t in case_tokens):
+        return 1
+    if any(t in v_low for t in control_tokens):
+        return 0
+    return None
+
+def convert_age(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    # Extract the first integer or float number found
+    m = re.search(r'(\d+(\.\d+)?)', v)
+    if not m:
+        return None
+    try:
+        return float(m.group(1))
+    except Exception:
+        return None
+
+def convert_gender(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    v_low = v.lower()
+    if v_low in ['female', 'f', 'woman', 'women']:
+        return 0
+    if v_low in ['male', 'm', 'man', 'men']:
+        return 1
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (skip because trait_row is None)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    clinical_preview = preview_df(selected_clinical_df)
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 11: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import json
+import pandas as pd
+import numpy as np
+
+# 1) Determine gene expression availability using lightweight heuristics
+def detect_gene_expression_availability(in_dir: str) -> bool:
+    if not os.path.isdir(in_dir):
+        return False
+    files = os.listdir(in_dir)
+    if not files:
+        return False
+
+    # If any typical gene expression matrix file is present (and not clearly miRNA/methylation)
+    expr_like = ['series_matrix', 'matrix', 'expression', 'counts', 'rsem', 'tpm', 'fpkm', 'readcounts']
+    exclude_like = ['mirna', 'microRNA', 'methyl', '450k', 'epic', 'methylation', 'metabol', 'proteom']
+    for f in files:
+        fl = f.lower()
+        if any(x in fl for x in expr_like):
+            if not any(x in fl for x in exclude_like):
+                return True
+    # Fallback: if directory has many GSM/GSE files, assume expression likely available
+    gsm_like = [f for f in files if f.lower().startswith(('gsm', 'gse'))]
+    return len(gsm_like) > 0
+
+is_gene_available = detect_gene_expression_availability(in_cohort_dir)
+
+# 2) Variable availability and conversion
+clinical_data = globals().get("clinical_data", None)
+
+def _split_cell(cell: any):
+    if pd.isna(cell):
+        return "", ""
+    s = str(cell).strip()
+    if ':' in s:
+        k, v = s.split(':', 1)
+        return k.strip().lower(), v.strip().lower()
+    return "", s.lower()
+
+def _row_values_and_keys(df: pd.DataFrame, row_id) -> (list, list):
+    row = df.loc[row_id]
+    keys, vals = [], []
+    for cell in row:
+        k, v = _split_cell(cell)
+        if v != "":
+            vals.append(v)
+        if k != "":
+            keys.append(k)
+    return keys, vals
+
+def _has_variation(values: list) -> bool:
+    uniq = set([v for v in values if v not in ("", "na", "n/a", "nan", "none", "unknown", "not available")])
+    return len(uniq) > 1
+
+trait_row = None
+age_row = None
+gender_row = None
+
+# Define conversion functions
+case_terms = [
+    'cancer', 'carcinoma', 'tumor', 'tumour', 'hnscc', 'oscc', 'scc', 'hnsc',
+    'malignant', 'neoplasm', 'lesion', 'metastasis', 'head and neck', 'oral cancer',
+    'squamous'
+]
+control_terms = ['normal', 'control', 'adjacent normal', 'non-tumor', 'nontumor', 'healthy', 'benign', 'noncancer']
+
+def convert_trait(x):
+    k, v = _split_cell(x)
+    if v == "":
+        return None
+    text = v.lower()
+    # Heuristic mapping
+    is_case = any(t in text for t in case_terms)
+    is_ctrl = any(t in text for t in control_terms)
+    if is_case and not is_ctrl:
+        return 1
+    if is_ctrl and not is_case:
+        return 0
+    # Exact words male/female should not be here; return None if ambiguous
+    return None
+
+def convert_age(x):
+    k, v = _split_cell(x)
+    if v == "" or v in ("na", "n/a", "nan", "none", "unknown", "not available"):
+        return None
+    # Extract first number (years)
+    m = re.search(r'(\d+(\.\d+)?)', v)
+    if not m:
+        return None
+    try:
+        val = float(m.group(1))
+        # Clip unreasonable ages
+        if 0 <= val <= 120:
+            return val
+    except Exception:
+        pass
+    return None
+
+def convert_gender(x):
+    k, v = _split_cell(x)
+    if v == "":
+        return None
+    v = v.lower()
+    # Normalize single-letter representations
+    if v in ("m", "male", "man", "boy"):
+        return 1
+    if v in ("f", "female", "woman", "girl"):
+        return 0
+    # Occasionally v may be things like 'gender: M/F'
+    if 'male' in v and 'female' not in v:
+        return 1
+    if 'female' in v and 'male' not in v:
+        return 0
+    return None
+
+def _score_trait_row(keys, vals):
+    # Require both case and control presence for trait row
+    case_count = sum(any(t in v for t in case_terms) for v in vals)
+    ctrl_count = sum(any(t in v for t in control_terms) for v in vals)
+    if min(case_count, ctrl_count) == 0:
+        return -1  # invalid
+    # Prefer informative keys
+    key_bonus = 0
+    key_text = ' '.join(set(keys))
+    for kw in ['disease', 'status', 'group', 'phenotype', 'tissue', 'tumor', 'diagnosis', 'condition', 'case', 'control']:
+        if kw in key_text:
+            key_bonus += 2
+    return (case_count + ctrl_count) + 3 * min(case_count, ctrl_count) + key_bonus
+
+def _looks_like_age(keys, vals):
+    key_text = ' '.join(set(keys))
+    if 'age' in key_text:
+        return True
+    # If majority of cells look numeric with 'year'
+    numeric_like = 0
+    for v in vals:
+        if re.search(r'\d', v):
+            numeric_like += 1
+    return numeric_like >= max(2, int(0.5 * len(vals)))
+
+def _looks_like_gender(keys, vals):
+    key_text = ' '.join(set(keys))
+    if ('gender' in key_text) or ('sex' in key_text):
+        return True
+    gender_like = 0
+    for v in vals:
+        if v in ("m", "male", "man", "boy", "f", "female", "woman", "girl") or ('male' in v) or ('female' in v):
+            gender_like += 1
+    return gender_like >= max(2, int(0.5 * len(vals)))
+
+if isinstance(clinical_data, pd.DataFrame) and clinical_data.shape[0] > 0:
+    # Identify rows
+    trait_candidates = []
+    age_candidates = []
+    gender_candidates = []
+
+    for row_id in clinical_data.index:
+        try:
+            keys, vals = _row_values_and_keys(clinical_data, row_id)
+        except Exception:
+            continue
+        # Trait candidates
+        score = _score_trait_row(keys, vals)
+        if score >= 0 and _has_variation([convert_trait(v) for v in clinical_data.loc[row_id].values]):
+            trait_candidates.append((row_id, score))
+        # Age candidates
+        if _looks_like_age(keys, vals):
+            # Variation after numeric conversion
+            converted = [convert_age(v) for v in clinical_data.loc[row_id].values]
+            if _has_variation([str(c) for c in converted if c is not None]):
+                age_candidates.append(row_id)
+        # Gender candidates
+        if _looks_like_gender(keys, vals):
+            converted = [convert_gender(v) for v in clinical_data.loc[row_id].values]
+            # Require both sexes present
+            if (1 in converted) and (0 in converted):
+                gender_candidates.append(row_id)
+
+    # Select best trait_row by score
+    if trait_candidates:
+        trait_candidates.sort(key=lambda x: x[1], reverse=True)
+        trait_row = trait_candidates[0][0]
+
+    # If multiple age rows, prefer one with 'age' in keys explicitly
+    if age_candidates:
+        if len(age_candidates) == 1:
+            age_row = age_candidates[0]
+        else:
+            # Choose the one with explicit 'age' mention in keys
+            winners = []
+            for rid in age_candidates:
+                keys, _ = _row_values_and_keys(clinical_data, rid)
+                if any('age' in k for k in keys):
+                    winners.append(rid)
+            age_row = winners[0] if winners else age_candidates[0]
+
+    # If multiple gender rows, prefer explicit 'gender'/'sex'
+    if gender_candidates:
+        if len(gender_candidates) == 1:
+            gender_row = gender_candidates[0]
+        else:
+            winners = []
+            for rid in gender_candidates:
+                keys, _ = _row_values_and_keys(clinical_data, rid)
+                if any(('gender' in k) or ('sex' in k) for k in keys):
+                    winners.append(rid)
+            gender_row = winners[0] if winners else gender_candidates[0]
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical Feature Extraction and preview/save if available
+if is_trait_available and isinstance(clinical_data, pd.DataFrame):
+    selected_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=int(trait_row),
+        convert_trait=convert_trait,
+        age_row=int(age_row) if age_row is not None else None,
+        convert_age=convert_age if age_row is not None else None,
+        gender_row=int(gender_row) if gender_row is not None else None,
+        convert_gender=convert_gender if gender_row is not None else None
+    )
+    clinical_preview = preview_df(selected_df, n=5, max_items=200)
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_df.to_csv(out_clinical_data_file)
+
+# Step 12: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1) Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2) Link clinical and genetic data (guard for missing clinical features)
+clinical_df = None
+if 'selected_clinical_df' in globals() and isinstance(selected_clinical_df, pd.DataFrame):
+    clinical_df = selected_clinical_df
+elif os.path.exists(out_clinical_data_file):
+    try:
+        clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+    except Exception:
+        clinical_df = None
+
+if isinstance(clinical_df, pd.DataFrame) and clinical_df.shape[0] > 0:
+    linked_data = geo_link_clinical_genetic_data(clinical_df, normalized_gene_data)
+else:
+    linked_data = pd.DataFrame()
+
+# Dynamic availability flags
+is_gene_available_dyn = (normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0)
+is_trait_available_dyn = (isinstance(linked_data, pd.DataFrame) and (len(linked_data.columns) > 0) and (trait in linked_data.columns))
+
+# 3) Handle missing values only if trait column is available
+if is_trait_available_dyn:
+    linked_data = handle_missing_values(linked_data, trait)
+    # 4) Assess bias and remove biased demographic features
+    is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+else:
+    # No trait available; construct empty df for validation
+    unbiased_linked_data = pd.DataFrame()
+    is_trait_biased = False  # placeholder; dataset will be marked unavailable due to missing trait
+
+# 5) Final validation and save cohort info
+note = ""
+if not is_trait_available_dyn:
+    note = "INFO: Trait not available in clinical annotations; unable to link clinical and genetic data."
+
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available_dyn,
+    is_trait_available=is_trait_available_dyn,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6) Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)
+
+# Step 13: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import numpy as np
+import pandas as pd
+
+# 1) Gene expression data availability (heuristic: assume TRUE for GEO series with parsed matrix)
+is_gene_available = True
+
+# 2) Variable availability scanning on clinical_data
+def _after_colon(x):
+    if x is None or (isinstance(x, float) and np.isnan(x)):
+        return None
+    s = str(x)
+    parts = s.split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    val = val.strip()
+    return val if val != "" else None
+
+def _clean_str(x):
+    if x is None:
+        return None
+    s = str(x).strip().lower()
+    s = re.sub(r'\s+', ' ', s)
+    return s
+
+def _parse_age(x):
+    if x is None:
+        return None
+    s = _clean_str(_after_colon(x))
+    if s is None or s in {"na", "n/a", "nan", "none", "unknown", ""}:
+        return None
+    # find a plausible age number
+    m = re.search(r'(\d+(\.\d+)?)', s)
+    if not m:
+        return None
+    try:
+        age = float(m.group(1))
+    except:
+        return None
+    if 0 <= age <= 120:
+        return age
+    return None
+
+def _parse_gender(x):
+    s = _clean_str(_after_colon(x))
+    if s is None:
+        return None
+    # common patterns
+    if s in {"male", "m", "man", "men"}:
+        return 1
+    if s in {"female", "f", "woman", "women"}:
+        return 0
+    # Sometimes values like "sex: Male", handled by _after_colon already
+    # handle single letters with extra text
+    if re.search(r'\bmale\b', s):
+        return 1
+    if re.search(r'\bfemale\b', s):
+        return 0
+    if re.fullmatch(r'm', s):
+        return 1
+    if re.fullmatch(r'f', s):
+        return 0
+    return None
+
+positive_terms = {
+    "cancer", "carcinoma", "tumor", "tumour", "squamous cell carcinoma",
+    "scc", "hnscc", "head and neck cancer", "oral cavity cancer",
+    "oropharyngeal carcinoma", "malignant", "metastatic", "primary tumor",
+    "primary tumour"
+}
+negative_terms = {
+    "normal", "non-tumor", "non tumour", "healthy", "control", "adjacent normal",
+    "benign", "normal tissue", "adjacent noncancerous", "adjacent non-cancerous"
+}
+case_terms = {"case"}
+control_terms = {"control"}
+
+def _trait_label_from_value(x):
+    s = _clean_str(_after_colon(x))
+    if s is None:
+        return None
+    # map to binary: presence of head and neck cancer = 1, absence = 0
+    # check explicit case/control first
+    if s in case_terms:
+        return 1
+    if s in control_terms:
+        return 0
+    # general positive/negative term containment
+    if any(t in s for t in positive_terms):
+        return 1
+    if any(t in s for t in negative_terms):
+        return 0
+    # sometimes values like "tumor", "normal"
+    if re.search(r'\btumou?r\b', s):
+        return 1
+    if re.search(r'\bnormal\b', s):
+        return 0
+    return None
+
+# Access clinical_data if available
+trait_row = None
+age_row = None
+gender_row = None
+
+if 'clinical_data' in globals() and isinstance(clinical_data, pd.DataFrame) and len(clinical_data) > 0:
+    # Identify candidate rows by scanning variability and recognizable patterns
+    trait_candidates = []
+    age_candidates = []
+    gender_candidates = []
+
+    for ridx in clinical_data.index:
+        vals = list(clinical_data.loc[ridx].values)
+        # Trait candidates: check if both 0 and 1 present after mapping
+        mapped = [ _trait_label_from_value(v) for v in vals ]
+        recognized = [m for m in mapped if m is not None]
+        if len(set(recognized)) >= 2:
+            # keep count of recognized values to rank
+            trait_candidates.append((ridx, len(recognized)))
+
+        # Age candidates: parse ages and check variability
+        ages = [ _parse_age(v) for v in vals ]
+        ages_clean = [a for a in ages if a is not None]
+        if len(set(ages_clean)) >= 2 and len(ages_clean) >= max(2, int(0.5 * len(vals))):
+            age_candidates.append((ridx, len(ages_clean)))
+
+        # Gender candidates: parse genders and check variability
+        genders = [ _parse_gender(v) for v in vals ]
+        genders_clean = [g for g in genders if g is not None]
+        if len(set(genders_clean)) >= 2 and len(genders_clean) >= max(2, int(0.5 * len(vals))):
+            gender_candidates.append((ridx, len(genders_clean)))
+
+    # Choose the best candidates by max recognized count
+    if trait_candidates:
+        trait_row = sorted(trait_candidates, key=lambda x: x[1], reverse=True)[0][0]
+    if age_candidates:
+        age_row = sorted(age_candidates, key=lambda x: x[1], reverse=True)[0][0]
+    if gender_candidates:
+        gender_row = sorted(gender_candidates, key=lambda x: x[1], reverse=True)[0][0]
+
+    # If trait row exists but constant after mapping (edge case), drop it
+    if trait_row is not None:
+        vals = list(clinical_data.loc[trait_row].values)
+        mapped = [ _trait_label_from_value(v) for v in vals ]
+        rec = [m for m in mapped if m is not None]
+        if len(set(rec)) <= 1:
+            trait_row = None
+
+# 2.2 Converters
+def convert_trait(x):
+    return _trait_label_from_value(x)
+
+def convert_age(x):
+    return _parse_age(x)
+
+def convert_gender(x):
+    return _parse_gender(x)
+
+# 3) Save initial metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical Feature Extraction if available
+if is_trait_available:
+    selected_clinical = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age if age_row is not None else None,
+        gender_row=gender_row,
+        convert_gender=convert_gender if gender_row is not None else None
+    )
+    _ = preview_df(selected_clinical)
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical.to_csv(out_clinical_data_file)
+
+# Step 14: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import json
+import gzip
+from io import TextIOWrapper
+from typing import Optional, Callable, List, Tuple, Dict, Any
+
+import pandas as pd
+import numpy as np
+
+# ---------------------------
+# Helpers
+# ---------------------------
+
+def _open_text_file(path):
+    if path.endswith(".gz"):
+        return TextIOWrapper(gzip.open(path, "rb"), encoding="utf-8", errors="ignore")
+    return open(path, "r", encoding="utf-8", errors="ignore")
+
+def is_gene_expression_available(in_dir: str) -> bool:
+    # Heuristic: inspect available text-like files for platform/assay hints
+    if not os.path.isdir(in_dir):
+        return False
+    text_exts = (".txt", ".csv", ".tsv", ".soft", ".series_matrix", ".json")
+    candidates = []
+    for root, _, files in os.walk(in_dir):
+        for f in files:
+            lower = f.lower()
+            if lower.endswith(text_exts) or lower.endswith(tuple(e + ".gz" for e in text_exts)):
+                candidates.append(os.path.join(root, f))
+
+    if not candidates:
+        return False
+
+    has_expression_hint = False
+    has_exclusion_hint = False
+    for path in candidates:
+        try:
+            with _open_text_file(path) as fh:
+                head = "".join([next(fh) for _ in range(200)])
+        except StopIteration:
+            pass
+        except Exception:
+            continue
+        s = head.lower()
+        # Exclusion: miRNA-only or methylation
+        if ("mirna" in s or "micro rna" in s or "micro-rna" in s or "microrna" in s) and ("mrna" not in s and "rna-seq" not in s and "transcript" not in s):
+            has_exclusion_hint = True
+        if ("methylation" in s) or ("cpg" in s) or ("illumina humanmethylation" in s):
+            has_exclusion_hint = True
+        # Inclusion hints for gene expression
+        if ("rna-seq" in s) or ("mrna" in s) or ("expression profiling by array" in s) or ("transcript" in s) or ("series_matrix_table_begin" in s):
+            has_expression_hint = True
+
+    if has_exclusion_hint and not has_expression_hint:
+        return False
+    # Default to True if we saw any plausible expression hints or cannot decisively exclude
+    return True
+
+def try_get_clinical_df() -> Optional[pd.DataFrame]:
+    # Use pre-existing variable if available
+    try:
+        if isinstance(clinical_data, pd.DataFrame):
+            return clinical_data
+    except NameError:
+        pass
+
+    # Try common filenames in the cohort directory
+    candidates = [
+        os.path.join(in_cohort_dir, "clinical_data.csv"),
+        os.path.join(in_cohort_dir, "clinical.csv"),
+        os.path.join(in_cohort_dir, "samples_clinical.csv"),
+        os.path.join(in_cohort_dir, "clinical_data.tsv"),
+        os.path.join(in_cohort_dir, "clinical_data.parquet"),
+        os.path.join(in_cohort_dir, "clinical_data.pkl"),
+    ]
+    for p in candidates:
+        if os.path.exists(p):
+            try:
+                if p.endswith(".csv"):
+                    return pd.read_csv(p, index_col=0)
+                if p.endswith(".tsv"):
+                    return pd.read_csv(p, sep="\t", index_col=0)
+                if p.endswith(".parquet"):
+                    return pd.read_parquet(p)
+                if p.endswith(".pkl"):
+                    return pd.read_pickle(p)
+            except Exception:
+                continue
+    return None
+
+def _parse_value(cell: Any) -> str:
+    if pd.isna(cell):
+        return ""
+    s = str(cell).strip()
+    # GEO often uses "key: value"
+    if ":" in s:
+        s = s.split(":", 1)[1].strip()
+    return s
+
+def _normalize(s: str) -> str:
+    return re.sub(r"\s+", " ", s.strip().lower())
+
+def detect_age_row(df: pd.DataFrame) -> Optional[int]:
+    candidates: List[Tuple[int, int, int]] = []  # (row_idx, n_numeric_unique, n_age_mentions)
+    for ridx in df.index:
+        vals = [str(v) for v in df.loc[ridx, :].tolist()]
+        vals_lower = [_normalize(v) for v in vals]
+        age_mentions = sum(("age" in v) for v in vals_lower)
+        # Extract numeric ages
+        nums = []
+        for v in vals_lower:
+            core = _parse_value(v)
+            # extract first number (potentially decimals)
+            m = re.search(r"(\d+(\.\d+)?)", core)
+            if m:
+                try:
+                    nums.append(float(m.group(1)))
+                except Exception:
+                    nums.append(np.nan)
+            else:
+                nums.append(np.nan)
+        nums = [x for x in nums if pd.notna(x)]
+        if len(set(nums)) >= 2 and age_mentions > 0:
+            candidates.append((ridx, len(set(nums)), age_mentions))
+    if not candidates:
+        return None
+    # Prefer more age mentions, then more numeric variability
+    candidates.sort(key=lambda x: (x[2], x[1]), reverse=True)
+    return candidates[0][0]
+
+def detect_gender_row(df: pd.DataFrame) -> Optional[int]:
+    def to_gender(v: str) -> Optional[int]:
+        x = _normalize(_parse_value(v))
+        if x in ("male", "m"):
+            return 1
+        if x in ("female", "f"):
+            return 0
+        # common variants
+        if "male" in x and "female" not in x:
+            return 1
+        if "female" in x and "male" not in x:
+            return 0
+        if "sex" in x:
+            if "male" in x:
+                return 1
+            if "female" in x:
+                return 0
+        return None
+
+    candidates: List[Tuple[int, int]] = []  # (row_idx, n_valid)
+    for ridx in df.index:
+        vals = [str(v) for v in df.loc[ridx, :].tolist()]
+        mapped = [to_gender(v) for v in vals]
+        valid = [m for m in mapped if m is not None]
+        if len(valid) >= 2 and len(set(valid)) == 2:  # both sexes present
+            candidates.append((ridx, len(valid)))
+    if not candidates:
+        return None
+    candidates.sort(key=lambda x: x[1], reverse=True)
+    return candidates[0][0]
+
+def detect_trait_row(df: pd.DataFrame) -> Optional[int]:
+    # We want binary case/control for Head_and_Neck_Cancer
+    cancer_kw = ["cancer", "carcinoma", "tumor", "tumour", "squamous", "hnscc", "oscc", "hn", "head and neck"]
+    control_kw = ["normal", "adjacent", "non-tumor", "nontumor", "noncancer", "control", "healthy", "benign"]
+
+    def to_trait(v: str) -> Optional[int]:
+        x = _normalize(_parse_value(v))
+        # explicit NA
+        if x in ("", "na", "n/a", "none", "unknown"):
+            return None
+        pos = any(kw in x for kw in cancer_kw)
+        neg = any(kw in x for kw in control_kw)
+        if pos and not neg:
+            return 1
+        if neg and not pos:
+            return 0
+        # group labels like case/control
+        if "case" in x and "control" not in x:
+            return 1
+        if "control" in x and "case" not in x:
+            return 0
+        # source_name_ch1 or tissue labels that imply tumor/normal
+        return None
+
+    candidates: List[Tuple[int, int, int]] = []  # (row_idx, n_valid, n_keyword_hits)
+    for ridx in df.index:
+        vals = [str(v) for v in df.loc[ridx, :].tolist()]
+        mapped = [to_trait(v) for v in vals]
+        valid = [m for m in mapped if m is not None]
+        if len(valid) >= 2 and len(set(valid)) == 2:
+            # keyword hits for ranking
+            vals_lower = [_normalize(_parse_value(v)) for v in vals]
+            kw_hits = sum(any(kw in v for kw in cancer_kw + control_kw) for v in vals_lower)
+            candidates.append((ridx, len(valid), kw_hits))
+    if not candidates:
+        return None
+    candidates.sort(key=lambda x: (x[2], x[1]), reverse=True)
+    return candidates[0][0]
+
+# ---------------------------
+# Step 1: Gene expression availability
+# ---------------------------
+is_gene_available = is_gene_expression_available(in_cohort_dir)
+
+# ---------------------------
+# Step 2: Variable availability and conversion functions
+# ---------------------------
+clinical_df = try_get_clinical_df()
+
+trait_row: Optional[int] = None
+age_row: Optional[int] = None
+gender_row: Optional[int] = None
+
+if clinical_df is not None and clinical_df.shape[0] > 0 and clinical_df.shape[1] > 0:
+    # Ensure row index are ints if possible; otherwise leave as is but we will return integer-like labels where applicable
+    # However, geo_select_clinical_features expects integer row identifiers that match the df index.
+    # We'll attempt to cast to int when safe; else keep as is.
+    try:
+        clinical_df.index = clinical_df.index.astype(int)
+    except Exception:
+        pass
+
+    # Detect rows
+    trait_row = detect_trait_row(clinical_df)
+    age_row = detect_age_row(clinical_df)
+    gender_row = detect_gender_row(clinical_df)
+
+# Conversion functions
+def convert_trait(x: Any) -> Optional[int]:
+    s = _normalize(_parse_value(x))
+    if s in ("", "na", "n/a", "none", "unknown"):
+        return None
+    pos_kw = ["cancer", "carcinoma", "tumor", "tumour", "squamous", "hnscc", "oscc", "malignant"]
+    neg_kw = ["normal", "adjacent", "non-tumor", "nontumor", "noncancer", "control", "healthy", "benign"]
+    pos = any(k in s for k in pos_kw)
+    neg = any(k in s for k in neg_kw)
+    if pos and not neg:
+        return 1
+    if neg and not pos:
+        return 0
+    if "case" in s and "control" not in s:
+        return 1
+    if "control" in s and "case" not in s:
+        return 0
+    # If dataset is case-only or control-only, return None to avoid constant feature
+    return None
+
+def convert_age(x: Any) -> Optional[float]:
+    s = _normalize(_parse_value(x))
+    if s in ("", "na", "n/a", "none", "unknown"):
+        return None
+    m = re.search(r"(\d+(\.\d+)?)", s)
+    if not m:
+        return None
+    try:
+        age_val = float(m.group(1))
+        # Heuristic sanity bounds
+        if 0 <= age_val <= 120:
+            return age_val
+        return None
+    except Exception:
+        return None
+
+def convert_gender(x: Any) -> Optional[int]:
+    s = _normalize(_parse_value(x))
+    if s in ("", "na", "n/a", "none", "unknown"):
+        return None
+    if s in ("male", "m"):
+        return 1
+    if s in ("female", "f"):
+        return 0
+    if "male" in s and "female" not in s:
+        return 1
+    if "female" in s and "male" not in s:
+        return 0
+    if "man" in s and "woman" not in s:
+        return 1
+    if "woman" in s and "man" not in s:
+        return 0
+    return None
+
+# If no clinical_df, we cannot have trait_row
+is_trait_available = trait_row is not None
+
+# ---------------------------
+# Step 3: Save metadata (initial filtering)
+# ---------------------------
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# ---------------------------
+# Step 4: Clinical feature extraction (only if trait_row is available)
+# ---------------------------
+if is_trait_available and clinical_df is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_df,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age if age_row is not None else None,
+        gender_row=gender_row,
+        convert_gender=convert_gender if gender_row is not None else None
+    )
+    preview = preview_df(selected_clinical_df, n=5)
+    print("Selected clinical features preview:", preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)

output/preprocess/Head_and_Neck_Cancer/code/GSE151181.py

@@ -0,0 +1,318 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE151181"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE151181"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE151181.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE151181.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE151181.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import pandas as pd
+
+# 1) Gene expression availability
+is_gene_available = True  # Title mentions "Gene and miRNA expression"; not pure miRNA-only.
+
+# 2) Variable availability and converters
+
+# Trait availability:
+# Use 'tissue type' (row 1): cancer vs non-neoplastic thyroid
+trait_row = 1
+
+# Age and gender are not available in the provided characteristics
+age_row = None
+gender_row = None
+
+def _extract_value(x):
+    if x is None or (isinstance(x, float) and pd.isna(x)):
+        return None
+    s = str(x)
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    return s.strip()
+
+def convert_trait(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    lv = v.lower()
+    # Control indicators
+    if 'non-neoplastic' in lv or lv in {'normal', 'normal thyroid', 'adjacent normal'}:
+        return 0
+    # Tumor/metastasis indicators
+    tumor_keywords = [
+        'primary tumor', 'metastasis', 'lymph node metastasis', 'synchronous lymph node metastasis',
+        'post rai', 'tumor'
+    ]
+    if any(k in lv for k in tumor_keywords):
+        return 1
+    return None
+
+def convert_age(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    lv = v.lower()
+    if lv in {'na', 'n/a', 'not available', 'unknown', ''}:
+        return None
+    # Extract first number as age
+    m = re.search(r'(\d+(\.\d+)?)', v)
+    if m:
+        try:
+            return float(m.group(1))
+        except:
+            return None
+    return None
+
+def convert_gender(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    lv = v.lower()
+    if lv in {'female', 'f'}:
+        return 0
+    if lv in {'male', 'm'}:
+        return 1
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (only if trait_row is available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age if age_row is not None else None,
+        gender_row=gender_row,
+        convert_gender=convert_gender if gender_row is not None else None
+    )
+    preview = preview_df(selected_clinical_df, n=5)
+    print("Preview of selected clinical features:", preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+# Based on the observed identifiers (e.g., '23064070'), these are not human gene symbols.
+# They appear to be platform-specific probe IDs or other numeric identifiers.
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Robust probe-to-gene mapping with graceful fallback
+
+import pandas as pd
+
+# 1) Determine the identifier column in the annotation that best matches the expression IDs
+expr_ids = set(gene_data.index.astype(str).str.strip())
+
+best_id_col = None
+best_overlap = 0
+for col in gene_annotation.columns:
+    try:
+        ann_values = set(gene_annotation[col].dropna().astype(str).str.strip())
+    except Exception:
+        continue
+    overlap = len(expr_ids & ann_values)
+    if overlap > best_overlap:
+        best_overlap = overlap
+        best_id_col = col
+
+# 2) Choose a gene symbol column by scanning all columns and counting extractable symbols
+gene_col = None
+best_symbol_count = -1
+for c in gene_annotation.columns:
+    try:
+        series = gene_annotation[c].dropna().astype(str)
+    except Exception:
+        continue
+    symbol_count = series.map(lambda x: len(extract_human_gene_symbols(x)) > 0).sum()
+    if symbol_count > best_symbol_count:
+        best_symbol_count = symbol_count
+        gene_col = c
+
+# Fallbacks if automatic detection fails
+if best_id_col is None:
+    best_id_col = 'ID' if 'ID' in gene_annotation.columns else gene_annotation.columns[0]
+if gene_col is None:
+    gene_col = 'GENE_SYMBOL' if 'GENE_SYMBOL' in gene_annotation.columns else gene_annotation.columns[0]
+
+print(f"Chosen ID column in annotation: {best_id_col}")
+print(f"Overlap with expression IDs: {best_overlap}")
+print(f"Chosen gene-symbol column: {gene_col} (rows with recognizable symbols: {best_symbol_count})")
+
+# 3) If no overlap with expression IDs, return an empty mapping gracefully
+if best_overlap == 0:
+    print("WARNING: No overlap between expression IDs and annotation IDs. Proceeding with empty gene-level data.")
+    gene_data = pd.DataFrame()
+else:
+    # Build mapping dataframe
+    mapping_df = get_gene_mapping(gene_annotation, prob_col=best_id_col, gene_col=gene_col)
+    # Apply mapping to convert probe-level to gene-level expression
+    mapped_gene_data = apply_gene_mapping(gene_data, mapping_df)
+
+    if mapped_gene_data.shape[0] == 0:
+        print("WARNING: Gene mapping produced an empty dataframe. Likely no usable gene symbols in the selected annotation column.")
+        gene_data = pd.DataFrame()
+    else:
+        gene_data = mapped_gene_data
+        print(f"Gene-level data shape: {gene_data.shape}")
+        print(f"First 10 mapped gene symbols: {list(gene_data.index[:10])}")
+
+# Step 7: Gene Identifier Mapping
+# Corrected Gene Identifier Mapping: prioritize true gene symbols
+
+import pandas as pd
+
+# Reload probe-level expression if gene_data was emptied previously
+if isinstance(gene_data, pd.DataFrame) and gene_data.shape[0] == 0:
+    probe_expr = get_genetic_data(matrix_file)
+else:
+    probe_expr = gene_data
+
+ann = gene_annotation.copy()
+
+# Remove obvious control probes if present
+if 'CONTROL_TYPE' in ann.columns:
+    control_mask = ann['CONTROL_TYPE'].astype(str).str.lower().isin(['pos', 'neg', 'positive', 'negative'])
+    ann = ann.loc[~control_mask]
+
+# Determine the ID column by overlap with expression probe IDs
+expr_ids = set(probe_expr.index.astype(str).str.strip())
+best_id_col = None
+best_overlap = 0
+for col in ann.columns:
+    try:
+        vals = set(ann[col].dropna().astype(str).str.strip())
+    except Exception:
+        continue
+    overlap = len(expr_ids & vals)
+    if overlap > best_overlap:
+        best_overlap = overlap
+        best_id_col = col
+
+# Force gene-symbol column to GENE_SYMBOL if available; otherwise fallback cautiously
+def count_recognizable_symbols(series: pd.Series) -> int:
+    try:
+        return series.dropna().astype(str).map(lambda x: len(extract_human_gene_symbols(x)) > 0).sum()
+    except Exception:
+        return 0
+
+preferred_symbol_cols = []
+if 'GENE_SYMBOL' in ann.columns:
+    preferred_symbol_cols.append('GENE_SYMBOL')
+# Conservative fallbacks if GENE_SYMBOL is unusable
+for alt in ['GENE_NAME', 'ACCESSION_STRING']:
+    if alt in ann.columns:
+        preferred_symbol_cols.append(alt)
+
+chosen_gene_col = None
+chosen_gene_count = -1
+
+# Try GENE_SYMBOL first; if mapping yields empty later, we'll fallback
+if 'GENE_SYMBOL' in ann.columns:
+    chosen_gene_col = 'GENE_SYMBOL'
+    chosen_gene_count = count_recognizable_symbols(ann['GENE_SYMBOL'])
+else:
+    # Choose the first fallback column with any recognizable symbols
+    for c in [c for c in preferred_symbol_cols if c != 'GENE_SYMBOL']:
+        cnt = count_recognizable_symbols(ann[c])
+        if cnt > 0:
+            chosen_gene_col = c
+            chosen_gene_count = cnt
+            break
+
+print(f"Chosen ID column in annotation: {best_id_col} (overlap with expression IDs: {best_overlap})")
+print(f"Chosen gene-symbol column: {chosen_gene_col} (rows with recognizable symbols: {chosen_gene_count})")
+
+# Build mapping and apply
+def build_and_apply_mapping(annotation_df: pd.DataFrame, id_col: str, sym_col: str, expr_df: pd.DataFrame) -> pd.DataFrame:
+    mapping_df = get_gene_mapping(annotation_df, prob_col=id_col, gene_col=sym_col)
+    if mapping_df.shape[0] == 0:
+        return pd.DataFrame()
+    # Drop obvious placeholders
+    mapping_df = mapping_df[mapping_df['Gene'].astype(str).str.strip().isin({'', '---', 'NA', 'NaN', 'nan'}) == False]
+    mapped = apply_gene_mapping(expr_df, mapping_df)
+    return mapped
+
+if (best_id_col is None) or (best_overlap == 0) or (chosen_gene_col is None):
+    print("WARNING: Unable to establish a valid probe-to-gene mapping. Producing empty gene-level data.")
+    gene_data = pd.DataFrame()
+else:
+    mapped_gene_data = build_and_apply_mapping(ann, best_id_col, chosen_gene_col, probe_expr)
+
+    # If primary choice (GENE_SYMBOL) fails, try conservative fallbacks
+    if (mapped_gene_data.shape[0] == 0) and ('GENE_SYMBOL' in ann.columns):
+        for fc in [c for c in preferred_symbol_cols if c != 'GENE_SYMBOL']:
+            mapped_fb = build_and_apply_mapping(ann, best_id_col, fc, probe_expr)
+            if mapped_fb.shape[0] > 0:
+                mapped_gene_data = mapped_fb
+                print(f"Gene-level data shape (using fallback {fc}): {mapped_gene_data.shape}")
+                break
+
+    if mapped_gene_data.shape[0] == 0:
+        print("WARNING: Gene mapping produced an empty dataframe after enforcing GENE_SYMBOL and conservative fallbacks.")
+        gene_data = pd.DataFrame()
+    else:
+        gene_data = mapped_gene_data
+        # Quick sanity check for common housekeeping genes
+        housekeeping = {'ACTB', 'GAPDH', 'RPLP0', 'RPL13A', 'HPRT1', 'B2M', 'TUBB', 'EEF1A1'}
+        present_housekeeping = [g for g in housekeeping if g in set(gene_data.index)]
+        print(f"Gene-level data shape: {gene_data.shape}")
+        print(f"First 10 mapped gene symbols: {list(gene_data.index[:10])}")
+        print(f"Housekeeping genes found (sanity check): {present_housekeeping}")

output/preprocess/Head_and_Neck_Cancer/code/GSE156915.py

@@ -0,0 +1,133 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE156915"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE156915"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE156915.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE156915.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE156915.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import re
+
+# 1) Gene expression availability (based on "Whole transcriptome" in background info)
+is_gene_available = True
+
+# 2) Variable availability based on the provided Sample Characteristics Dictionary
+#    No explicit or inferable fields for the trait "Head_and_Neck_Cancer", age, or gender were provided.
+trait_row = None
+age_row = None
+gender_row = None
+
+# 2.2) Conversion functions
+
+def _get_value_after_colon(x):
+    if x is None:
+        return None
+    s = str(x).strip()
+    if ':' in s:
+        s = s.split(':', 1)[1].strip()
+    return s if s != '' else None
+
+def convert_trait(x):
+    """
+    Binary: Head_and_Neck_Cancer case=1, non-case=0.
+    Map known H&N terms to 1; explicit 'normal'/'control' to 0; otherwise None.
+    """
+    val = _get_value_after_colon(x)
+    if val is None:
+        return None
+    v = val.lower()
+    hn_terms = [
+        'head and neck', 'hnscc', 'oral', 'oropharyngeal', 'laryngeal', 'hypopharyngeal',
+        'nasopharyngeal', 'tongue', 'buccal mucosa'
+    ]
+    if any(t in v for t in hn_terms):
+        return 1
+    if 'control' in v or 'normal' in v or 'healthy' in v or 'non-cancer' in v:
+        return 0
+    # Non-H&N cancers (e.g., colorectal) are not our target trait -> treat as non-informative here
+    return None
+
+def convert_age(x):
+    """
+    Continuous age in years. Extract numeric value.
+    """
+    val = _get_value_after_colon(x)
+    if val is None:
+        return None
+    m = re.search(r'(\d+(\.\d+)?)', val)
+    if not m:
+        return None
+    try:
+        return float(m.group(1))
+    except Exception:
+        return None
+
+def convert_gender(x):
+    """
+    Binary gender: female=0, male=1.
+    """
+    val = _get_value_after_colon(x)
+    if val is None:
+        return None
+    v = val.strip().lower()
+    if v in ['female', 'f', 'woman', 'women']:
+        return 0
+    if v in ['male', 'm', 'man', 'men']:
+        return 1
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (skip because trait_row is None)
+# If trait_row were available:
+# selected_clinical_df = geo_select_clinical_features(
+#     clinical_df=clinical_data,
+#     trait=trait,
+#     trait_row=trait_row,
+#     convert_trait=convert_trait,
+#     age_row=age_row,
+#     convert_age=convert_age,
+#     gender_row=gender_row,
+#     convert_gender=convert_gender
+# )
+# preview = preview_df(selected_clinical_df, n=5)
+# os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+# selected_clinical_df.to_csv(out_clinical_data_file)

output/preprocess/Head_and_Neck_Cancer/code/GSE184944.py

@@ -0,0 +1,124 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE184944"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE184944"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE184944.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE184944.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE184944.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import re
+import pandas as pd
+
+# 1. Gene Expression Data Availability
+# NanoString immune gene expression profiling => gene expression available
+is_gene_available = True
+
+# 2. Variable Availability and Data Type Conversion
+
+# 2.1 Data Availability based on Sample Characteristics Dictionary:
+# {0: ['histology: LL', 'histology: PL', 'histology: PEL', 'histology: EL'],
+#  1: ['smoking status: F', 'smoking status: N', 'smoking status: C'],
+#  2: ['gender: F', 'gender: M']}
+
+# Trait is Head_and_Neck_Cancer; dataset contains leukoplakia subtypes without explicit cancer status => trait not available
+trait_row = None
+
+# No age key provided
+age_row = None
+
+# Gender available at key 2
+gender_row = 2
+
+# 2.2 Data Type Conversion
+
+def convert_trait(x):
+    # Head_and_Neck_Cancer status not explicitly provided in this dataset; cannot be inferred reliably from leukoplakia subtype
+    return None
+
+def convert_age(x):
+    if x is None:
+        return None
+    try:
+        # Extract substring after colon if present
+        s = str(x)
+        if ':' in s:
+            s = s.split(':', 1)[1].strip()
+        # Extract first numeric (integer or float)
+        m = re.search(r'[-+]?\d*\.?\d+', s)
+        if not m:
+            return None
+        val = float(m.group())
+        # Return int if whole number
+        return int(val) if val.is_integer() else val
+    except Exception:
+        return None
+
+def convert_gender(x):
+    if x is None:
+        return None
+    s = str(x)
+    if ':' in s:
+        s = s.split(':', 1)[1].strip()
+    s_lower = s.strip().lower()
+    if s_lower in {'f', 'female', '0'}:
+        return 0
+    if s_lower in {'m', 'male', '1'}:
+        return 1
+    return None
+
+# 3. Save Metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4. Clinical Feature Extraction (skip if trait_row is None)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    _ = preview_df(selected_clinical_df)
+    # Save clinical data
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)

output/preprocess/Head_and_Neck_Cancer/code/GSE201777.py

@@ -0,0 +1,190 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE201777"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE201777"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE201777.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE201777.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE201777.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+
+# 1) Gene Expression Data Availability
+is_gene_available = True  # Microarray gene expression was performed on RNA samples.
+
+# 2) Variable Availability and Data Type Conversion
+
+# Use tissue to infer sample-level cancer status:
+# - Tumor -> 1
+# - Mucosa (non-tumoral) -> 0
+# - Lymph node -> None (avoid misclassification due to metastasis-negative patients' lymph nodes)
+trait_row = 1  # 'tissue' field
+age_row = None
+gender_row = None
+
+def _after_colon(x):
+    if x is None:
+        return None
+    s = str(x)
+    return s.split(":", 1)[1].strip() if ":" in s else s.strip()
+
+def convert_trait(x):
+    v = _after_colon(x)
+    if not v:
+        return None
+    v_low = v.lower()
+    if 'tumor' in v_low or 'tumour' in v_low:
+        return 1
+    if 'mucosa' in v_low:
+        return 0
+    if 'lymph' in v_low:
+        return None
+    if v_low in {'na', 'n/a', 'unknown', 'none', ''}:
+        return None
+    return None
+
+def convert_age(x):
+    v = _after_colon(x)
+    if not v:
+        return None
+    m = re.search(r'(\d+(\.\d+)?)', v)
+    if m:
+        try:
+            val = float(m.group(1))
+            return int(val) if val.is_integer() else val
+        except:
+            return None
+    return None
+
+def convert_gender(x):
+    v = _after_colon(x)
+    if not v:
+        return None
+    v_low = v.lower()
+    if v_low in {'male', 'm', 'man'}:
+        return 1
+    if v_low in {'female', 'f', 'woman'}:
+        return 0
+    return None
+
+# 3) Save Metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical Feature Extraction (only if trait is available)
+if is_trait_available:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    clinical_preview = preview_df(selected_clinical_df)
+    print(clinical_preview)
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Identify the appropriate columns in the annotation dataframe
+probe_id_col = 'ID'
+gene_symbol_col = 'Gene Symbol'
+
+# Build mapping dataframe (Probe ID -> Gene Symbol)
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_id_col, gene_col=gene_symbol_col)
+
+# Apply mapping to convert probe-level data to gene-level data
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data (fix variable name)
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Assess bias and remove biased demographic features
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort info
+note = ("INFO: Trait derived from 'tissue' field (Tumor=1, Mucosa=0); lymph node samples were set to NaN and removed "
+        "during filtering. No age/gender fields available in clinical annotations for this series.")
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=True,
+    is_trait_available=True,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data only if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Head_and_Neck_Cancer/code/GSE212250.py

@@ -0,0 +1,111 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE212250"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE212250"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE212250.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE212250.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE212250.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+# Step 1: Assess gene expression data availability based on background info (CUT&RUN epigenetic profiling)
+is_gene_available = False  # CUT&RUN histone mark profiling; not suitable gene expression data
+
+# Step 2: Variable availability and converters
+# From the sample characteristics, all samples are head and neck cancer cell lines; no age/gender provided.
+trait_row = None   # Everyone has Head_and_Neck_Cancer -> constant; treat as not available
+age_row = None     # No age field observed
+gender_row = None  # No gender field observed
+
+# Converters (robust, though not used due to unavailability)
+def _after_colon(x):
+    if x is None:
+        return None
+    s = str(x)
+    parts = s.split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    return val.strip()
+
+def convert_trait(x):
+    # Binary: 1 = cancer, 0 = non-cancer; unknown -> None
+    val = _after_colon(x)
+    if val is None or val == "":
+        return None
+    v = val.lower()
+    cancer_keywords = ["cancer", "tumor", "carcinoma", "metastatic", "metastasis", "cisplatin", "hn", "oral"]
+    control_keywords = ["normal", "healthy", "control", "adjacent", "benign", "non-cancer", "noncancer"]
+    if any(k in v for k in cancer_keywords):
+        return 1
+    if any(k in v for k in control_keywords):
+        return 0
+    return None
+
+def convert_age(x):
+    # Continuous age in years; extract numeric
+    import re
+    val = _after_colon(x)
+    if val is None or val == "":
+        return None
+    nums = re.findall(r"[0-9]+\.?[0-9]*", val)
+    if not nums:
+        return None
+    try:
+        age = float(nums[0])
+        # Basic sanity check
+        if 0 <= age <= 120:
+            return age
+        return None
+    except Exception:
+        return None
+
+def convert_gender(x):
+    # Binary: female=0, male=1
+    val = _after_colon(x)
+    if val is None or val == "":
+        return None
+    v = val.strip().lower()
+    if v in {"male", "m", "man", "men"}:
+        return 1
+    if v in {"female", "f", "woman", "women"}:
+        return 0
+    return None
+
+# Step 3: Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# Step 4: Clinical feature extraction skipped because trait_row is None

output/preprocess/Head_and_Neck_Cancer/code/GSE218109.py

@@ -0,0 +1,167 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE218109"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE218109"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE218109.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE218109.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE218109.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import re
+
+# 1) Gene expression availability
+is_gene_available = True  # Transcriptional profiling of ESCC tumors (likely mRNA expression)
+
+# 2) Variable availability (from provided Sample Characteristics Dictionary)
+# Trait (Head_and_Neck_Cancer): all samples are ESCC tumors => constant -> not available
+trait_row = None
+
+# Age and Gender are available
+age_row = 1
+gender_row = 0
+
+# 2.2) Data type conversion functions
+def convert_trait(x):
+    # Not used since trait_row is None; return None defensively
+    return None
+
+def convert_age(x):
+    if x is None:
+        return None
+    s = str(x)
+    # Extract substring after colon if present
+    if ":" in s:
+        s = s.split(":", 1)[1]
+    # Extract first number (int or float)
+    m = re.search(r'[-+]?\d*\.?\d+', s)
+    if not m:
+        return None
+    try:
+        val = float(m.group())
+        return val
+    except Exception:
+        return None
+
+def convert_gender(x):
+    if x is None:
+        return None
+    s = str(x)
+    if ":" in s:
+        s = s.split(":", 1)[1]
+    s = s.strip().lower()
+    if s in {"f", "female", "woman", "women"}:
+        return 0
+    if s in {"m", "male", "man", "men"}:
+        return 1
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (skip because trait_row is None)
+# If trait_row were available:
+# if trait_row is not None:
+#     selected_df = geo_select_clinical_features(
+#         clinical_df=clinical_data,
+#         trait=trait,
+#         trait_row=trait_row,
+#         convert_trait=convert_trait,
+#         age_row=age_row,
+#         convert_age=convert_age,
+#         gender_row=gender_row,
+#         convert_gender=convert_gender
+#     )
+#     _ = preview_df(selected_df)
+#     os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+#     selected_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Determine the appropriate columns for mapping: probe ID ('ID') and gene symbols ('GENE_SYMBOL')
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='GENE_SYMBOL')
+
+# Apply the mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save gene expression data
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# Since trait_row was determined to be None in Step 2, trait is unavailable for this cohort.
+is_gene_available = True
+is_trait_available = False
+
+# 2–4. Skip linking, missing value handling, and bias judgment due to missing trait.
+# 5. Final validation and save cohort info (record as unavailable for association due to missing trait)
+note = ("WARNING: Trait not available for this cohort. All samples are ESCC tumors "
+        "(constant tissue), so the specified trait cannot be constructed; "
+        "clinical linking skipped. Saved normalized gene data only.")
+_ = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available,
+    is_biased=False,                 # Not applicable since trait is unavailable
+    df=normalized_gene_data.T,       # Provide a valid dataframe for metadata recording
+    note=note
+)
+
+# 6. Do not save linked data since the dataset is not usable for association without trait information.

output/preprocess/Head_and_Neck_Cancer/code/GSE244580.py

@@ -0,0 +1,256 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+cohort = "GSE244580"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Head_and_Neck_Cancer"
+in_cohort_dir = "../DATA/GEO/Head_and_Neck_Cancer/GSE244580"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/GSE244580.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/GSE244580.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/GSE244580.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import pandas as pd
+
+# 1) Gene expression availability (microarray gene expression per background info)
+is_gene_available = True
+
+# 2) Variable availability
+# Sample characteristics show only disease state categories; no explicit age/gender present.
+trait_row = 0
+age_row = None
+gender_row = None
+
+# 2.2) Converters
+def _extract_value(x):
+    if x is None:
+        return None
+    try:
+        s = str(x)
+    except Exception:
+        return None
+    parts = s.split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    return val.strip()
+
+def convert_trait(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    vl = v.lower()
+    # Cancer cases: oropharyngeal cancer, peritumoral tissue, lymph node of cancer patients
+    if ("cancer" in vl) or ("peritumor" in vl) or ("peritumour" in vl):
+        return 1
+    # Non-cancer control: chronic tonsillitis
+    if ("tonsillitis" in vl):
+        return 0
+    return None
+
+def convert_age(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    # Extract first number as age
+    m = re.search(r"(\d+(\.\d+)?)", v)
+    if not m:
+        return None
+    try:
+        age = float(m.group(1))
+        # sanity check for human age
+        if 0 <= age <= 120:
+            return age
+    except Exception:
+        return None
+    return None
+
+def convert_gender(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    vl = v.strip().lower()
+    if vl in {"male", "m"}:
+        return 1
+    if vl in {"female", "f"}:
+        return 0
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical Feature Extraction
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    preview = preview_df(selected_clinical_df)
+    print(preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Ensure we are starting from the original probe-level matrix
+probe_expr_df = get_genetic_data(matrix_file)
+
+# Identify the probe ID column in annotation
+id_col = 'ID' if 'ID' in gene_annotation.columns else gene_annotation.columns[0]
+
+# Prefer true gene symbol columns (case-insensitive), else fall back to GB_ACC, else SPOT_ID
+def _norm(s: str) -> str:
+    return re.sub(r'[\s_]+', '', str(s)).lower()
+
+symbol_like_targets = {'genesymbol', 'symbol'}
+symbol_cols = [c for c in gene_annotation.columns if _norm(c) in symbol_like_targets]
+
+selected_mode = None
+gene_col = None
+if len(symbol_cols) > 0:
+    gene_col = symbol_cols[0]
+    selected_mode = 'symbol'
+elif 'GB_ACC' in gene_annotation.columns and gene_annotation['GB_ACC'].notna().sum() > 0:
+    gene_col = 'GB_ACC'
+    selected_mode = 'refseq'
+elif 'SPOT_ID' in gene_annotation.columns and gene_annotation['SPOT_ID'].notna().sum() > 0:
+    gene_col = 'SPOT_ID'
+    selected_mode = 'spot'
+else:
+    # As a last resort, pick any non-ID column
+    non_id_cols = [c for c in gene_annotation.columns if c != id_col]
+    gene_col = non_id_cols[0] if non_id_cols else id_col
+    selected_mode = 'fallback'
+
+print(f"Annotation columns available: {list(gene_annotation.columns)}")
+print(f"Identifier column selected: {id_col}")
+print(f"Selected mapping column: {gene_col} (mode={selected_mode})")
+
+# Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=id_col, gene_col=gene_col)
+
+# Apply mapping
+if selected_mode == 'symbol':
+    # Use library function that extracts/normalizes human symbols and aggregates
+    gene_data = apply_gene_mapping(expression_df=probe_expr_df, mapping_df=mapping_df)
+else:
+    # Avoid extract_human_gene_symbols to prevent dropping NR_/XR_/coordinates.
+    map_df = mapping_df[mapping_df['ID'].isin(probe_expr_df.index)].copy()
+    map_df = map_df.dropna(subset=['Gene'])
+    # Aggregate probe expressions per mapped identifier
+    merged = map_df.set_index('ID').join(probe_expr_df, how='inner')
+    expr_cols = [c for c in merged.columns if c not in ['Gene']]
+    gene_data = merged.groupby('Gene')[expr_cols].sum()
+
+# Diagnostics
+n_genes = gene_data.shape[0]
+example_genes = list(gene_data.index[:10])
+print(f"Mapped features: {n_genes}. Example identifiers: {example_genes}")
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save gene data
+initial_gene_count = gene_data.shape[0]
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+processed_linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Determine bias where possible
+if processed_linked_data.empty or trait not in processed_linked_data.columns:
+    is_trait_biased = True
+    unbiased_linked_data = processed_linked_data
+else:
+    is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(processed_linked_data, trait)
+
+# 5. Final validation and save cohort info
+post_norm_gene_count = normalized_gene_data.shape[0]
+note = (
+    f"INFO: Mapping used RefSeq accessions (GB_ACC) from platform; gene-symbol normalization reduced features "
+    f"from {initial_gene_count} to {post_norm_gene_count}. This platform lacks explicit gene symbols, "
+    f"so normalization likely dropped many features."
+)
+
+is_gene_available_final = post_norm_gene_count > 0
+is_trait_available_final = True  # trait was extracted in prior steps
+
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available_final,
+    is_trait_available=is_trait_available_final,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Head_and_Neck_Cancer/code/TCGA.py

@@ -0,0 +1,255 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Head_and_Neck_Cancer"
+
+# Input paths
+tcga_root_dir = "../DATA/TCGA"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/TCGA.csv"
+out_gene_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/gene_data/TCGA.csv"
+out_clinical_data_file = "./output/z3/preprocess/Head_and_Neck_Cancer/clinical_data/TCGA.csv"
+json_path = "./output/z3/preprocess/Head_and_Neck_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+import os
+import pandas as pd
+
+# 1) Select the most appropriate TCGA cohort directory for Head and Neck Cancer
+subdirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+subdirs_lc = {d: d.lower() for d in subdirs}
+
+# Preference keywords in order
+preference_keywords = ['head_and_neck_cancer', 'head_and_neck', '(hnsc)', 'hnsc', 'scchn']
+
+def pick_best_dir(subdirs_map, keywords):
+    # Score directories based on keyword presence and specificity
+    scored = []
+    for d, dl in subdirs_map.items():
+        score = 0
+        for i, kw in enumerate(keywords):
+            if kw in dl:
+                score += (len(keywords) - i) * 10  # earlier keywords get higher weight
+        if '(hnsc)' in dl:
+            score += 5
+        if score > 0:
+            scored.append((score, len(dl), d))
+    if not scored:
+        return None
+    # Sort by score desc, then length desc (more specific/longer name)
+    scored.sort(key=lambda x: (x[0], x[1]), reverse=True)
+    return scored[0][2]
+
+selected_subdir = pick_best_dir(subdirs_lc, preference_keywords)
+
+if selected_subdir is None:
+    print("No suitable TCGA subdirectory found for Head and Neck Cancer. Skipping this trait.")
+    # Record unusable cohort
+    _ = validate_and_save_cohort_info(
+        is_final=False,
+        cohort="TCGA_Head_and_Neck_Cancer",
+        info_path=json_path,
+        is_gene_available=False,
+        is_trait_available=False
+    )
+else:
+    print(f"Selected TCGA subdirectory: {selected_subdir}")
+    cohort_dir = os.path.join(tcga_root_dir, selected_subdir)
+
+    # 2) Identify clinical and genetic data file paths
+    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)
+
+    # 3) Load both files as DataFrames
+    clinical_df = pd.read_csv(clinical_file_path, sep='\t', index_col=0, low_memory=False)
+    genetic_df = pd.read_csv(genetic_file_path, sep='\t', index_col=0, low_memory=False)
+
+    # 4) Print clinical data column names
+    print(list(clinical_df.columns))
+
+# Step 2: Find Candidate Demographic Features
+import os
+import re
+import pandas as pd
+
+# Locate clinical file
+cohort_dir = os.path.join(tcga_root_dir, "TCGA_Head_and_Neck_Cancer_(HNSC)")
+clinical_file_path, _ = tcga_get_relevant_filepaths(cohort_dir)
+
+# Load clinical dataframe
+clinical_df = pd.read_csv(clinical_file_path, sep='\t', index_col=0, dtype=str)
+
+# Identify candidate columns
+cols_lower = {c: c.lower() for c in clinical_df.columns}
+
+def is_age_col(cl: str) -> bool:
+    # Exclude 'stage' false positives and capture standalone 'age' tokens or explicit birth proxies
+    if 'stage' in cl:
+        return False
+    if 'days_to_birth' in cl:
+        return True
+    return re.search(r'(^|[^a-z])age([^a-z]|$)', cl) is not None
+
+candidate_age_cols = [c for c, cl in cols_lower.items() if is_age_col(cl)]
+candidate_gender_cols = [
+    c for c, cl in cols_lower.items()
+    if ('gender' in cl) or (re.search(r'(^|[^a-z])sex([^a-z]|$)', cl) is not None)
+]
+
+# Print candidate lists (strict format)
+print(f"candidate_age_cols = {candidate_age_cols}")
+print(f"candidate_gender_cols = {candidate_gender_cols}")
+
+# Preview extracted data for age and gender
+age_preview = preview_df(clinical_df[candidate_age_cols]) if candidate_age_cols else {}
+gender_preview = preview_df(clinical_df[candidate_gender_cols]) if candidate_gender_cols else {}
+
+print(age_preview)
+print(gender_preview)
+
+# Step 3: Select Demographic Features
+# Select age and gender columns from candidate lists with sensible defaults and fallbacks
+try:
+    candidate_age_cols
+except NameError:
+    candidate_age_cols = []
+try:
+    candidate_gender_cols
+except NameError:
+    candidate_gender_cols = []
+
+# Preferred order for age columns based on typical TCGA clinical data
+preferred_age_cols = [
+    "age_at_initial_pathologic_diagnosis",
+    "age_at_diagnosis",
+    "age",
+    "days_to_birth",
+]
+
+# Choose age_col
+age_col = None
+for col in preferred_age_cols:
+    if col in candidate_age_cols:
+        age_col = col
+        break
+
+# Choose gender_col
+gender_col = "gender" if "gender" in candidate_gender_cols else None
+
+# Attempt to find preview dictionaries created in previous steps to print sample values
+def find_preview_values_for(col_name: str):
+    if col_name is None:
+        return None
+    for var_name, obj in globals().items():
+        if isinstance(obj, dict) and col_name in obj:
+            vals = obj.get(col_name, None)
+            if isinstance(vals, list):
+                return vals
+    return None
+
+age_preview_vals = find_preview_values_for(age_col)
+gender_preview_vals = find_preview_values_for(gender_col)
+
+# Explicitly print out the chosen columns and any available preview values
+print(f"Chosen age_col: {age_col}")
+if age_preview_vals is not None:
+    print(f"Preview values for {age_col}: {age_preview_vals}")
+
+print(f"Chosen gender_col: {gender_col}")
+if gender_preview_vals is not None:
+    print(f"Preview values for {gender_col}: {gender_preview_vals}")
+
+# Step 4: Feature Engineering and Validation
+import os
+import pandas as pd
+
+# Ensure clinical and genetic data are loaded
+try:
+    clinical_df
+    genetic_df
+except NameError:
+    cohort_dir = os.path.join(tcga_root_dir, "TCGA_Head_and_Neck_Cancer_(HNSC)")
+    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)
+    clinical_df = pd.read_csv(clinical_file_path, sep='\t', index_col=0, dtype=str, low_memory=False)
+    genetic_df = pd.read_csv(genetic_file_path, sep='\t', index_col=0, low_memory=False)
+
+# Ensure chosen demographic columns exist; if not, set None
+try:
+    age_col
+except NameError:
+    age_col = None
+try:
+    gender_col
+except NameError:
+    gender_col = None
+
+if age_col not in clinical_df.columns:
+    age_col = None
+if gender_col not in clinical_df.columns:
+    gender_col = None
+
+# 1) Extract and standardize clinical features (Trait, Age, Gender)
+selected_clinical_df = tcga_select_clinical_features(
+    clinical_df=clinical_df,
+    trait=trait,
+    age_col=age_col,
+    gender_col=gender_col
+)
+
+# 2) Normalize gene symbols; restrict to HNSC samples present in clinical data; save normalized gene data
+# Convert expression values to numeric (coerce non-numeric to NaN)
+genetic_df = genetic_df.apply(pd.to_numeric, errors='coerce')
+
+# Normalize gene symbols in index and aggregate duplicates
+normalized_gene_df = normalize_gene_symbols_in_index(genetic_df)
+
+# Match expression columns to clinical samples (HNSC cohort)
+matched_samples = sorted(list(set(normalized_gene_df.columns).intersection(set(selected_clinical_df.index))))
+normalized_gene_df_matched = normalized_gene_df.loc[:, matched_samples]
+
+# Save normalized gene data (for this cohort's samples)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_df_matched.to_csv(out_gene_data_file)
+
+# 3) Link clinical and genetic data on sample IDs
+gene_df_T = normalized_gene_df_matched.T  # samples x genes
+# Inner join on samples to ensure alignment
+linked_data = pd.concat([selected_clinical_df.loc[matched_samples], gene_df_T], axis=1, join='inner')
+
+# 4) Handle missing values systematically
+processed_df = handle_missing_values(linked_data, trait_col=trait)
+
+# 5) Determine bias in trait and demographic features; remove biased demographics
+trait_biased, final_df = judge_and_remove_biased_features(processed_df, trait)
+
+# 6) Final validation and save cohort info
+# Determine availability flags
+covariate_cols = [trait, 'Age', 'Gender']
+gene_cols_final = [c for c in final_df.columns if c not in covariate_cols]
+is_gene_available = len(gene_cols_final) > 0
+is_trait_available = (trait in final_df.columns) and (final_df[trait].notna().sum() > 0) and (len(final_df) > 0)
+
+note = ("INFO: Gene symbols normalized using NCBI synonym mapping; gene matrix restricted to HNSC samples; "
+        "Trait derived from TCGA sample barcode (01-09 tumor = 1, 10-19 normal = 0).")
+
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort="TCGA_Head_and_Neck_Cancer",
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available,
+    is_biased=trait_biased,
+    df=final_df,
+    note=note
+)
+
+# 7) Save usable linked data
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    # Save the cleaned and validated dataset
+    final_df.to_csv(out_data_file)
+
+# Ensure 'linked_data' points to the final processed dataset for downstream use
+linked_data = final_df

output/preprocess/Head_and_Neck_Cancer/cohort_info.json

@@ -1,112 +1 @@
-{
-  "GSE244580": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 30
-  },
-  "GSE218109": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": true,
-    "has_gender": true,
-    "sample_size": 36
-  },
-  "GSE212250": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": false,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE201777": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 47
-  },
-  "GSE184944": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": true,
-    "sample_size": 49
-  },
-  "GSE156915": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": false,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE151181": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": false,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE151179": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 52
-  },
-  "GSE148320": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 57
-  },
-  "GSE104006": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": true,
-    "has_gender": true,
-    "sample_size": 34
-  },
-  "TCGA": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": true,
-    "has_gender": true,
-    "sample_size": 566
-  }
-}
+{"GSE244580": {"is_usable": false, "is_gene_available": false, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": "INFO: Mapping used RefSeq accessions (GB_ACC) from platform; gene-symbol normalization reduced features from 16346 to 0. This platform lacks explicit gene symbols, so normalization likely dropped many features."}, "GSE218109": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": "WARNING: Trait not available for this cohort. All samples are ESCC tumors (constant tissue), so the specified trait cannot be constructed; clinical linking skipped. Saved normalized gene data only."}, "GSE212250": {"is_usable": false, "is_gene_available": false, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE201777": {"is_usable": true, "is_gene_available": true, "is_trait_available": true, "is_available": true, "is_biased": false, "has_age": false, "has_gender": false, "sample_size": 31, "note": "INFO: Trait derived from 'tissue' field (Tumor=1, Mucosa=0); lymph node samples were set to NaN and removed during filtering. No age/gender fields available in clinical annotations for this series."}, "GSE184944": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE156915": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE151179": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE148320": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "TCGA_Head_and_Neck_Cancer": {"is_usable": true, "is_gene_available": true, "is_trait_available": true, "is_available": true, "is_biased": false, "has_age": true, "has_gender": true, "sample_size": 566, "note": "INFO: Gene symbols normalized using NCBI synonym mapping; gene matrix restricted to HNSC samples; Trait derived from TCGA sample barcode (01-09 tumor = 1, 10-19 normal = 0)."}}

output/preprocess/Heart_rate/clinical_data/GSE18583.csv

@@ -1,42 +1,2 @@
-Sample,Heart_rate,Gender
-GSM462215,173.0,1.0
-GSM462216,155.0,1.0
-GSM462217,183.0,1.0
-GSM462218,149.0,1.0
-GSM462219,146.0,1.0
-GSM462220,157.0,1.0
-GSM462221,162.0,1.0
-GSM462222,170.0,1.0
-GSM462223,165.0,1.0
-GSM462224,144.0,1.0
-GSM462225,167.0,1.0
-GSM462226,165.0,1.0
-GSM462227,191.0,1.0
-GSM462228,149.0,1.0
-GSM462229,160.0,1.0
-GSM462230,177.0,1.0
-GSM462231,174.0,1.0
-GSM462232,165.0,1.0
-GSM462233,173.0,1.0
-GSM462234,190.0,1.0
-GSM462235,190.0,1.0
-GSM462236,190.0,1.0
-GSM462237,169.0,1.0
-GSM462238,160.0,1.0
-GSM462239,,1.0
-GSM462240,,1.0
-GSM462241,,1.0
-GSM462242,,1.0
-GSM462243,,1.0
-GSM462244,,1.0
-GSM462245,,1.0
-GSM462246,,1.0
-GSM462247,,1.0
-GSM462248,,1.0
-GSM462249,,1.0
-GSM462250,,1.0
-GSM462251,,1.0
-GSM462252,,1.0
-GSM462253,,1.0
-GSM462254,,1.0
-GSM462255,,1.0
+,GSM462215,GSM462216,GSM462217,GSM462218,GSM462219,GSM462220,GSM462221,GSM462222,GSM462223,GSM462224,GSM462225,GSM462226,GSM462227,GSM462228,GSM462229,GSM462230,GSM462231,GSM462232,GSM462233,GSM462234,GSM462235,GSM462236,GSM462237,GSM462238,GSM462239,GSM462240,GSM462241,GSM462242,GSM462243,GSM462244,GSM462245,GSM462246,GSM462247,GSM462248,GSM462249,GSM462250,GSM462251,GSM462252,GSM462253,GSM462254,GSM462255
+Heart_rate,173.0,155.0,183.0,149.0,146.0,157.0,162.0,170.0,165.0,144.0,167.0,165.0,191.0,149.0,160.0,177.0,174.0,165.0,173.0,190.0,190.0,190.0,169.0,160.0,,,,,,,,,,,,,,,,,

output/preprocess/Heart_rate/clinical_data/GSE34788.csv

@@ -1,121 +1,2 @@
-Sample,Heart_rate,Gender
-GSM855302,,
-GSM855303,,
-GSM855304,,
-GSM855305,,
-GSM855306,,
-GSM855307,,
-GSM855308,,
-GSM855309,,
-GSM855310,,
-GSM855311,,
-GSM855312,,
-GSM855313,,
-GSM855314,,
-GSM855315,,
-GSM855316,,
-GSM855317,,
-GSM855318,,
-GSM855319,,
-GSM855320,,
-GSM855321,,
-GSM855322,,
-GSM855323,,
-GSM855324,,
-GSM855325,,
-GSM855326,,
-GSM855327,,
-GSM855328,,
-GSM855329,,
-GSM855330,,
-GSM855331,,
-GSM855332,,
-GSM855333,,
-GSM855334,,
-GSM855335,,
-GSM855336,,
-GSM855337,,
-GSM855338,,
-GSM855339,,
-GSM855340,,
-GSM855341,,
-GSM855342,,
-GSM855343,,
-GSM855344,,
-GSM855345,,
-GSM855346,,
-GSM855347,,
-GSM855348,,
-GSM855349,,
-GSM855350,,
-GSM855351,,
-GSM855352,,
-GSM855353,,
-GSM855354,,
-GSM855355,,
-GSM855356,,
-GSM855357,,
-GSM855358,,
-GSM855359,,
-GSM855360,,
-GSM855361,,
-GSM855362,,
-GSM855363,,
-GSM855364,,
-GSM855365,,
-GSM855366,,
-GSM855367,,
-GSM855368,,
-GSM855369,,
-GSM855370,,
-GSM855371,,
-GSM855372,,
-GSM855373,,
-GSM855374,,
-GSM855375,,
-GSM855376,,
-GSM855377,,
-GSM855378,,
-GSM855379,,
-GSM855380,,
-GSM855381,,
-GSM855382,,
-GSM855383,,
-GSM855384,,
-GSM855385,,
-GSM855386,,
-GSM855387,,
-GSM855388,,
-GSM855389,,
-GSM855390,,
-GSM855391,,
-GSM855392,,
-GSM855393,,
-GSM855394,,
-GSM855395,,
-GSM855396,,
-GSM855397,,
-GSM855398,,
-GSM855399,,
-GSM855400,,
-GSM855401,,
-GSM855402,,
-GSM855403,,
-GSM855404,,
-GSM855405,,
-GSM855406,,
-GSM855407,,
-GSM855408,,
-GSM855409,,
-GSM855410,,
-GSM855411,,
-GSM855412,,
-GSM855413,,
-GSM855414,,
-GSM855415,,
-GSM855416,,
-GSM855417,,
-GSM855418,,
-GSM855419,,
-GSM855420,,
-GSM855421,,
+,GSM855302,GSM855303,GSM855304,GSM855305,GSM855306,GSM855307,GSM855308,GSM855309,GSM855310,GSM855311,GSM855312,GSM855313,GSM855314,GSM855315,GSM855316,GSM855317,GSM855318,GSM855319,GSM855320,GSM855321,GSM855322,GSM855323,GSM855324,GSM855325,GSM855326,GSM855327,GSM855328,GSM855329,GSM855330,GSM855331,GSM855332,GSM855333,GSM855334,GSM855335,GSM855336,GSM855337,GSM855338,GSM855339,GSM855340,GSM855341,GSM855342,GSM855343,GSM855344,GSM855345,GSM855346,GSM855347,GSM855348,GSM855349,GSM855350,GSM855351,GSM855352,GSM855353,GSM855354,GSM855355,GSM855356,GSM855357,GSM855358,GSM855359,GSM855360,GSM855361,GSM855362,GSM855363,GSM855364,GSM855365,GSM855366,GSM855367,GSM855368,GSM855369,GSM855370,GSM855371,GSM855372,GSM855373,GSM855374,GSM855375,GSM855376,GSM855377,GSM855378,GSM855379,GSM855380,GSM855381,GSM855382,GSM855383,GSM855384,GSM855385,GSM855386,GSM855387,GSM855388,GSM855389,GSM855390,GSM855391,GSM855392,GSM855393,GSM855394,GSM855395,GSM855396,GSM855397,GSM855398,GSM855399,GSM855400,GSM855401,GSM855402,GSM855403,GSM855404,GSM855405,GSM855406,GSM855407,GSM855408,GSM855409,GSM855410,GSM855411,GSM855412,GSM855413,GSM855414,GSM855415,GSM855416,GSM855417,GSM855418,GSM855419,GSM855420,GSM855421
+Heart_rate,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0

output/preprocess/Heart_rate/clinical_data/GSE35661.csv

@@ -1,42 +1,2 @@
-Sample,Heart_rate,Gender
-GSM462215,173.0,1.0
-GSM462216,155.0,1.0
-GSM462217,183.0,1.0
-GSM462218,149.0,1.0
-GSM462219,146.0,1.0
-GSM462220,157.0,1.0
-GSM462221,162.0,1.0
-GSM462222,170.0,1.0
-GSM462223,165.0,1.0
-GSM462224,144.0,1.0
-GSM462225,167.0,1.0
-GSM462226,165.0,1.0
-GSM462227,191.0,1.0
-GSM462228,149.0,1.0
-GSM462229,160.0,1.0
-GSM462230,177.0,1.0
-GSM462231,174.0,1.0
-GSM462232,165.0,1.0
-GSM462233,173.0,1.0
-GSM462234,190.0,1.0
-GSM462235,190.0,1.0
-GSM462236,190.0,1.0
-GSM462237,169.0,1.0
-GSM462238,160.0,1.0
-GSM462239,,1.0
-GSM462240,,1.0
-GSM462241,,1.0
-GSM462242,,1.0
-GSM462243,,1.0
-GSM462244,,1.0
-GSM462245,,1.0
-GSM462246,,1.0
-GSM462247,,1.0
-GSM462248,,1.0
-GSM462249,,1.0
-GSM462250,,1.0
-GSM462251,,1.0
-GSM462252,,1.0
-GSM462253,,1.0
-GSM462254,,1.0
-GSM462255,,1.0
+,GSM462215,GSM462216,GSM462217,GSM462218,GSM462219,GSM462220,GSM462221,GSM462222,GSM462223,GSM462224,GSM462225,GSM462226,GSM462227,GSM462228,GSM462229,GSM462230,GSM462231,GSM462232,GSM462233,GSM462234,GSM462235,GSM462236,GSM462237,GSM462238,GSM462239,GSM462240,GSM462241,GSM462242,GSM462243,GSM462244,GSM462245,GSM462246,GSM462247,GSM462248,GSM462249,GSM462250,GSM462251,GSM462252,GSM462253,GSM462254,GSM462255
+Heart_rate,173.0,155.0,183.0,149.0,146.0,157.0,162.0,170.0,165.0,144.0,167.0,165.0,191.0,149.0,160.0,177.0,174.0,165.0,173.0,190.0,190.0,190.0,169.0,160.0,,,,,,,,,,,,,,,,,

output/preprocess/Heart_rate/code/GSE117070.py

@@ -0,0 +1,212 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+cohort = "GSE117070"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Heart_rate"
+in_cohort_dir = "../DATA/GEO/Heart_rate/GSE117070"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/GSE117070.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/GSE117070.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/GSE117070.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+# Step 1: Determine data availability
+is_gene_available = True  # Affymetrix gene expression microarrays indicate gene expression data is available
+trait_row = None          # Heart_rate not available in provided sample characteristics
+age_row = None            # No age information available
+gender_row = None         # No gender information available
+
+# Step 2: Define conversion functions
+def _extract_after_colon(x):
+    if x is None:
+        return None
+    try:
+        s = str(x)
+    except Exception:
+        return None
+    parts = s.split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    val = val.strip().strip('"').strip()
+    return val if val != "" else None
+
+def convert_trait(x):
+    # Heart rate is continuous (e.g., bpm). Return float if present, else None.
+    val = _extract_after_colon(x)
+    if val is None:
+        return None
+    # Heuristic parsing: extract first numeric (could include decimal)
+    import re
+    m = re.search(r"(-?\d+(\.\d+)?)", val)
+    if m:
+        try:
+            return float(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def convert_age(x):
+    # Age is continuous; return years as float if present.
+    val = _extract_after_colon(x)
+    if val is None:
+        return None
+    import re
+    m = re.search(r"(\d+(\.\d+)?)", val)
+    if m:
+        try:
+            return float(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def convert_gender(x):
+    # Binary: female -> 0, male -> 1
+    val = _extract_after_colon(x)
+    if val is None:
+        return None
+    v = val.lower()
+    if v in {"male", "m"}:
+        return 1
+    if v in {"female", "f"}:
+        return 0
+    # Handle common alternatives
+    if "male" in v:
+        return 1
+    if "female" in v:
+        return 0
+    return None
+
+# Step 3: Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# Step 4: Clinical feature extraction (skip because trait_row is None)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    preview = preview_df(selected_clinical_df, n=5)
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+requires_gene_mapping = True
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Select appropriate columns based on observations:
+# - Probe identifiers in expression data match 'ID' in gene_annotation (e.g., '1007_s_at')
+# - Gene symbols are stored in 'Gene Symbol'
+id_col = 'ID'
+gene_symbol_col = 'Gene Symbol'
+
+# 2. Extract mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=id_col, gene_col=gene_symbol_col)
+
+# 3. Apply mapping to convert probe-level data to gene-level data
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1. Normalize gene symbols and save normalized gene data
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2-6. Proceed only if clinical data with the target trait exists; otherwise, record unavailability
+has_selected = ('selected_clinical_data' in globals()) and isinstance(globals()['selected_clinical_data'], pd.DataFrame)
+trait_present = has_selected and (trait in globals()['selected_clinical_data'].index)
+
+if has_selected and trait_present:
+    # 2. Link clinical and genetic data
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
+
+    # 3. Handle missing values with the specified strategy
+    linked_data = handle_missing_values(linked_data, trait)
+
+    # 4. Determine bias and remove biased demographic features
+    is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+    # 5. Final validation and save cohort info
+    is_usable = validate_and_save_cohort_info(
+        is_final=True,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=True,
+        is_trait_available=True,
+        is_biased=is_trait_biased,
+        df=unbiased_linked_data,
+        note="INFO: Clinical features extracted and linked with gene expression."
+    )
+
+    # 6. Save linked data only if usable
+    if is_usable:
+        os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+        unbiased_linked_data.to_csv(out_data_file)
+else:
+    # Trait is unavailable; perform final validation accordingly and do not save linked data
+    _ = validate_and_save_cohort_info(
+        is_final=True,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=True,
+        is_trait_available=False,
+        is_biased=False,
+        df=normalized_gene_data.T,
+        note="INFO: Trait 'Heart_rate' unavailable in clinical annotations; only 'status: pre/post-training' present. Skipping linking."
+    )

output/preprocess/Heart_rate/code/GSE12385.py

@@ -0,0 +1,213 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+cohort = "GSE12385"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Heart_rate"
+in_cohort_dir = "../DATA/GEO/Heart_rate/GSE12385"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/GSE12385.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/GSE12385.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/GSE12385.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+# Step 1: Determine data availability based on provided background and characteristics
+is_gene_available = True  # Agilent Whole Human Genome Oligo Microarrays -> mRNA expression data
+# Trait is Heart_rate; no such field in characteristics -> not available
+trait_row = None
+age_row = 1  # Age (years)
+# Gender is constant "Male" across the cohort -> not useful (constant feature)
+gender_row = None
+
+# Step 2: Define conversion functions
+import re
+from typing import Optional
+
+def _extract_value(text: str) -> str:
+    if text is None:
+        return ""
+    # Typical GEO format "Label: value"
+    parts = str(text).split(":", 1)
+    return parts[1].strip() if len(parts) > 1 else str(text).strip()
+
+def convert_trait(x) -> Optional[float]:
+    # Generic heart rate parser; kept for completeness though trait not available in this cohort
+    val = _extract_value(x)
+    if not val:
+        return None
+    # Extract first float-like number
+    m = re.search(r"[-+]?\d*\.?\d+", val)
+    if m:
+        try:
+            return float(m.group(0))
+        except:
+            return None
+    # Map common textual statuses if ever encountered (unlikely here)
+    low_map = {"resting": None, "na": None, "n/a": None, "unknown": None}
+    v = val.lower()
+    if v in low_map:
+        return low_map[v]
+    return None
+
+def convert_age(x) -> Optional[float]:
+    val = _extract_value(x)
+    if not val:
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+", val)
+    if m:
+        try:
+            return float(m.group(0))
+        except:
+            return None
+    return None
+
+def convert_gender(x) -> Optional[int]:
+    val = _extract_value(x).strip().lower()
+    if not val:
+        return None
+    # Normalize
+    if val in {"m", "male"}:
+        return 1
+    if val in {"f", "female"}:
+        return 0
+    if val in {"man"}:
+        return 1
+    if val in {"woman"}:
+        return 0
+    if val in {"unknown", "na", "n/a", "other"}:
+        return None
+    # Try to catch forms like "sex=male"
+    if "male" in val:
+        return 1
+    if "female" in val:
+        return 0
+    return None
+
+# Step 3: Initial filtering and save metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# Step 4: Clinical feature extraction (skip because trait_row is None)
+# If in other cohorts trait_row is available, uncomment and use:
+# if trait_row is not None:
+#     selected_clinical_df = geo_select_clinical_features(
+#         clinical_df=clinical_data,
+#         trait=trait,
+#         trait_row=trait_row,
+#         convert_trait=convert_trait,
+#         age_row=age_row,
+#         convert_age=convert_age,
+#         gender_row=gender_row,
+#         convert_gender=convert_gender
+#     )
+#     preview = preview_df(selected_clinical_df)
+#     os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+#     selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Decide the appropriate columns for mapping
+probe_col = 'ID'  # Matches gene_data index
+symbol_col = 'GENE_SYMBOL' if 'GENE_SYMBOL' in gene_annotation.columns else 'GENE'
+
+# Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=symbol_col)
+
+# Apply mapping to convert probe-level data to gene-level data
+gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize the obtained gene data and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2-6. Proceed only if clinical data with the trait is available; otherwise skip linking and record metadata
+has_selected = 'selected_clinical_data' in globals() and selected_clinical_data is not None
+trait_available = (trait_row is not None) and has_selected and (trait in getattr(selected_clinical_data, 'index', []))
+
+if trait_available:
+    # 2. Link the clinical and genetic data
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
+
+    # 3. Handle missing values in the linked data
+    linked_data = handle_missing_values(linked_data, trait)
+
+    # 4. Determine whether the trait and some demographic features are severely biased, and remove biased features.
+    is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+    # 5. Conduct quality check and save the cohort information.
+    is_usable = validate_and_save_cohort_info(
+        is_final=True,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=True,
+        is_trait_available=True,
+        is_biased=is_trait_biased,
+        df=unbiased_linked_data,
+        note="INFO: Linked data generated and evaluated."
+    )
+
+    # 6. If the linked data is usable, save it
+    if is_usable:
+        os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+        unbiased_linked_data.to_csv(out_data_file)
+else:
+    # Trait not available: record metadata as not usable for linking stage
+    validate_and_save_cohort_info(
+        is_final=False,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=True,
+        is_trait_available=False
+    )

output/preprocess/Heart_rate/code/GSE18583.py

@@ -0,0 +1,235 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+cohort = "GSE18583"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Heart_rate"
+in_cohort_dir = "../DATA/GEO/Heart_rate/GSE18583"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/GSE18583.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/GSE18583.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/GSE18583.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+
+# 1) Gene expression availability
+is_gene_available = True  # Series title indicates baseline skeletal muscle gene expression (mRNA microarray)
+
+# 2) Variable availability
+# Trait: Heart_rate -> 'heart rate (bpm)' at key 2
+trait_row = 2
+
+# Age: not available in the sample characteristics
+age_row = None
+
+# Gender: key 0 exists but all subjects are male (constant -> not useful)
+gender_row = None
+
+# 2.2) Converters
+_num_pattern = re.compile(r'[-+]?\d*\.?\d+')
+
+def _value_after_colon(x):
+    if x is None:
+        return None
+    parts = str(x).split(':', 1)
+    return parts[1].strip() if len(parts) == 2 else str(x).strip()
+
+def convert_trait(x):
+    v = _value_after_colon(x)
+    if v is None:
+        return None
+    m = _num_pattern.search(v)
+    if not m:
+        return None
+    try:
+        return float(m.group())
+    except Exception:
+        return None
+
+def convert_age(x):
+    v = _value_after_colon(x)
+    if v is None:
+        return None
+    m = _num_pattern.search(v)
+    if not m:
+        return None
+    try:
+        age = float(m.group())
+        if 0 < age < 120:
+            return age
+        return None
+    except Exception:
+        return None
+
+def convert_gender(x):
+    v = _value_after_colon(x)
+    if v is None:
+        return None
+    v = v.strip().lower()
+    if v in {'male', 'm', 'man', 'boy'}:
+        return 1
+    if v in {'female', 'f', 'woman', 'girl'}:
+        return 0
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (only if trait available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=None,
+        gender_row=gender_row,
+        convert_gender=None
+    )
+    preview = preview_df(selected_clinical_df)
+    print(preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+import re
+import pandas as pd
+
+# 1) Decide identifier and gene symbol columns
+# From preview, 'ID' matches expression IDs (e.g., ENST..._at) and 'SPOT_ID' contains corresponding ENST IDs without suffix.
+id_col = 'ID'
+gene_col = 'SPOT_ID'
+
+# 2) Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=id_col, gene_col=gene_col)
+
+# 3) Apply mapping to convert probe-level to gene-level (here, gene identifiers are ENST IDs; no true symbols available)
+#    Implement mapping without extracting human symbols to avoid dropping ENST IDs.
+#    Handle potential one-to-many mappings by equally splitting probe expression among mapped targets.
+mapping_df = mapping_df[mapping_df['ID'].isin(gene_data.index)].copy()
+
+# Split potential multi-target annotations; here typically single ENST per probe but keep generic handling
+def _split_genes(s: str):
+    if pd.isna(s):
+        return []
+    parts = [p.strip() for p in re.split(r'\s*[;/,|]\s*', str(s)) if p and p.strip()]
+    return parts if parts else [str(s).strip()]
+
+mapping_df['Gene_list'] = mapping_df['Gene'].apply(_split_genes)
+mapping_df['num_genes'] = mapping_df['Gene_list'].apply(len)
+
+mapping_expanded = mapping_df[['ID', 'Gene_list', 'num_genes']].explode('Gene_list').rename(columns={'Gene_list': 'Gene'})
+mapping_expanded = mapping_expanded.dropna(subset=['Gene']).set_index('ID')
+
+merged = mapping_expanded.join(gene_data, how='inner')
+expr_cols = [c for c in merged.columns if c not in ['Gene', 'num_genes']]
+
+# Distribute expression equally among mapped genes for each probe
+merged[expr_cols] = merged[expr_cols].div(merged['num_genes'].replace(0, 1), axis=0)
+
+# Sum contributions per gene
+gene_data = merged.groupby('Gene')[expr_cols].sum()
+
+# Step 7: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1) Normalize gene symbols; fall back to transcript-level data if normalization drops all rows
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data.copy())
+if normalized_gene_data.shape[0] == 0:
+    used_gene_data = gene_data.copy()
+    note = ("WARNING: Platform uses transcript identifiers (e.g., ENST); gene symbol normalization "
+            "dropped all features. Proceeding with transcript-level data without normalization.")
+else:
+    used_gene_data = normalized_gene_data
+    note = "INFO: Gene symbols normalized using NCBI synonym mapping."
+
+# Save the best available gene matrix
+used_gene_data.to_csv(out_gene_data_file)
+
+# 2) Load clinical data and link
+selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, used_gene_data)
+
+# 3) Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4) Evaluate bias and remove biased demographic features
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# Ensure booleans are Python bool for JSON serialization
+is_trait_biased = bool(is_trait_biased)
+is_gene_available_final = bool(used_gene_data.shape[0] > 0)
+is_trait_available_final = bool((trait in selected_clinical_df.index) and bool(selected_clinical_df.loc[trait].notna().any()))
+
+# 5) Final validation and save cohort info
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available_final,
+    is_trait_available=is_trait_available_final,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6) Save linked data only if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Heart_rate/code/GSE236927.py

@@ -0,0 +1,198 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+cohort = "GSE236927"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Heart_rate"
+in_cohort_dir = "../DATA/GEO/Heart_rate/GSE236927"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/GSE236927.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/GSE236927.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/GSE236927.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import re
+import pandas as pd
+
+# 1) Gene expression data availability
+is_gene_available = True  # PBMC microarray transcriptomics -> gene expression data available
+
+# 2) Variable availability and converters
+# Trait: Heart_rate
+# Not found in provided sample characteristics dictionary; background mentions HR collected but not in GEO characteristics
+trait_row = None
+
+def convert_trait(x):
+    if x is None:
+        return None
+    s = str(x).strip().lower()
+    # Try to extract after colon if present
+    if ':' in s:
+        s_val = s.split(':', 1)[1].strip()
+    else:
+        s_val = s
+    # Heuristics for heart rate (beats per minute)
+    # Known patterns might include "heart rate", "pulse", etc., but since trait_row is None, this is a fallback parser.
+    # Accept pure numbers or numbers followed by bpm
+    s_val = re.sub(r'\(.*?\)', '', s_val).strip()  # remove parentheses content
+    s_val = s_val.replace('bpm', '').strip()
+    try:
+        val = float(s_val)
+        # Basic sanity check for human resting HR range
+        if 20 <= val <= 220:
+            return val
+        return None
+    except:
+        return None
+
+# Age: appears under keys 1 and 2; key 2 contains the bulk of ages (though mixed with BMI). We'll use key 2.
+age_row = 2
+
+def convert_age(x):
+    if x is None:
+        return None
+    s = str(x).strip().lower()
+    # Extract the part after colon if exists
+    if ':' in s:
+        header, value = s.split(':', 1)
+        header = header.strip()
+        value = value.strip()
+        # Only parse if it's age-like
+        if 'age' not in header:
+            return None
+        s_val = value
+    else:
+        # No header; try to parse as a number directly (rare)
+        s_val = s
+    # Clean and parse number
+    s_val = re.sub(r'[^\d\.]+', ' ', s_val).strip()
+    # Handle cases like "79" or "79.0"
+    try:
+        val = float(s_val)
+        if 0 < val < 120:
+            return val
+        return None
+    except:
+        return None
+
+# Gender: females only (constant) per background and sample dictionary => not useful
+gender_row = None
+
+def convert_gender(x):
+    if x is None:
+        return None
+    s = str(x).strip().lower()
+    # Extract after colon if present
+    if ':' in s:
+        s_val = s.split(':', 1)[1].strip()
+    else:
+        s_val = s
+    # Normalize
+    if s_val in ['female', 'f', '0']:
+        return 0
+    if s_val in ['male', 'm', '1']:
+        return 1
+    # Sometimes formats like "sex (female/male): Female"
+    if 'female' in s_val and 'male' not in s_val:
+        return 0
+    if 'male' in s_val and 'female' not in s_val:
+        return 1
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical Feature Extraction (skip if trait not available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    _ = preview_df(selected_clinical_df)
+    # Save clinical features
+    out_dir = os.path.dirname(out_clinical_data_file)
+    os.makedirs(out_dir, exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Map probe IDs to gene symbols and aggregate to gene-level expression
+
+# Identify columns: probe IDs match 'ID' (e.g., ILMN_1651228) and gene symbols in 'Symbol'
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Symbol')
+
+# Preserve original probe-level data, then apply mapping to get gene-level data
+probe_level_data = gene_data
+gene_data = apply_gene_mapping(probe_level_data, mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# Normalize gene symbols and save normalized gene data
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# Trait not available -> skip linking and final validation; record initial metadata only
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=True,
+    is_trait_available=False
+)

output/preprocess/Heart_rate/code/GSE34788.py

@@ -0,0 +1,181 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+cohort = "GSE34788"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Heart_rate"
+in_cohort_dir = "../DATA/GEO/Heart_rate/GSE34788"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/GSE34788.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/GSE34788.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/GSE34788.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+
+# 1) Gene expression data availability
+is_gene_available = True  # mRNA microarray per background info
+
+# 2) Variable availability and converters
+
+# Sample Characteristics Dictionary indicates:
+# 6: ['heart rate: Low responders', 'heart rate: High responders']
+trait_row = 6  # heart rate responder status available and varying
+age_row = None  # no age field present
+gender_row = None  # only 'gender: female' -> constant, thus not available
+
+# Converters
+def _after_colon(value: str) -> str:
+    if value is None:
+        return ""
+    parts = str(value).split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    return val.strip()
+
+def convert_trait(x):
+    val = _after_colon(x).strip().lower()
+    if not val:
+        return None
+    # Map any "high" to 1, any "low" to 0
+    if "high" in val:
+        return 1
+    if "low" in val:
+        return 0
+    return None
+
+def convert_age(x):
+    # Not used because age_row is None, but keep robust parser
+    val = _after_colon(x)
+    # Extract first number (int/float)
+    m = re.search(r"[-+]?\d*\.?\d+", val)
+    if m:
+        try:
+            return float(m.group())
+        except Exception:
+            return None
+    return None
+
+def convert_gender(x):
+    # Not used because gender_row is None
+    val = _after_colon(x).strip().lower()
+    if val in {"female", "f", "woman", "women"}:
+        return 0
+    if val in {"male", "m", "man", "men"}:
+        return 1
+    return None
+
+# 3) Save metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4) Clinical feature extraction (only if trait available)
+if trait_row is not None:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    preview = preview_df(selected_clinical_df, n=5)
+    print(preview)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+# Determine mapping columns based on previews:
+# - Probe/feature identifiers match the 'ID' column (e.g., '7892501' style).
+# - Gene symbols are embedded in 'gene_assignment' text; extract_human_gene_symbols handles parsing.
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='gene_assignment')
+
+# Apply mapping to convert probe-level data to gene-level expression
+gene_data = apply_gene_mapping(gene_data, mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save gene data
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Bias evaluation and removal of biased covariates
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort info
+note = "INFO: Only trait available; Age/Gender not provided in clinical annotations."
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=True,
+    is_trait_available=True,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Heart_rate/code/GSE35661.py

@@ -0,0 +1,379 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+cohort = "GSE35661"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Heart_rate"
+in_cohort_dir = "../DATA/GEO/Heart_rate/GSE35661"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/GSE35661.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/GSE35661.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/GSE35661.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+import os
+import re
+import pandas as pd
+
+# 1. Gene Expression Data Availability
+is_gene_available = True  # Affymetrix U133+2 arrays indicate mRNA expression data
+
+# 2. Variable Availability and Data Type Conversion
+
+# 2.1 Identify rows for variables
+trait_row = 2       # 'heart rate (bpm)'
+age_row = None      # No age information available
+gender_row = None   # Only 'male' present (constant), thus not useful
+
+# 2.2 Converters
+def _extract_after_colon(value: str) -> str:
+    if value is None or (isinstance(value, float) and pd.isna(value)):
+        return ""
+    s = str(value)
+    return s.split(":", 1)[1].strip() if ":" in s else s.strip()
+
+def _to_float_from_text(txt: str):
+    if not txt:
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?", txt)
+    if not m:
+        return None
+    try:
+        return float(m.group(0))
+    except Exception:
+        return None
+
+def convert_trait(x):
+    # Expected format like 'heart rate (bpm): 165'
+    txt = _extract_after_colon(x)
+    return _to_float_from_text(txt)
+
+def convert_age(x):
+    # Not used (age_row is None), provided for API completeness
+    txt = _extract_after_colon(x)
+    return _to_float_from_text(txt)
+
+def convert_gender(x):
+    # Not used (gender_row is None), provided for API completeness
+    txt = _extract_after_colon(x).lower()
+    if txt in {"male", "m"}:
+        return 1
+    if txt in {"female", "f"}:
+        return 0
+    return None
+
+# 3. Save Metadata (initial filtering)
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# 4. Clinical Feature Extraction
+if is_trait_available:
+    selected_clinical_df = geo_select_clinical_features(
+        clinical_df=clinical_data,
+        trait=trait,
+        trait_row=trait_row,
+        convert_trait=convert_trait,
+        age_row=age_row,
+        convert_age=None,
+        gender_row=gender_row,
+        convert_gender=None
+    )
+    clinical_preview = preview_df(selected_clinical_df, n=5)
+
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_df.to_csv(out_clinical_data_file)
+
+# Step 3: Gene Data Extraction
+# 1. Use the get_genetic_data function from the library to get the gene_data from the matrix_file previously defined.
+gene_data = get_genetic_data(matrix_file)
+
+# 2. Print the first 20 row IDs (gene or probe identifiers) for future observation.
+print(gene_data.index[:20])
+
+# Step 4: Gene Identifier Review
+print("requires_gene_mapping = True")
+
+# Step 5: Gene Annotation
+# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.
+gene_annotation = get_gene_annotation(soft_file)
+
+# 2. Use the 'preview_df' function from the library to preview the data and print out the results.
+print("Gene annotation preview:")
+print(preview_df(gene_annotation))
+
+# Step 6: Gene Identifier Mapping
+import re
+import pandas as pd
+
+# Preserve the original probe/transcript-level expression dataframe
+expr_df = gene_data.copy()
+
+# 1) Decide identifier and gene symbol columns and build mapping dataframe
+# Identify the gene symbol column
+symbol_col = None
+for c in gene_annotation.columns:
+    cl = c.lower().strip()
+    if cl == 'gene symbol' or 'gene symbol' in cl or cl == 'gene_symbol':
+        symbol_col = c
+        break
+if symbol_col is None:
+    raise RuntimeError("Could not find a 'Gene Symbol' column in the platform annotation (SOFT).")
+
+selected_identifier_info = None
+mapping_df = None
+overlap_ids = 0
+
+# Attempt A: Use SOFT 'ID' (Affy probe IDs) vs expression IDs
+if 'ID' in gene_annotation.columns:
+    tmp_map = get_gene_mapping(gene_annotation, prob_col='ID', gene_col=symbol_col)
+    overlap_ids_A = tmp_map['ID'].isin(expr_df.index).sum()
+    if overlap_ids_A > 0:
+        mapping_df = tmp_map
+        overlap_ids = tmp_map['ID'].isin(expr_df.index).sum()
+        selected_identifier_info = "SOFT column 'ID' (Affymetrix probe IDs)"
+    # If zero, likely mismatch because expression index looks like ENSTxxx_at
+
+# Attempt B: Parse Ensembl Transcript IDs (ENST) from any annotation column, map to Gene Symbol
+if mapping_df is None:
+    # Prioritize columns explicitly mentioning Ensembl
+    candidate_cols = [c for c in gene_annotation.columns if 'ensembl' in c.lower()]
+    # If none found, scan all columns for ENST tokens
+    if not candidate_cols:
+        candidate_cols = list(gene_annotation.columns)
+
+    enst_cols = []
+    for col in candidate_cols:
+        # Check if column contains any ENST token
+        try:
+            has_enst = gene_annotation[col].astype(str).str.contains(r'ENST\d+', regex=True, na=False).any()
+        except Exception:
+            has_enst = False
+        if has_enst:
+            enst_cols.append(col)
+
+    if enst_cols:
+        rows = []
+        # Build mapping rows from ENST tokens to Gene Symbol; align to matrix IDs by appending '_at'
+        for col in enst_cols:
+            sub = gene_annotation[[col, symbol_col]].dropna(subset=[col, symbol_col])
+            # Iterate rows to extract all ENST tokens per cell
+            for val, gene_sym in zip(sub[col].astype(str), sub[symbol_col].astype(str)):
+                tokens = re.findall(r'ENST\d+', val)
+                if tokens:
+                    for t in set(tokens):
+                        rows.append((t + '_at', gene_sym))
+        if rows:
+            map_df = pd.DataFrame(rows, columns=['ID', 'Gene']).drop_duplicates()
+            # Keep only IDs present in expression data
+            map_df = map_df[map_df['ID'].isin(expr_df.index)]
+            overlap_ids_B = map_df['ID'].nunique()
+            if overlap_ids_B > 0:
+                mapping_df = map_df
+                overlap_ids = overlap_ids_B
+                selected_identifier_info = f"Parsed ENST tokens from columns: {enst_cols} (appended '_at')"
+
+# Ensure we do not proceed without a valid mapping
+if mapping_df is None or overlap_ids == 0:
+    raise RuntimeError(
+        "Failed to construct a valid mapping from platform annotation to the matrix identifiers.\n"
+        f"Matrix index example: {list(expr_df.index[:5])}\n"
+        f"Available annotation columns: {list(gene_annotation.columns)}\n"
+        "Tried: (A) 'ID' vs matrix IDs; (B) parsing ENST tokens and appending '_at'."
+    )
+
+# 2-3) Apply mapping to convert probe/transcript-level data to gene-level expression
+print(f"Selected identifier source: {selected_identifier_info}")
+print(f"Selected symbol column: {symbol_col}")
+print(f"Number of unique matrix IDs covered by mapping: {overlap_ids}")
+
+gene_data = apply_gene_mapping(expr_df, mapping_df)
+
+print(f"Gene-level matrix shape: {gene_data.shape}")
+# Optional quick preview of mapped genes
+print(f"First 5 mapped genes: {list(gene_data.index[:5])}")
+
+# Step 7: Gene Identifier Mapping
+import re
+import pandas as pd
+
+# Start from the previously extracted matrix-level expression data
+expr_df = gene_data.copy()
+
+# 1) Try to build a mapping from platform annotation: ENST*_at -> Gene Symbol
+mapping_df = None
+strategy_info = ""
+
+def find_symbol_column(df: pd.DataFrame) -> str:
+    for c in df.columns:
+        if c.strip().lower() == 'gene symbol':
+            return c
+    for c in df.columns:
+        cl = c.strip().lower()
+        if 'gene symbol' in cl or cl in ('gene_symbol', 'genesymbol'):
+            return c
+    raise RuntimeError("No 'Gene Symbol' column found.")
+
+try:
+    symbol_col = find_symbol_column(gene_annotation)
+    enst_cols = []
+    for col in gene_annotation.columns:
+        try:
+            if gene_annotation[col].astype(str).str.contains(r'ENST\d+', regex=True, na=False).any():
+                enst_cols.append(col)
+        except Exception:
+            continue
+
+    if enst_cols:
+        rows = []
+        sub = gene_annotation[enst_cols + [symbol_col]].dropna(subset=[symbol_col])
+        for _, r in sub.iterrows():
+            sym = str(r[symbol_col])
+            for col in enst_cols:
+                val = str(r[col])
+                tokens = re.findall(r'ENST\d+', val)
+                for t in set(tokens):
+                    rows.append((t + '_at', sym))
+        if rows:
+            map_df = pd.DataFrame(rows, columns=['ID', 'Gene']).drop_duplicates()
+            map_df = map_df[map_df['ID'].isin(expr_df.index)]
+            if map_df['ID'].nunique() > 0:
+                mapping_df = map_df
+                strategy_info = f"Parsed ENST tokens from annotation columns {enst_cols} mapped to '{symbol_col}'."
+except Exception:
+    pass
+
+# 2) Fallback: if no valid mapping to gene symbols, proceed with transcript-level IDs as features
+if mapping_df is None or mapping_df['ID'].nunique() == 0:
+    mapping_df = pd.DataFrame({
+        'ID': expr_df.index.astype(str),
+        'Gene': expr_df.index.astype(str)
+    })
+    strategy_info = "No valid platform mapping found; proceeding with transcript-level IDs (ENST*_at) as features."
+
+def apply_mapping_no_symbol_filter(expression_df: pd.DataFrame, mapping_df: pd.DataFrame) -> pd.DataFrame:
+    m = mapping_df.copy()
+    m = m[m['ID'].isin(expression_df.index)]
+    # Split multi-gene entries if present (e.g., 'A /// B')
+    def _split_genes(g):
+        if g is None or (isinstance(g, float) and pd.isna(g)):
+            return []
+        parts = re.split(r'\s*///\s*|;|,', str(g))
+        parts = [p.strip() for p in parts if p.strip()]
+        return parts if parts else [str(g).strip()]
+    m['Gene'] = m['Gene'].apply(_split_genes)
+    m['num_genes'] = m['Gene'].apply(len).clip(lower=1)
+    m = m.explode('Gene')
+    m = m.dropna(subset=['Gene'])
+    m = m.set_index('ID')
+    merged = m.join(expression_df)
+    expr_cols = [c for c in merged.columns if c not in ['Gene', 'num_genes']]
+    merged[expr_cols] = merged[expr_cols].div(merged['num_genes'], axis=0)
+    gene_expression_df = merged.groupby('Gene')[expr_cols].sum()
+    return gene_expression_df
+
+gene_data = apply_mapping_no_symbol_filter(expr_df, mapping_df)
+
+# Diagnostics
+covered_ids = mapping_df['ID'].isin(expr_df.index).sum()
+print(strategy_info)
+print(f"Number of unique matrix IDs covered by mapping: {covered_ids}")
+print(f"Gene-level matrix shape: {gene_data.shape}")
+print(f"First 5 mapped gene identifiers: {list(gene_data.index[:5])}")
+
+# Step 8: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1. Normalize gene symbols with fallback if normalization removes most rows
+note = "INFO: "
+try:
+    normalized_gene_data_attempt = normalize_gene_symbols_in_index(gene_data.copy())
+    original_rows = len(gene_data)
+    normalized_rows = len(normalized_gene_data_attempt)
+    retention_ratio = (normalized_rows / original_rows) if original_rows > 0 else 0.0
+except Exception as e:
+    normalized_gene_data_attempt = pd.DataFrame()
+    retention_ratio = 0.0
+
+# Fallback to transcript-level features if too few rows remain after normalization
+if (normalized_gene_data_attempt is None) or (len(normalized_gene_data_attempt) == 0) or (retention_ratio < 0.05):
+    normalized_gene_data = gene_data.copy()
+    note = ("WARNING: Gene symbol normalization skipped; platform mapping unavailable and normalization "
+            f"retained {normalized_rows if 'normalized_rows' in locals() else 0} of {original_rows if 'original_rows' in locals() else 'NA'} "
+            "rows (<5%). Proceeding with transcript-level (ENST*_at) features.")
+else:
+    normalized_gene_data = normalized_gene_data_attempt
+    note = (f"INFO: Gene symbols normalized using synonym table; retained "
+            f"{len(normalized_gene_data)} of {len(gene_data)} rows ({retention_ratio:.1%}).")
+
+# Ensure output directory exists and save gene data
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link clinical and genetic data
+# Ensure the clinical dataframe variable is available (fallback to reloading from disk if necessary)
+if 'selected_clinical_df' not in globals():
+    selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Bias assessment (only if data is non-empty and has columns)
+if len(linked_data) > 0 and trait in linked_data.columns:
+    is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+else:
+    # Create a minimal placeholder to pass to validator
+    unbiased_linked_data = linked_data
+    is_trait_biased = True
+    note = ("ERROR: Linked dataset is empty or missing the trait column after preprocessing; "
+            "cannot perform bias assessment.")
+
+# 5. Final validation and save cohort info
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=True,
+    is_trait_available=True,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Heart_rate/code/GSE72462.py

@@ -0,0 +1,94 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+cohort = "GSE72462"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Heart_rate"
+in_cohort_dir = "../DATA/GEO/Heart_rate/GSE72462"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/GSE72462.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/GSE72462.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/GSE72462.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+from tools.preprocess import *
+# 1. Identify the paths to the SOFT file and the matrix file
+soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
+
+# 2. Read the matrix file to obtain background information and sample characteristics data
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)
+
+# 3. Obtain the sample characteristics dictionary from the clinical dataframe
+sample_characteristics_dict = get_unique_values_by_row(clinical_data)
+
+# 4. Explicitly print out all the background information and the sample characteristics dictionary
+print("Background Information:")
+print(background_info)
+print("Sample Characteristics Dictionary:")
+print(sample_characteristics_dict)
+
+# Step 2: Dataset Analysis and Clinical Feature Extraction
+# Determine data availability based on provided background info and sample characteristics
+is_gene_available = True  # Gene expression microarray on muscle biopsies
+trait_row = None  # No heart rate data present in sample characteristics
+age_row = 3       # 'age: ...'
+gender_row = 2    # 'Sex: female/male'
+
+# Define conversion functions
+def _after_colon(value):
+    if value is None:
+        return None
+    # Split once by colon and take the part after it, else use the whole string
+    parts = str(value).split(":", 1)
+    v = parts[1] if len(parts) > 1 else parts[0]
+    return v.strip().strip('"').strip()
+
+def convert_trait(value):
+    # No heart rate information in this dataset
+    return None
+
+def convert_age(value):
+    v = _after_colon(value)
+    if v is None or v == '':
+        return None
+    # Extract leading number (int or float)
+    import re
+    m = re.search(r"[-+]?\d*\.?\d+", v)
+    if not m:
+        return None
+    try:
+        num = float(m.group())
+        return num
+    except:
+        return None
+
+def convert_gender(value):
+    v = _after_colon(value)
+    if v is None or v == '':
+        return None
+    v_low = v.lower()
+    if v_low in ['male', 'm']:
+        return 1
+    if v_low in ['female', 'f']:
+        return 0
+    return None
+
+# Initial filtering and save cohort metadata
+is_trait_available = trait_row is not None
+_ = validate_and_save_cohort_info(
+    is_final=False,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available
+)
+
+# Clinical feature extraction is skipped because trait data is not available (trait_row is None).

output/preprocess/Heart_rate/code/TCGA.py

@@ -0,0 +1,81 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Heart_rate"
+
+# Input paths
+tcga_root_dir = "../DATA/TCGA"
+
+# Output paths
+out_data_file = "./output/z3/preprocess/Heart_rate/TCGA.csv"
+out_gene_data_file = "./output/z3/preprocess/Heart_rate/gene_data/TCGA.csv"
+out_clinical_data_file = "./output/z3/preprocess/Heart_rate/clinical_data/TCGA.csv"
+json_path = "./output/z3/preprocess/Heart_rate/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+import os
+import re
+import pandas as pd
+
+# Given list of subdirectories from TCGA Xena dataset
+subdirs = ['TCGA_lower_grade_glioma_and_glioblastoma_(GBMLGG)', 'TCGA_Uterine_Carcinosarcoma_(UCS)', 'TCGA_Thyroid_Cancer_(THCA)', 'TCGA_Thymoma_(THYM)', 'TCGA_Testicular_Cancer_(TGCT)', 'TCGA_Stomach_Cancer_(STAD)', 'TCGA_Sarcoma_(SARC)', 'TCGA_Rectal_Cancer_(READ)', 'TCGA_Prostate_Cancer_(PRAD)', 'TCGA_Pheochromocytoma_Paraganglioma_(PCPG)', 'TCGA_Pancreatic_Cancer_(PAAD)', 'TCGA_Ovarian_Cancer_(OV)', 'TCGA_Ocular_melanomas_(UVM)', 'TCGA_Mesothelioma_(MESO)', 'TCGA_Melanoma_(SKCM)', 'TCGA_Lung_Squamous_Cell_Carcinoma_(LUSC)', 'TCGA_Lung_Cancer_(LUNG)', 'TCGA_Lung_Adenocarcinoma_(LUAD)', 'TCGA_Lower_Grade_Glioma_(LGG)', 'TCGA_Liver_Cancer_(LIHC)', 'TCGA_Large_Bcell_Lymphoma_(DLBC)', 'TCGA_Kidney_Papillary_Cell_Carcinoma_(KIRP)', 'TCGA_Kidney_Clear_Cell_Carcinoma_(KIRC)', 'TCGA_Kidney_Chromophobe_(KICH)', 'TCGA_Head_and_Neck_Cancer_(HNSC)', 'TCGA_Glioblastoma_(GBM)', 'TCGA_Esophageal_Cancer_(ESCA)', 'TCGA_Endometrioid_Cancer_(UCEC)', 'TCGA_Colon_and_Rectal_Cancer_(COADREAD)', 'TCGA_Colon_Cancer_(COAD)', 'TCGA_Cervical_Cancer_(CESC)', 'TCGA_Breast_Cancer_(BRCA)', 'TCGA_Bladder_Cancer_(BLCA)', 'TCGA_Bile_Duct_Cancer_(CHOL)', 'TCGA_Adrenocortical_Cancer_(ACC)', 'TCGA_Acute_Myeloid_Leukemia_(LAML)']
+
+# Heart-rate related terms to match as whole words to avoid false positives
+hr_terms = ['heart rate', 'heart', 'cardiac', 'pulse', 'tachycardia', 'bradycardia', 'arrhythmia', 'palpitation', 'sinus']
+# Build a regex that enforces word boundaries and is case-insensitive
+hr_pattern = re.compile(r'\b(' + '|'.join(map(re.escape, hr_terms)) + r')\b', flags=re.IGNORECASE)
+
+def contains_hr_term(name: str) -> bool:
+    # Normalize separators to spaces so word boundaries behave sensibly
+    normalized = re.sub(r'[_()/\-]+', ' ', name)
+    return bool(hr_pattern.search(normalized))
+
+# Try to find a cohort directory relevant to "Heart_rate"
+matches = [d for d in subdirs if contains_hr_term(d)]
+
+selected_dir = None
+if matches:
+    # Choose the most specific match (longest name as a naive proxy for specificity)
+    selected_dir = sorted(matches, key=lambda x: len(x), reverse=True)[0]
+
+clinical_df = pd.DataFrame()
+genetic_df = pd.DataFrame()
+
+if selected_dir is None:
+    print("No TCGA cohort directory name matches heart-rate related terms. Skipping this trait.")
+    validate_and_save_cohort_info(
+        is_final=False,
+        cohort="TCGA",
+        info_path=json_path,
+        is_gene_available=False,
+        is_trait_available=False
+    )
+else:
+    cohort_dir = os.path.join(tcga_root_dir, selected_dir)
+    clinical_path, genetic_path = tcga_get_relevant_filepaths(cohort_dir)
+
+    # Load clinical data first to perform a quick robustness check for HR-related fields
+    clinical_df = pd.read_csv(clinical_path, sep='\t', index_col=0, low_memory=False)
+
+    # Check if clinical columns contain heart-rate related fields
+    col_text = ' '.join(map(str, clinical_df.columns)).lower()
+    has_hr_field = any(term in col_text for term in ['heart rate', 'heartrate', 'pulse', 'tachycardia', 'bradycardia', 'arrhythmia', 'palpitation'])
+
+    if not has_hr_field:
+        print(f"Selected directory '{selected_dir}' does not contain heart-rate related clinical fields. Skipping this trait.")
+        # Record and skip without loading genetic data or printing columns
+        validate_and_save_cohort_info(
+            is_final=False,
+            cohort="TCGA",
+            info_path=json_path,
+            is_gene_available=False,
+            is_trait_available=False
+        )
+        clinical_df = pd.DataFrame()
+        genetic_df = pd.DataFrame()
+    else:
+        # Proceed to load genetic data and print clinical columns
+        genetic_df = pd.read_csv(genetic_path, sep='\t', index_col=0, low_memory=False)
+        print(list(clinical_df.columns))

output/preprocess/Heart_rate/cohort_info.json

@@ -1,82 +1 @@
-{
-  "GSE72462": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": false,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE35661": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 24
-  },
-  "GSE34788": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": false,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE236927": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 95
-  },
-  "GSE18583": {
-    "is_usable": true,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": false,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 24
-  },
-  "GSE12385": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": false,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE117070": {
-    "is_usable": false,
-    "is_gene_available": true,
-    "is_trait_available": false,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "TCGA": {
-    "is_usable": false,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": true,
-    "has_age": true,
-    "has_gender": true,
-    "sample_size": 79
-  }
-}
+{"GSE72462": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE35661": {"is_usable": true, "is_gene_available": true, "is_trait_available": true, "is_available": true, "is_biased": false, "has_age": false, "has_gender": false, "sample_size": 24, "note": "WARNING: Gene symbol normalization skipped; platform mapping unavailable and normalization retained 0 of 25710 rows (<5%). Proceeding with transcript-level (ENST*_at) features."}, "GSE34788": {"is_usable": true, "is_gene_available": true, "is_trait_available": true, "is_available": true, "is_biased": false, "has_age": false, "has_gender": false, "sample_size": 120, "note": "INFO: Only trait available; Age/Gender not provided in clinical annotations."}, "GSE236927": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE18583": {"is_usable": true, "is_gene_available": true, "is_trait_available": true, "is_available": true, "is_biased": false, "has_age": false, "has_gender": false, "sample_size": 24, "note": "INFO: Gene symbols normalized using NCBI synonym mapping."}, "GSE12385": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE117070": {"is_usable": false, "is_gene_available": true, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": "INFO: Trait 'Heart_rate' unavailable in clinical annotations; only 'status: pre/post-training' present. Skipping linking."}, "TCGA": {"is_usable": false, "is_gene_available": false, "is_trait_available": false, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}}

output/preprocess/Height/clinical_data/GSE106800.csv

@@ -1,48 +1,3 @@
-Sample,Height,Age,Gender
-GSM2850460,,24.0,1.0
-GSM2850461,,24.0,1.0
-GSM2850462,,24.0,1.0
-GSM2850463,,24.0,1.0
-GSM2850464,,24.0,1.0
-GSM2850465,,24.0,1.0
-GSM2850466,,24.0,1.0
-GSM2850467,,24.0,1.0
-GSM2850468,,21.0,1.0
-GSM2850469,,21.0,1.0
-GSM2850470,,21.0,1.0
-GSM2850471,,21.0,1.0
-GSM2850472,,20.0,1.0
-GSM2850473,,20.0,1.0
-GSM2850474,,20.0,1.0
-GSM2850475,,20.0,1.0
-GSM2850476,,22.0,1.0
-GSM2850477,,22.0,1.0
-GSM2850478,,22.0,1.0
-GSM2850479,,22.0,1.0
-GSM2850480,,20.0,1.0
-GSM2850481,,20.0,1.0
-GSM2850482,,20.0,1.0
-GSM2850483,,19.0,1.0
-GSM2850484,,19.0,1.0
-GSM2850485,,19.0,1.0
-GSM2850486,,19.0,1.0
-GSM2850487,,22.0,1.0
-GSM2850488,,22.0,1.0
-GSM2850489,,22.0,1.0
-GSM2850490,,22.0,1.0
-GSM2850491,,21.0,1.0
-GSM2850492,,21.0,1.0
-GSM2850493,,21.0,1.0
-GSM2850494,,21.0,1.0
-GSM2850495,,26.0,1.0
-GSM2850496,,26.0,1.0
-GSM2850497,,26.0,1.0
-GSM2850498,,26.0,1.0
-GSM2850499,,20.0,1.0
-GSM2850500,,20.0,1.0
-GSM2850501,,20.0,1.0
-GSM2850502,,20.0,1.0
-GSM2850503,,29.0,1.0
-GSM2850504,,29.0,1.0
-GSM2850505,,29.0,1.0
-GSM2850506,,29.0,1.0
+,GSM2850460,GSM2850461,GSM2850462,GSM2850463,GSM2850464,GSM2850465,GSM2850466,GSM2850467,GSM2850468,GSM2850469,GSM2850470,GSM2850471,GSM2850472,GSM2850473,GSM2850474,GSM2850475,GSM2850476,GSM2850477,GSM2850478,GSM2850479,GSM2850480,GSM2850481,GSM2850482,GSM2850483,GSM2850484,GSM2850485,GSM2850486,GSM2850487,GSM2850488,GSM2850489,GSM2850490,GSM2850491,GSM2850492,GSM2850493,GSM2850494,GSM2850495,GSM2850496,GSM2850497,GSM2850498,GSM2850499,GSM2850500,GSM2850501,GSM2850502,GSM2850503,GSM2850504,GSM2850505,GSM2850506
+Height,1.72,1.72,1.72,1.72,1.85,1.85,1.85,1.85,1.74,1.74,1.74,1.74,1.71,1.71,1.71,1.71,1.89,1.89,1.89,1.89,1.76,1.76,1.76,1.91,1.91,1.91,1.91,1.9,1.9,1.9,1.9,1.82,1.82,1.82,1.82,1.9,1.9,1.9,1.9,1.74,1.74,1.74,1.74,1.88,1.88,1.88,1.88
+Age,24.0,24.0,24.0,24.0,24.0,24.0,24.0,24.0,21.0,21.0,21.0,21.0,20.0,20.0,20.0,20.0,22.0,22.0,22.0,22.0,20.0,20.0,20.0,19.0,19.0,19.0,19.0,22.0,22.0,22.0,22.0,21.0,21.0,21.0,21.0,26.0,26.0,26.0,26.0,20.0,20.0,20.0,20.0,29.0,29.0,29.0,29.0

output/preprocess/Height/clinical_data/GSE131835.csv

@@ -0,0 +1,4 @@
+,GSM3821249,GSM3821251,GSM3821252,GSM3821254,GSM3821255,GSM3821257,GSM3821258,GSM3821259,GSM3821261,GSM3821262,GSM3821264,GSM3821265,GSM3821267,GSM3821268,GSM3821269,GSM3821271,GSM3821273,GSM3821274,GSM3821276,GSM3821277,GSM3821278,GSM3821280,GSM3821281,GSM3821283,GSM3821284,GSM3821285,GSM3821287,GSM3821288,GSM3821290,GSM3821291,GSM3821292,GSM3821293,GSM3821295,GSM3821296,GSM3821298,GSM3821299,GSM3821301,GSM3821302,GSM3821304,GSM3821305,GSM3821307,GSM3821308,GSM3821309,GSM3821311,GSM3821312,GSM3821314,GSM3821315,GSM3821316
+Height,178.0,170.0,166.0,160.0,180.0,163.0,178.0,183.0,180.0,172.0,169.0,160.0,166.0,170.0,158.0,158.0,172.0,183.0,163.0,173.0,173.0,169.0,193.0,152.0,170.0,158.0,158.0,152.0,193.0,167.0,167.0,168.0,168.0,177.0,165.0,165.0,158.0,179.0,177.0,158.0,183.0,183.0,190.0,170.0,179.0,190.0,170.0,170.0
+Age,51.0,64.0,62.0,78.0,47.0,59.0,51.0,57.0,47.0,58.0,53.0,78.0,62.0,64.0,49.0,49.0,58.0,57.0,59.0,54.0,54.0,53.0,60.0,56.0,64.0,41.0,41.0,56.0,60.0,56.0,56.0,76.0,76.0,81.0,48.0,48.0,65.0,68.0,81.0,65.0,65.0,65.0,72.0,64.0,68.0,72.0,59.0,59.0
+Gender,1.0,1.0,0.0,0.0,1.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,1.0,0.0,1.0,1.0,0.0,1.0,0.0,1.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,0.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0