Add files using upload-large-folder tool

output/preprocess/LDL_Cholesterol_Levels/code/GSE111567.py

@@ -0,0 +1,136 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "LDL_Cholesterol_Levels"
+cohort = "GSE111567"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/LDL_Cholesterol_Levels"
+in_cohort_dir = "../DATA/GEO/LDL_Cholesterol_Levels/GSE111567"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/GSE111567.csv"
+out_gene_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/gene_data/GSE111567.csv"
+out_clinical_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/clinical_data/GSE111567.csv"
+json_path = "./output/z4/preprocess/LDL_Cholesterol_Levels/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 (Illumina HumanHT-12 v4 microarray mentioned in background)
+is_gene_available = True
+
+# 2) Variable availability based on provided Sample Characteristics Dictionary
+# trait (LDL_Cholesterol_Levels): not available in the dictionary
+trait_row = None
+
+# age: not available in the dictionary
+age_row = None
+
+# gender: available at key 0
+gender_row = 0
+
+# 2.2) Conversion functions
+def _after_colon(value):
+    if value is None:
+        return None
+    s = str(value)
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    return s.strip()
+
+def convert_trait(x):
+    """
+    Convert LDL cholesterol values to a continuous variable in mg/dL if units are provided.
+    Heuristics:
+    - Extract numeric value after colon.
+    - If 'mmol' present, convert to mg/dL using factor 38.67.
+    - If units not specified, assume mg/dL.
+    """
+    s = _after_colon(x)
+    if not s:
+        return None
+    s_lower = s.lower()
+    if any(tok in s_lower for tok in ["na", "n/a", "not available", "unknown", "missing"]):
+        return None
+    # extract first float-like number
+    m = re.search(r"[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?", s_lower)
+    if not m:
+        return None
+    val = float(m.group())
+    # unit handling
+    if "mmol" in s_lower:
+        val = val * 38.67  # mmol/L to mg/dL for LDL-C
+    # basic sanity check
+    if val <= 0 or val > 1000:
+        return None
+    return val
+
+def convert_age(x):
+    """
+    Convert age to continuous years.
+    Extract numeric value; if out of plausible human range, return None.
+    """
+    s = _after_colon(x)
+    if not s:
+        return None
+    s_lower = s.lower()
+    if any(tok in s_lower for tok in ["na", "n/a", "not available", "unknown", "missing"]):
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?", s_lower)
+    if not m:
+        return None
+    age = float(m.group())
+    # assume already in years; discard implausible values
+    if age <= 0 or age > 120:
+        return None
+    return age
+
+def convert_gender(x):
+    """
+    Convert gender to binary: female=0, male=1.
+    """
+    s = _after_colon(x)
+    if not s:
+        return None
+    t = s.strip().lower()
+    # handle common variants
+    if t in ["f", "female", "woman", "girl", "frau", "fem"]:
+        return 0
+    if t in ["m", "male", "man", "boy", "herr", "masc"]:
+        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 (trait data not available in this cohort's sample characteristics)

output/preprocess/LDL_Cholesterol_Levels/code/GSE181339.py

@@ -0,0 +1,125 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "LDL_Cholesterol_Levels"
+cohort = "GSE181339"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/LDL_Cholesterol_Levels"
+in_cohort_dir = "../DATA/GEO/LDL_Cholesterol_Levels/GSE181339"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/GSE181339.csv"
+out_gene_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/gene_data/GSE181339.csv"
+out_clinical_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/clinical_data/GSE181339.csv"
+json_path = "./output/z4/preprocess/LDL_Cholesterol_Levels/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 Data Availability
+is_gene_available = True  # Agilent GE platform with PBMC transcriptomics -> gene expression data available
+
+# 2. Variable Availability and Data Type Conversion
+
+# Keys from Sample Characteristics Dictionary:
+# 0: gender
+# 1: group (OW/OB, NW, MONW)
+# 2: age
+# 3: fasting time
+# 4: timepoint
+
+# Trait (LDL_Cholesterol_Levels): not explicitly available; cannot be reliably inferred from 'group'
+trait_row = None
+
+# Age and Gender availability
+age_row = 2
+gender_row = 0
+
+def _extract_value(cell):
+    if cell is None:
+        return None
+    # Split on first colon
+    parts = str(cell).split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    return val.strip()
+
+def convert_trait(x):
+    # LDL not explicitly available in sample characteristics; return None
+    return None
+
+def convert_age(x):
+    val = _extract_value(x)
+    if val is None or val == "":
+        return None
+    # Extract first integer/float from the string
+    m = re.search(r'[-+]?\d*\.?\d+', val)
+    if not m:
+        return None
+    try:
+        num = float(m.group())
+        # Age in years; keep as float
+        if num <= 0 or num > 120:
+            # Implausible human age; mark unknown
+            return None
+        return num
+    except Exception:
+        return None
+
+def convert_gender(x):
+    val = _extract_value(x)
+    if val is None or val == "":
+        return None
+    v = val.strip().lower()
+    # Map to Female=0, Male=1
+    if v in {"woman", "female", "f"}:
+        return 0
+    if v in {"man", "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 (skip because 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 = preview_df(selected_clinical_df, n=5)
+    selected_clinical_df.to_csv(out_clinical_data_file, index=True)

output/preprocess/LDL_Cholesterol_Levels/code/GSE28893.py

@@ -0,0 +1,205 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "LDL_Cholesterol_Levels"
+cohort = "GSE28893"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/LDL_Cholesterol_Levels"
+in_cohort_dir = "../DATA/GEO/LDL_Cholesterol_Levels/GSE28893"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/GSE28893.csv"
+out_gene_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/gene_data/GSE28893.csv"
+out_clinical_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/clinical_data/GSE28893.csv"
+json_path = "./output/z4/preprocess/LDL_Cholesterol_Levels/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 data availability
+is_gene_available = True  # Illumina gene expression array in primary human liver tissue
+
+# 2) Variable availability (rows) inferred from provided sample characteristics
+trait_row = None         # LDL cholesterol not present in characteristics; background mentions related eQTLs but no trait values here
+age_row = 1              # 'age: ...'
+gender_row = 2           # 'gender: M/F'
+
+# 2.2) Data type conversion functions
+def _after_colon(value):
+    if value is None:
+        return None
+    parts = str(value).split(":", 1)
+    return parts[1].strip() if len(parts) > 1 else str(value).strip()
+
+def convert_trait(x):
+    # Trait would be continuous if present; attempt numeric parse, else None
+    v = _after_colon(x)
+    if v is None:
+        return None
+    v_clean = v.strip()
+    if v_clean.lower() in {"na", "nan", "none", "", "unknown"}:
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?", v_clean)
+    if m:
+        try:
+            return float(m.group(0))
+        except:
+            return None
+    return None
+
+def convert_age(x):
+    # Continuous numeric age in years
+    v = _after_colon(x)
+    if v is None:
+        return None
+    s = v.strip().lower()
+    if s in {"na", "nan", "none", "", "unknown"}:
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+(?:[eE][-+]?\d+)?", s)
+    if m:
+        try:
+            return float(m.group(0))
+        except:
+            return None
+    return None
+
+def convert_gender(x):
+    # Binary: female->0, male->1
+    v = _after_colon(x)
+    if v is None:
+        return None
+    s = v.strip().lower()
+    if s in {"m", "male"}:
+        return 1
+    if s in {"f", "female"}:
+        return 0
+    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 (skip because trait not 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, 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
+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
+# Map probe IDs to gene symbols using columns observed in annotation preview: 'ID' (probe) and 'Symbol' (gene symbol)
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Symbol')
+
+# Convert probe-level data to gene-level expression by applying the mapping
+gene_data = apply_gene_mapping(gene_data, mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save gene-level expression
+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)
+
+# Determine trait availability from previous steps (trait_row defined in Step 2)
+try:
+    trait_available = (trait_row is not None)
+except NameError:
+    trait_available = False
+
+# 2-6. Proceed only if trait/clinical data were extracted previously; otherwise, record metadata and skip linking
+if trait_available and 'selected_clinical_df' in globals():
+    # 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 checks 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=""
+    )
+
+    # 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 not available: record metadata accordingly; do not save linked data
+    note = "INFO: Trait not available in clinical annotations; saved gene-level data only."
+    _ = 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,  # ignored since dataset is not available for the trait
+        df=normalized_gene_data.T,  # placeholder df to avoid abnormality override
+        note=note
+    )

output/preprocess/LDL_Cholesterol_Levels/code/GSE34945.py

@@ -0,0 +1,123 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "LDL_Cholesterol_Levels"
+cohort = "GSE34945"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/LDL_Cholesterol_Levels"
+in_cohort_dir = "../DATA/GEO/LDL_Cholesterol_Levels/GSE34945"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/GSE34945.csv"
+out_gene_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/gene_data/GSE34945.csv"
+out_clinical_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/clinical_data/GSE34945.csv"
+json_path = "./output/z4/preprocess/LDL_Cholesterol_Levels/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 data availability
+# Background indicates SNP genotyping (Illumina GoldenGate), not gene expression.
+is_gene_available = False
+
+# 2) Variable availability and converters
+# From the provided sample characteristics:
+# - No LDL Cholesterol data field found -> trait not available
+# - No age field -> not available
+# - No gender field -> not available
+trait_row = None
+age_row = None
+gender_row = None
+
+# Choose data types:
+# - trait (LDL) would be continuous if available
+# - age would be continuous if available
+# - gender is binary if available
+
+def _extract_after_colon(x: str) -> str:
+    if x is None:
+        return None
+    parts = str(x).split(":", 1)
+    return parts[1].strip() if len(parts) == 2 else str(x).strip()
+
+def _to_float(s: str):
+    if s is None:
+        return None
+    s = s.strip()
+    # Remove common units and commas
+    s_clean = re.sub(r'[^\d\.\-]+', ' ', s).strip()
+    # Attempt to find a float in the string
+    m = re.search(r'[-+]?\d*\.?\d+', s_clean)
+    if not m:
+        return None
+    try:
+        return float(m.group(0))
+    except:
+        return None
+
+# Convert LDL cholesterol (continuous, mg/dL typically)
+def convert_trait(x):
+    v = _extract_after_colon(x)
+    if v is None:
+        return None
+    # Return numeric value if present; otherwise None
+    return _to_float(v)
+
+# Convert age (continuous, years)
+def convert_age(x):
+    v = _extract_after_colon(x)
+    if v is None:
+        return None
+    return _to_float(v)
+
+# Convert gender (binary: female -> 0, male -> 1)
+def convert_gender(x):
+    v = _extract_after_colon(x)
+    if v is None:
+        return None
+    v_low = v.strip().lower()
+    # Common variants
+    if v_low in {"male", "m", "man", "boy"}:
+        return 1
+    if v_low in {"female", "f", "woman", "girl"}:
+        return 0
+    # Heuristic: startswith letters
+    if v_low.startswith("m"):
+        return 1
+    if v_low.startswith("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 is skipped because trait_row is None (no clinical trait data available)

output/preprocess/LDL_Cholesterol_Levels/code/TCGA.py

@@ -0,0 +1,65 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "LDL_Cholesterol_Levels"
+
+# Input paths
+tcga_root_dir = "../DATA/TCGA"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/TCGA.csv"
+out_gene_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/gene_data/TCGA.csv"
+out_clinical_data_file = "./output/z4/preprocess/LDL_Cholesterol_Levels/clinical_data/TCGA.csv"
+json_path = "./output/z4/preprocess/LDL_Cholesterol_Levels/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+import os
+import pandas as pd
+
+# Step 1: Find the most relevant TCGA subdirectory for LDL/Cholesterol trait
+subdirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+
+# Define synonyms for LDL Cholesterol related traits
+synonyms_priority = ['ldl', 'cholesterol', 'lipid', 'hypercholesterolemia', 'dyslipidemia']
+
+def match_score(dir_name: str) -> int:
+    name = dir_name.lower()
+    for i, term in enumerate(synonyms_priority):
+        if term in name:
+            return i  # lower is better (more specific)
+    return len(synonyms_priority) + 1  # no match
+
+# Rank directories by best matching term
+ranked = sorted(subdirs, key=lambda d: match_score(d))
+best_dir = ranked[0] if ranked else None
+selected_dir = None if match_score(best_dir) > len(synonyms_priority) else best_dir
+
+clinical_df = pd.DataFrame()
+genetic_df = pd.DataFrame()
+
+if selected_dir is None:
+    print("No suitable TCGA cohort directory found for trait 'LDL_Cholesterol_Levels'. Skipping this trait.")
+    # Record unusable trait for TCGA
+    _ = validate_and_save_cohort_info(
+        is_final=False,
+        cohort="TCGA",
+        info_path=json_path,
+        is_gene_available=False,
+        is_trait_available=False
+    )
+else:
+    print(f"Selected TCGA cohort directory: {selected_dir}")
+    cohort_dir = os.path.join(tcga_root_dir, selected_dir)
+
+    # Step 2: Identify clinicalMatrix and PANCAN files
+    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)
+
+    # Step 3: Load both files
+    clinical_df = pd.read_csv(clinical_file_path, sep='\t', index_col=0, compression='infer', low_memory=False)
+    genetic_df = pd.read_csv(genetic_file_path, sep='\t', index_col=0, compression='infer', low_memory=False)
+
+    # Step 4: Print clinical data column names
+    print("Clinical data columns:")
+    print(list(clinical_df.columns))

output/preprocess/LDL_Cholesterol_Levels/cohort_info.json

@@ -1,52 +1 @@
-{
-  "GSE34945": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": true,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE28893": {
-    "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
-  },
-  "GSE181339": {
-    "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": 78
-  },
-  "GSE111567": {
-    "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
-  },
-  "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": 423
-  }
-}
+{"GSE34945": {"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}, "GSE28893": {"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 not available in clinical annotations; saved gene-level data only."}, "GSE181339": {"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}, "GSE111567": {"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": {"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/Large_B-cell_Lymphoma/code/GSE173263.py

@@ -0,0 +1,194 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Large_B-cell_Lymphoma"
+cohort = "GSE173263"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Large_B-cell_Lymphoma"
+in_cohort_dir = "../DATA/GEO/Large_B-cell_Lymphoma/GSE173263"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/GSE173263.csv"
+out_gene_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/gene_data/GSE173263.csv"
+out_clinical_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/clinical_data/GSE173263.csv"
+json_path = "./output/z4/preprocess/Large_B-cell_Lymphoma/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
+
+# Determine data availability based on sample characteristics and background info
+is_gene_available = True  # GEP indicated; this is a gene expression dataset
+
+# Trait within-cohort: response to R-CHOP (two unique values under key 2)
+trait_row = 2
+age_row = None
+gender_row = None
+
+# Conversion functions
+def _extract_value(x):
+    if x is None:
+        return None
+    parts = str(x).split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    return val.strip()
+
+def convert_trait(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    v = val.lower()
+    # Map response to R-CHOP: Early failure (refractory/early relapse) = 1; remission/responding = 0
+    failure_terms = {"early failure", "refractory", "nonresponder", "non-responder", "progressive disease", "pd", "relapse", "early relapse", "failure"}
+    response_terms = {"remission", "responder", "responding", "complete response", "cr", "partial response", "pr", "response"}
+    if v in failure_terms or any(t in v for t in ["early failure", "refractory", "early relapse", "nonrespond"]):
+        return 1
+    if v in response_terms or any(t in v for t in ["remission", "respond"]):
+        return 0
+    return None
+
+def convert_age(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+", val)
+    if not m:
+        return None
+    try:
+        return float(m.group())
+    except Exception:
+        return None
+
+def convert_gender(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    v = val.strip().lower()
+    if v in {"female", "f", "woman", "women", "girl"}:
+        return 0
+    if v in {"male", "m", "man", "men", "boy"}:
+        return 1
+    return None
+
+# 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
+)
+
+# Clinical feature extraction
+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)
+
+# Save clinical features
+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 for probe IDs and gene symbols
+probe_col = 'ID'
+gene_symbol_col = 'Gene Symbol'
+
+# Create mapping dataframe from annotation
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_symbol_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 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)
+
+# 3. Handle missing values in the linked data
+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_gene_available_final = (normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0)
+is_trait_available_final = trait in selected_clinical_df.index
+
+notes = []
+if 'Age' not in selected_clinical_df.index:
+    notes.append("Age unavailable")
+if 'Gender' not in selected_clinical_df.index:
+    notes.append("Gender unavailable")
+note = "INFO: " + ("; ".join(notes) if notes else "No additional notes")
+
+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 the 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/Large_B-cell_Lymphoma/code/GSE182362.py

@@ -0,0 +1,123 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Large_B-cell_Lymphoma"
+cohort = "GSE182362"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Large_B-cell_Lymphoma"
+in_cohort_dir = "../DATA/GEO/Large_B-cell_Lymphoma/GSE182362"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/GSE182362.csv"
+out_gene_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/gene_data/GSE182362.csv"
+out_clinical_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/clinical_data/GSE182362.csv"
+json_path = "./output/z4/preprocess/Large_B-cell_Lymphoma/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 gene expression data availability
+# Background suggests mRNA-level changes (mTOR, TLR5) post miR-155 manipulation in BJAB cells -> likely gene expression data.
+is_gene_available = True
+
+# Step 2: Variable availability and conversion functions
+# From the sample characteristics, all samples are DLBCL-derived BJAB cell line; no human-level trait variation, age, or gender.
+trait_row = None
+age_row = None
+gender_row = None
+
+def _extract_value_after_colon(x: str) -> str:
+    if x is None:
+        return ''
+    if isinstance(x, str):
+        parts = x.split(':', 1)
+        return parts[1].strip() if len(parts) == 2 else x.strip()
+    return str(x)
+
+def convert_trait(x):
+    # Map DLBCL presence to 1 and non-DLBCL to 0 when applicable; otherwise None.
+    v = _extract_value_after_colon(x).lower()
+    if not v:
+        return None
+    # Heuristics for DLBCL
+    if 'diffuse large b-cell lymphoma' in v or 'dlbcl' in v or 'large b-cell lymphoma' in v:
+        return 1
+    if 'healthy' in v or 'control' in v or 'normal' in v:
+        return 0
+    # Ignore treatment-only fields (not the trait of interest)
+    if 'transfected' in v:
+        return None
+    return None
+
+def convert_age(x):
+    v = _extract_value_after_colon(x).lower()
+    if not v:
+        return None
+    # Extract first number as age in years
+    import re
+    m = re.search(r'(\d+(\.\d+)?)', v)
+    if m:
+        try:
+            age = float(m.group(1))
+            # Filter implausible ages for humans
+            if 0 <= age <= 120:
+                return age
+        except Exception:
+            return None
+    return None
+
+def convert_gender(x):
+    v = _extract_value_after_colon(x).lower()
+    if not v:
+        return None
+    if v in ['male', 'm']:
+        return 1
+    if v in ['female', 'f']:
+        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 were 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,
+#     gender_row=gender_row,
+#     convert_gender=convert_gender
+# )
+# preview = preview_df(selected_clinical)
+# os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+# selected_clinical.to_csv(out_clinical_data_file)

output/preprocess/Large_B-cell_Lymphoma/code/GSE197977.py

@@ -0,0 +1,169 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Large_B-cell_Lymphoma"
+cohort = "GSE197977"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Large_B-cell_Lymphoma"
+in_cohort_dir = "../DATA/GEO/Large_B-cell_Lymphoma/GSE197977"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/GSE197977.csv"
+out_gene_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/gene_data/GSE197977.csv"
+out_clinical_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/clinical_data/GSE197977.csv"
+json_path = "./output/z4/preprocess/Large_B-cell_Lymphoma/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 Dictionary
+
+# 1) Gene expression availability:
+# The study analyzes immune gene expression in tumor microenvironment; this is gene expression (not miRNA/methylation).
+is_gene_available = True
+
+# 2) Variable availability and conversion functions
+
+# Trait: Large B-cell Lymphoma. All samples are LBCL patients; no controls are present.
+# No explicit disease status key is provided, and the trait is constant across samples.
+trait_row = None  # Not available for association (constant / not explicitly recorded)
+
+# Age: No age-related key present in the sample characteristics dictionary.
+age_row = None
+
+# Gender: No gender-related key present in the sample characteristics dictionary.
+gender_row = None
+
+# Conversion functions
+def _after_colon(x):
+    if x is None:
+        return None
+    s = str(x)
+    if ':' in s:
+        return s.split(':', 1)[1].strip()
+    return s.strip()
+
+def convert_trait(x):
+    # Map disease/control if encountered; otherwise None.
+    v = _after_colon(x)
+    if v is None or v == '' or v.lower() in {'na', 'nan', 'none', 'missing', 'unknown'}:
+        return None
+    val = v.lower()
+    # Positive disease indicators
+    pos_terms = {
+        'lbcl', 'large b-cell lymphoma', 'large b cell lymphoma', 'dlbcl',
+        'diffuse large b-cell lymphoma', 'diffuse large b cell lymphoma',
+        'b-cell lymphoma', 'b cell lymphoma', 'lymphoma'
+    }
+    neg_terms = {'normal', 'healthy', 'control', 'non-tumor', 'non tumour', 'non-cancer', 'non cancer', 'benign'}
+    if any(t in val for t in pos_terms):
+        return 1
+    if any(t in val for t in neg_terms):
+        return 0
+    return None
+
+def convert_age(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    v = v.strip()
+    if v.lower() in {'na', 'nan', 'none', ''}:
+        return None
+    # Extract leading numeric part
+    import re
+    m = re.search(r'(\d+(\.\d+)?)', v)
+    if m:
+        try:
+            return float(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def convert_gender(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    val = v.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
+# Skip because trait_row is None (no usable clinical trait for association).
+# 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)
+# 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 the appropriate columns for mapping based on previous previews:
+# - Expression data uses numeric 'ID's like '1', '2', ...
+# - Gene annotation preview shows matching 'ID' and gene symbols under 'ORF'
+probe_col = 'ID'
+gene_symbol_col = 'ORF'
+
+# Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_col, gene_col=gene_symbol_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/Large_B-cell_Lymphoma/code/GSE243973.py

@@ -0,0 +1,174 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Large_B-cell_Lymphoma"
+cohort = "GSE243973"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Large_B-cell_Lymphoma"
+in_cohort_dir = "../DATA/GEO/Large_B-cell_Lymphoma/GSE243973"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/GSE243973.csv"
+out_gene_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/gene_data/GSE243973.csv"
+out_clinical_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/clinical_data/GSE243973.csv"
+json_path = "./output/z4/preprocess/Large_B-cell_Lymphoma/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
+is_gene_available = True  # nCounter 780-gene panel indicates transcriptomic (gene expression) data
+
+# Identify rows for variables in the sample characteristics dictionary
+trait_row = 0     # 'disease state: large B-cell lymphoma' vs 'healthy control'
+age_row = None    # Not present in the provided characteristics
+gender_row = None # Not present in the provided characteristics
+
+# Conversion helpers
+import re
+
+def _extract_after_colon(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_after_colon(x)
+    if v is None:
+        return None
+    vl = v.lower()
+    # Primary mapping based on disease state
+    if "healthy" in vl or "control" in vl:
+        return 0
+    if "lymphoma" in vl or "lbcl" in vl:
+        return 1
+    # Fallback using cell type phrasing if encountered
+    if "donor" in vl or "healthy control" in vl:
+        return 0
+    if "patient" in vl:
+        return 1
+    return None
+
+def convert_age(x):
+    v = _extract_after_colon(x)
+    if v is None:
+        return None
+    vl = v.lower()
+    if vl in {"na", "n/a", "unknown", ""}:
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+", vl)
+    return float(m.group()) if m else None
+
+def convert_gender(x):
+    v = _extract_after_colon(x)
+    if v is None:
+        return None
+    vl = v.strip().lower()
+    if vl in {"na", "n/a", "unknown", ""}:
+        return None
+    if vl in {"male", "m", "man"}:
+        return 1
+    if vl in {"female", "f", "woman"}:
+        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 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
+    )
+    # Preview and save
+    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
+requires_gene_mapping = False
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Data Normalization and Linking
+# Ensure required modules are available
+import os
+
+# 1. Normalize the obtained gene data with the 'normalize_gene_symbols_in_index' function from the library.
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+
+# Ensure output directory exists before saving 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 with the 'geo_link_clinical_genetic_data' function from the library.
+# Fix variable name: use selected_clinical_df from Step 2
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, 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.
+note = ("INFO: Age and Gender not provided in the series characteristics; "
+        "Nanostring nCounter 780-gene panel with patient and healthy control samples.")
+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. 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)

output/preprocess/Large_B-cell_Lymphoma/code/GSE248835.py

@@ -0,0 +1,220 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Large_B-cell_Lymphoma"
+cohort = "GSE248835"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Large_B-cell_Lymphoma"
+in_cohort_dir = "../DATA/GEO/Large_B-cell_Lymphoma/GSE248835"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/GSE248835.csv"
+out_gene_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/gene_data/GSE248835.csv"
+out_clinical_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/clinical_data/GSE248835.csv"
+json_path = "./output/z4/preprocess/Large_B-cell_Lymphoma/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 pandas as pd
+import numpy as np
+
+# 1) Gene expression availability (based on series summary and context, this is gene expression data)
+is_gene_available = True
+
+# 2) Variable availability and conversion functions
+
+# Since this is a LBCL-only cohort, the disease trait is constant (not useful for association)
+trait_row = None
+
+# No explicit age or gender fields found in the provided sample characteristics dictionary
+age_row = None
+gender_row = None
+
+def _extract_value(x):
+    if x is None:
+        return None
+    if isinstance(x, str):
+        # Split by colon and take the value part
+        parts = x.split(":", 1)
+        val = parts[1] if len(parts) > 1 else parts[0]
+        val = val.strip()
+        if val in ["", "NA", "NaN", "null", "None", "Missing"]:
+            return None
+        return val
+    return x
+
+def convert_trait(x):
+    # Not used because trait_row is None; provide a generic mapper for LBCL if ever needed
+    val = _extract_value(x)
+    if val is None:
+        return None
+    v = str(val).lower()
+    # Map LBCL-related histologies to 1
+    if any(k in v for k in ["dlbcl", "large b-cell", "lbcl", "hgbcl", "hgb l", "hgb l", "hgb l", "hgb l", "hgb l", "hgb l", "hgb l"]):
+        return 1
+    return None
+
+def convert_age(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    try:
+        age = float(str(val).replace(",", "").strip())
+        # Filter out implausible ages
+        if 0 < age < 120:
+            return age
+        return None
+    except Exception:
+        return None
+
+def convert_gender(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    v = str(val).strip().lower()
+    # Female -> 0, Male -> 1
+    if v in ["female", "f", "woman", "women", "girl"]:
+        return 0
+    if v in ["male", "m", "man", "men", "boy"]:
+        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 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)
+    # 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)
+
+# 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 for mapping:
+# - Probe/ID column: 'ID' (matches numeric IDs in the expression data)
+# - Gene symbol column: 'Gene_Signature_Name' (contains gene/signature names; gene symbols can be extracted)
+
+# 2. Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Gene_Signature_Name')
+
+# 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 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)
+
+# Determine trait availability from earlier step context
+is_trait_available = (trait_row is not None)
+
+if is_trait_available:
+    # Recompute clinical features if trait is available (robust to prior steps)
+    selected_clinical_data = 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
+    )
+    # 2. Link the clinical and genetic data
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, 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)
+
+    # 5. Final validation and save 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: Linked data generated with normalized gene symbols."
+    )
+
+    # 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 not available: skip linking; still perform final validation and record metadata
+    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,  # ignored since is_available will be False
+        df=normalized_gene_data.T,  # pass a non-degenerate df to avoid abnormality override
+        note="INFO: No trait found in clinical annotations; linking and downstream steps skipped."
+    )

output/preprocess/Large_B-cell_Lymphoma/code/TCGA.py

@@ -0,0 +1,375 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Large_B-cell_Lymphoma"
+
+# Input paths
+tcga_root_dir = "../DATA/TCGA"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/TCGA.csv"
+out_gene_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/gene_data/TCGA.csv"
+out_clinical_data_file = "./output/z4/preprocess/Large_B-cell_Lymphoma/clinical_data/TCGA.csv"
+json_path = "./output/z4/preprocess/Large_B-cell_Lymphoma/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+# Select the most relevant TCGA cohort directory for the trait
+dirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+san = lambda s: s.lower().replace('-', '').replace(' ', '').replace('_', '')
+trait_key = san(trait)
+
+candidates = []
+for d in dirs:
+    dl = d.lower()
+    sdl = san(d)
+    contains_exact_trait = trait_key in sdl
+    contains_dlbc = 'dlbc' in dl
+    contains_large = 'large' in dl
+    contains_bcell = ('bcell' in sdl) or ('b-cell' in dl)
+    contains_lymphoma = 'lymphoma' in dl
+    score = (
+        int(contains_exact_trait),
+        int(contains_dlbc),
+        int(contains_large and contains_bcell and contains_lymphoma),
+        int(contains_lymphoma)
+    )
+    if any(score):
+        candidates.append((score, d))
+
+selected_dir = None
+if candidates:
+    # Sort by score descending (most specific first)
+    candidates.sort(key=lambda x: x[0], reverse=True)
+    selected_dir = candidates[0][1]
+
+if selected_dir is None:
+    # No suitable cohort found; mark as unavailable for this trait
+    validate_and_save_cohort_info(
+        is_final=False,
+        cohort="TCGA_unmatched",
+        info_path=json_path,
+        is_gene_available=False,
+        is_trait_available=False
+    )
+else:
+    cohort_dir = os.path.join(tcga_root_dir, selected_dir)
+    # Identify clinical and genetic file paths
+    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)
+
+    # Robust TSV reader with gzip support
+    def read_tsv_auto(fp):
+        compression = 'gzip' if fp.endswith('.gz') else None
+        return pd.read_csv(fp, sep='\t', index_col=0, low_memory=False, compression=compression)
+
+    clinical_df = read_tsv_auto(clinical_file_path)
+    genetic_df = read_tsv_auto(genetic_file_path)
+
+    # Print clinical column names for inspection
+    print(clinical_df.columns.tolist())
+
+# Step 2: Find Candidate Demographic Features
+# Use the provided column name list directly (no file I/O)
+columns_list = ['_INTEGRATION', '_PATIENT', '_cohort', '_primary_disease', '_primary_site', 'age_at_initial_pathologic_diagnosis', 'b_lymphocyte_genotyping_method', 'bcr_followup_barcode', 'bcr_patient_barcode', 'bcr_sample_barcode', 'bone_marrow_biopsy_done', 'bone_marrow_involvement', 'bone_marrow_sample_histology', 'clinical_stage', 'days_to_birth', 'days_to_collection', 'days_to_death', 'days_to_initial_pathologic_diagnosis', 'days_to_last_followup', 'days_to_new_tumor_event_after_initial_treatment', 'eastern_cancer_oncology_group', 'ebv_positive_malignant_cells_percent', 'ebv_status_malignant_cells_method', 'epstein_barr_viral_status', 'extranodal_involvement', 'extranodal_involvment_site_other', 'extranodal_sites_involvement_number', 'first_progression_histology_type', 'first_progression_histology_type_other', 'first_recurrence_biopsy_confirmed', 'follicular_percent', 'followup_case_report_form_submission_reason', 'followup_treatment_success', 'form_completion_date', 'gender', 'genetic_abnormality_method_other', 'genetic_abnormality_results', 'genetic_abnormality_results_other', 'genetic_abnormality_tested', 'genetic_abnormality_tested_other', 'height', 'histological_type', 'history_immunological_disease', 'history_immunological_disease_other', 'history_immunosuppresive_rx', 'history_immunosuppressive_rx_other', 'history_of_neoadjuvant_treatment', 'history_relevant_infectious_dx', 'hiv_status', 'icd_10', 'icd_o_3_histology', 'icd_o_3_site', 'igh_genotype_results', 'immunophenotypic_analysis_method', 'immunophenotypic_analysis_results', 'immunophenotypic_analysis_tested', 'informed_consent_verified', 'initial_weight', 'intermediate_dimension', 'is_ffpe', 'ldh_level', 'ldh_norm_range_upper', 'longest_dimension', 'lost_follow_up', 'lymph_node_involvement_site', 'maximum_tumor_bulk_anatomic_site', 'maximum_tumor_dimension', 'mib1_positive_percentage_range', 'new_neoplasm_event_occurrence_anatomic_site', 'new_neoplasm_occurrence_anatomic_site_text', 'new_tumor_event_after_initial_treatment', 'oct_embedded', 'other_dx', 'pathology_report_file_name', 'patient_id', 'person_neoplasm_cancer_status', 'pet_scan_results', 'primary_therapy_outcome_success', 'radiation_therapy', 'sample_type', 'sample_type_id', 'shortest_dimension', 'system_version', 'targeted_molecular_therapy', 'tissue_prospective_collection_indicator', 'tissue_retrospective_collection_indicator', 'tissue_source_site', 'tumor_tissue_site', 'vial_number', 'vital_status', 'weight', 'year_of_initial_pathologic_diagnosis', '_GENOMIC_ID_TCGA_DLBC_exp_HiSeqV2', '_GENOMIC_ID_TCGA_DLBC_exp_HiSeqV2_exon', '_GENOMIC_ID_TCGA_DLBC_PDMRNAseq', '_GENOMIC_ID_TCGA_DLBC_hMethyl450', '_GENOMIC_ID_TCGA_DLBC_miRNA_HiSeq', '_GENOMIC_ID_TCGA_DLBC_gistic2thd', '_GENOMIC_ID_TCGA_DLBC_PDMRNAseqCNV', '_GENOMIC_ID_data/public/TCGA/DLBC/miRNA_HiSeq_gene', '_GENOMIC_ID_TCGA_DLBC_gistic2', '_GENOMIC_ID_TCGA_DLBC_mutation_bcm_gene', '_GENOMIC_ID_TCGA_DLBC_exp_HiSeqV2_percentile', '_GENOMIC_ID_TCGA_DLBC_RPPA', '_GENOMIC_ID_TCGA_DLBC_exp_HiSeqV2_PANCAN']
+
+# Identify candidate demographic columns from the list
+candidate_age_cols = []
+for col in columns_list:
+    low = col.lower()
+    if (
+        low == "days_to_birth"
+        or low.startswith("age")
+        or "age_at_" in low
+        or "_age_" in low
+    ):
+        # Avoid false positives such as 'stage' and 'percentage'
+        if "stage" in low or "percentage" in low:
+            continue
+        candidate_age_cols.append(col)
+
+candidate_gender_cols = []
+for col in columns_list:
+    low = col.lower()
+    if low in ("gender", "sex"):
+        candidate_gender_cols.append(col)
+
+# Print required lists in strict format
+print(f"candidate_age_cols = {candidate_age_cols}")
+print(f"candidate_gender_cols = {candidate_gender_cols}")
+
+# Preview extracted data (if clinical_df is available in memory)
+if 'clinical_df' in globals() and isinstance(globals().get('clinical_df'), pd.DataFrame):
+    available_age_cols = [c for c in candidate_age_cols if c in clinical_df.columns]
+    if available_age_cols:
+        age_preview = preview_df(clinical_df[available_age_cols], n=5)
+        print(age_preview)
+    available_gender_cols = [c for c in candidate_gender_cols if c in clinical_df.columns]
+    if available_gender_cols:
+        gender_preview = preview_df(clinical_df[available_gender_cols], n=5)
+        print(gender_preview)
+
+# Step 3: Select Demographic Features
+import pandas as pd
+
+# Safely retrieve candidate lists
+age_candidates = globals().get("candidate_age_cols", []) or []
+gender_candidates = globals().get("candidate_gender_cols", []) or []
+
+def _get_preview_dict(candidates: list) -> dict:
+    # Try to find a dict in globals that contains keys overlapping with candidates and list-like values
+    best_match = None
+    best_overlap = 0
+    for name, obj in globals().items():
+        if isinstance(obj, dict) and any(k in candidates for k in obj.keys()):
+            # Ensure values are list-like for overlapping keys
+            overlap_keys = [k for k in obj.keys() if k in candidates and isinstance(obj[k], list)]
+            overlap = len(overlap_keys)
+            if overlap > best_overlap:
+                best_overlap = overlap
+                best_match = obj
+    return best_match or {}
+
+def _missing_ratio(vals: list) -> float:
+    if not isinstance(vals, list) or len(vals) == 0:
+        return 1.0
+    def _is_missing(v):
+        if v is None:
+            return True
+        if isinstance(v, float) and pd.isna(v):
+            return True
+        if isinstance(v, str) and v.strip().lower() in {"", "na", "n/a", "null", "none"}:
+            return True
+        return False
+    missing = sum(_is_missing(v) for v in vals)
+    return missing / len(vals)
+
+def _numeric_ratio(vals: list) -> float:
+    if not isinstance(vals, list) or len(vals) == 0:
+        return 0.0
+    def _is_number(v):
+        try:
+            float(str(v).replace(",", "").strip())
+            return True
+        except Exception:
+            return False
+    numeric = sum(_is_number(v) for v in vals)
+    return numeric / len(vals)
+
+def _gender_valid_ratio(vals: list) -> float:
+    if not isinstance(vals, list) or len(vals) == 0:
+        return 0.0
+    valid_tokens = {"male", "female", "m", "f"}
+    def _is_valid(v):
+        if v is None:
+            return False
+        s = str(v).strip().lower()
+        return s in valid_tokens
+    valid = sum(_is_valid(v) for v in vals)
+    return valid / len(vals)
+
+# Retrieve preview dictionaries if available
+age_preview_dict = _get_preview_dict(age_candidates)
+gender_preview_dict = _get_preview_dict(gender_candidates)
+
+age_col = None
+gender_col = None
+
+# Select age column
+if len(age_candidates) > 0:
+    # Prefer columns that represent age directly
+    preferred_age_order = [
+        "age_at_initial_pathologic_diagnosis",
+        "age_at_diagnosis",
+    ]
+    # Build a scored list of candidates based on missingness and numeric content
+    scored = []
+    for col in age_candidates:
+        vals = age_preview_dict.get(col, [])
+        miss = _missing_ratio(vals)
+        numr = _numeric_ratio(vals)
+        # Score: higher is better; penalize missingness. Prefer direct age columns by bonus.
+        bonus = 0.2 if col in preferred_age_order else 0.0
+        score = (numr - miss) + bonus
+        scored.append((score, -miss, numr, col))
+    if scored:
+        # Sort by highest score, then lowest missingness
+        scored.sort(reverse=True)
+        best_col = scored[0][3]
+        # Apply a sanity threshold on missingness (use preview to avoid heavily missing)
+        best_vals = age_preview_dict.get(best_col, [])
+        if _missing_ratio(best_vals) < 0.6:
+            age_col = best_col
+        else:
+            age_col = None
+else:
+    age_col = None
+
+# Select gender column
+if len(gender_candidates) > 0:
+    # Prefer straightforward 'gender' column
+    preferred_gender_order = ["gender", "sex"]
+    scored = []
+    for col in gender_candidates:
+        vals = gender_preview_dict.get(col, [])
+        miss = _missing_ratio(vals)
+        valid = _gender_valid_ratio(vals)
+        bonus = 0.2 if col in preferred_gender_order else 0.0
+        score = (valid - miss) + bonus
+        scored.append((score, -miss, valid, col))
+    if scored:
+        scored.sort(reverse=True)
+        best_col = scored[0][3]
+        best_vals = gender_preview_dict.get(best_col, [])
+        if _missing_ratio(best_vals) < 0.6:
+            gender_col = best_col
+        else:
+            gender_col = None
+else:
+    gender_col = None
+
+# Explicitly print chosen columns and their preview values (first 5), if available
+print(f"Chosen age_col: {age_col}")
+if age_col is not None:
+    print(f"Preview values for {age_col}: {age_preview_dict.get(age_col, [])[:5] if isinstance(age_preview_dict.get(age_col, []), list) else 'N/A'}")
+else:
+    print("Preview values for age_col: None (no suitable age column found)")
+
+print(f"Chosen gender_col: {gender_col}")
+if gender_col is not None:
+    print(f"Preview values for {gender_col}: {gender_preview_dict.get(gender_col, [])[:5] if isinstance(gender_preview_dict.get(gender_col, []), list) else 'N/A'}")
+else:
+    print("Preview values for gender_col: None (no suitable gender column found)")
+
+# Step 4: Feature Engineering and Validation
+import os
+import re
+import pandas as pd
+
+# 1) Extract and standardize clinical features
+def _to_int_or_none(x):
+    try:
+        return int(re.search(r'\d+', str(x)).group())
+    except Exception:
+        return None
+
+def _extract_sample_type_from_barcode(barcode: str) -> int:
+    # TCGA barcode pattern: TCGA-XX-XXXX-SS... where SS are two digits (01-09 tumor, 10-19 normal)
+    try:
+        parts = str(barcode).split('-')
+        if len(parts) >= 4:
+            code2 = parts[3][:2]
+            return int(code2) if code2.isdigit() else None
+    except Exception:
+        pass
+    # Fallback: search any two-digit token
+    m = re.search(r'-(\d{2})', str(barcode))
+    if m:
+        try:
+            return int(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def _trait_from_sample_type_code(code: int) -> int:
+    if code is None:
+        return None
+    if 1 <= code <= 9:
+        return 1
+    if 10 <= code <= 19:
+        return 0
+    return None
+
+# Determine trait from sample_type_id if available, else parse barcode
+if 'sample_type_id' in clinical_df.columns:
+    st_codes = clinical_df['sample_type_id'].apply(_to_int_or_none)
+else:
+    st_codes = clinical_df.index.to_series().apply(_extract_sample_type_from_barcode)
+
+trait_series = st_codes.apply(_trait_from_sample_type_code).rename(trait)
+
+age_series = clinical_df[age_col].apply(tcga_convert_age).rename("Age") if (isinstance(age_col, str) and age_col in clinical_df.columns) else None
+gender_series = clinical_df[gender_col].apply(tcga_convert_gender).rename("Gender") if (isinstance(gender_col, str) and gender_col in clinical_df.columns) else None
+
+feature_list = [trait_series]
+if age_series is not None:
+    feature_list.append(age_series)
+if gender_series is not None:
+    feature_list.append(gender_series)
+clinical_selected_df = pd.concat(feature_list, axis=1)
+
+# 2) Normalize gene symbols in expression data and save
+# Detect orientation and ensure we have genes as rows, samples as columns for normalization
+barcode_pat = re.compile(r"^TCGA-[A-Z0-9]{2}-[A-Z0-9]{4}-\d{2}")
+
+def _ratio_barcode_like(vals):
+    if len(vals) == 0:
+        return 0.0
+    cnt = 0
+    tot = 0
+    for v in vals:
+        s = str(v)
+        if s and s != 'nan':
+            tot += 1
+            if bool(barcode_pat.match(s)):
+                cnt += 1
+    return (cnt / tot) if tot > 0 else 0.0
+
+cols_match_ratio = _ratio_barcode_like(genetic_df.columns)
+idx_match_ratio = _ratio_barcode_like(genetic_df.index)
+
+# If many columns look like barcodes -> genes x samples; else if index looks like barcodes -> samples x genes -> transpose
+if cols_match_ratio >= idx_match_ratio:
+    gene_by_sample = genetic_df.copy()
+else:
+    gene_by_sample = genetic_df.T.copy()
+
+# Keep only valid TCGA sample columns
+valid_sample_cols = [c for c in gene_by_sample.columns if isinstance(c, str) and barcode_pat.match(c)]
+gene_by_sample = gene_by_sample[valid_sample_cols]
+
+# Normalize gene symbols (index assumed to be gene symbols)
+gene_by_sample_norm = normalize_gene_symbols_in_index(gene_by_sample)
+
+# Save normalized gene matrix (genes x samples)
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+gene_by_sample_norm.to_csv(out_gene_data_file)
+
+# 3) Link clinical and genetic data on sample IDs
+samples_common = clinical_selected_df.index.intersection(gene_by_sample_norm.columns)
+clinical_selected_df_aligned = clinical_selected_df.loc[samples_common]
+expr_samples_by_genes = gene_by_sample_norm.T.loc[samples_common]  # samples x genes
+
+linked_data_raw = pd.concat([clinical_selected_df_aligned, expr_samples_by_genes], axis=1)
+linked_data = linked_data_raw.copy()
+
+# 4) Handle missing values systematically
+linked_data = handle_missing_values(linked_data, trait_col=trait)
+
+# 5) Determine bias and remove biased demographic features
+is_biased, linked_data = judge_and_remove_biased_features(linked_data, trait=trait)
+
+# 6) Final quality validation and save cohort info
+cohort_name = globals().get('selected_dir', 'TCGA_Large_Bcell_Lymphoma_(DLBC)')
+is_gene_available = gene_by_sample_norm.shape[0] > 0 and gene_by_sample_norm.shape[1] > 0
+is_trait_available = trait in clinical_selected_df.columns and clinical_selected_df[trait].notna().sum() > 0
+
+note = ("INFO: Trait derived from sample_type_id when available; otherwise parsed from TCGA barcode. "
+        "Gene matrix orientation auto-detected and normalized using NCBI synonym mapping; "
+        "kept only valid TCGA barcode columns.")
+
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort_name,
+    info_path=json_path,
+    is_gene_available=bool(is_gene_available),
+    is_trait_available=bool(is_trait_available),
+    is_biased=is_biased,
+    df=linked_data,
+    note=note
+)
+
+# 7) Save the usable linked dataset
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    linked_data.to_csv(out_data_file)

output/preprocess/Liver_Cancer/clinical_data/GSE174570.csv

@@ -1,2 +1,2 @@
-0,1
-1.0,
+,GSM5319834,GSM5319835,GSM5319836,GSM5319837,GSM5319838,GSM5319839,GSM5319840,GSM5319841,GSM5319842,GSM5319843,GSM5319844,GSM5319845,GSM5319846,GSM5319847,GSM5319848,GSM5319849,GSM5319850,GSM5319851,GSM5319852,GSM5319853,GSM5319854,GSM5319855,GSM5319856,GSM5319857,GSM5319858,GSM5319859,GSM5319860,GSM5319861,GSM5319862,GSM5319863,GSM5319864,GSM5319865,GSM5319866,GSM5319867,GSM5319868,GSM5319869,GSM5319870,GSM5319871,GSM5319872,GSM5319873,GSM5319874,GSM5319875,GSM5319876,GSM5319877,GSM5319878,GSM5319879,GSM5319880,GSM5319881,GSM5319882,GSM5319883,GSM5319884,GSM5319885,GSM5319886,GSM5319887,GSM5319888,GSM5319889,GSM5319890,GSM5319891,GSM5319892,GSM5319893,GSM5319894,GSM5319895,GSM5319896,GSM5319897,GSM5319898,GSM5319899,GSM5319900,GSM5319901,GSM5319902,GSM5319903,GSM5319904,GSM5319905,GSM5319906,GSM5319907,GSM5319908,GSM5319909,GSM5319910,GSM5319911,GSM5319912,GSM5319913,GSM5319914,GSM5319915,GSM5319916,GSM5319917,GSM5319918,GSM5319919,GSM5319920,GSM5319921,GSM5319922,GSM5319923,GSM5319924,GSM5319925,GSM5319926,GSM5319927,GSM5319928,GSM5319929,GSM5319930,GSM5319931,GSM5319932,GSM5319933,GSM5319934,GSM5319935,GSM5319936,GSM5319937,GSM5319938,GSM5319939,GSM5319940,GSM5319941,GSM5319942,GSM5319943,GSM5319944,GSM5319945,GSM5319946,GSM5319947
+Liver_Cancer,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0

output/preprocess/Liver_Cancer/clinical_data/GSE209875.csv

@@ -1,4 +1,4 @@
-GSM1,GSM2,GSM3,GSM4,GSM5,GSM6,GSM7,GSM8,GSM9,GSM10,GSM11,GSM12,GSM13,GSM14,GSM15,GSM16,GSM17,GSM18,GSM19,GSM20,GSM21
-1.0,1.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0
-63.0,34.0,73.0,76.0,71.0,68.0,39.0,31.0,48.0,66.0,62.0,75.0,65.0,55.0,,,,,,,
-1.0,0.0,,,,,,,,,,,,,,,,,,,
+,GSM6380607,GSM6380608,GSM6380609,GSM6380610,GSM6380611,GSM6380612,GSM6380613,GSM6380614,GSM6380615,GSM6380616,GSM6380617,GSM6380618,GSM6380619,GSM6380620,GSM6380621,GSM6380622,GSM6380623,GSM6380624,GSM6380625,GSM6380626,GSM6380627,GSM6380628,GSM6380629,GSM6380630,GSM6380631,GSM6380632,GSM6380633,GSM6380634,GSM6380635,GSM6380636,GSM6380637,GSM6380638,GSM6380639,GSM6380640,GSM6380641,GSM6380642,GSM6380643,GSM6380644,GSM6380645,GSM6380646,GSM6380647,GSM6380648,GSM6380649,GSM6380650,GSM6380651,GSM6380652,GSM6380653,GSM6380654,GSM6380655,GSM6380656,GSM6380657,GSM6380658,GSM6380659,GSM6380660,GSM6380661,GSM6380662,GSM6380663,GSM6380664,GSM6380665,GSM6380666,GSM6380667,GSM6380668,GSM6380669,GSM6380670,GSM6380671,GSM6380672,GSM6380673,GSM6380674,GSM6380675,GSM6380676,GSM6380677,GSM6380678
+Liver_Cancer,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0
+Age,63.0,34.0,73.0,76.0,71.0,68.0,39.0,31.0,73.0,48.0,63.0,34.0,73.0,76.0,71.0,68.0,39.0,31.0,73.0,48.0,66.0,66.0,68.0,62.0,63.0,71.0,66.0,75.0,71.0,75.0,66.0,66.0,68.0,62.0,63.0,71.0,66.0,75.0,71.0,75.0,65.0,55.0,76.0,75.0,65.0,55.0,76.0,75.0,63.0,34.0,73.0,76.0,71.0,68.0,39.0,31.0,73.0,48.0,66.0,66.0,68.0,62.0,63.0,71.0,66.0,75.0,71.0,75.0,65.0,55.0,76.0,75.0
+Gender,1.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,0.0,1.0,0.0,0.0,0.0,1.0,0.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,0.0,1.0,0.0

output/preprocess/Liver_Cancer/clinical_data/GSE218438.csv

@@ -0,0 +1,2 @@
+,GSM6744665,GSM6744666,GSM6744667,GSM6744668,GSM6744669,GSM6744670,GSM6744671,GSM6744672,GSM6744673,GSM6744674,GSM6744675,GSM6744676,GSM6744677,GSM6744678,GSM6744679,GSM6744680,GSM6744681,GSM6744682,GSM6744683,GSM6744684,GSM6744685,GSM6744686,GSM6744687,GSM6744688,GSM6744689,GSM6744690,GSM6744691,GSM6744692,GSM6744693,GSM6744694,GSM6744695,GSM6744696,GSM6744697,GSM6744698,GSM6744699,GSM6744700,GSM6744701,GSM6744702,GSM6744703,GSM6744704,GSM6744705,GSM6744706,GSM6744707,GSM6744708,GSM6744709,GSM6744710,GSM6744711,GSM6744712,GSM6744713,GSM6744714,GSM6744715,GSM6744716,GSM6744717,GSM6744718,GSM6744719,GSM6744720,GSM6744721,GSM6744722,GSM6744723,GSM6744724,GSM6744725,GSM6744726,GSM6744727,GSM6744728,GSM6744729,GSM6744730,GSM6744731,GSM6744732,GSM6744733,GSM6744734,GSM6744735,GSM6744736,GSM6744737,GSM6744738,GSM6744739,GSM6744740,GSM6744741,GSM6744742,GSM6744743,GSM6744744,GSM6744745,GSM6744746,GSM6744747,GSM6744748,GSM6744749,GSM6744750,GSM6744751,GSM6744752,GSM6744753,GSM6744754,GSM6744755,GSM6744756,GSM6744757,GSM6744758,GSM6744759,GSM6744760,GSM6744761,GSM6744762,GSM6744763,GSM6744764,GSM6744765,GSM6744766,GSM6744767,GSM6744768,GSM6744769,GSM6744770,GSM6744771,GSM6744772,GSM6744773,GSM6744774,GSM6744775,GSM6744776,GSM6744777,GSM6744778,GSM6744779,GSM6744780,GSM6744781,GSM6744782,GSM6744783,GSM6744784,GSM6744785,GSM6744786,GSM6744787,GSM6744788,GSM6744789,GSM6744790,GSM6744791,GSM6744792,GSM6744793,GSM6744794,GSM6744795,GSM6744796,GSM6744797,GSM6744798,GSM6744799,GSM6744800,GSM6744801,GSM6744802,GSM6744803,GSM6744804,GSM6744805,GSM6744806,GSM6744807,GSM6744808,GSM6744809,GSM6744810,GSM6744811,GSM6744812,GSM6744813,GSM6744814,GSM6744815,GSM6744816,GSM6744817,GSM6744818,GSM6744819,GSM6744820,GSM6744821,GSM6744822,GSM6744823,GSM6744824,GSM6744825,GSM6744826,GSM6744827,GSM6744828,GSM6744829,GSM6744830,GSM6744831,GSM6744832,GSM6744833,GSM6744834,GSM6744835,GSM6744836,GSM6744837,GSM6744838,GSM6744839,GSM6744840,GSM6744841,GSM6744842,GSM6744843,GSM6744844,GSM6744845,GSM6744846,GSM6744847,GSM6744848,GSM6744849,GSM6744850,GSM6744851,GSM6744852,GSM6744853,GSM6744854,GSM6744855,GSM6744856,GSM6744857,GSM6744858,GSM6744859,GSM6744860,GSM6744861,GSM6744862,GSM6744863,GSM6744864,GSM6744865,GSM6744866,GSM6744867,GSM6744868,GSM6744869,GSM6744870,GSM6744871,GSM6744872,GSM6744873,GSM6744874,GSM6744875,GSM6744876,GSM6744877,GSM6744878,GSM6744879,GSM6744880,GSM6744881,GSM6744882,GSM6744883,GSM6744884,GSM6744885,GSM6744886,GSM6744887,GSM6744888,GSM6744889,GSM6744890,GSM6744891,GSM6744892,GSM6744893,GSM6744894,GSM6744895,GSM6744896,GSM6744897,GSM6744898,GSM6744899,GSM6744900,GSM6744901,GSM6744902,GSM6744903,GSM6744904,GSM6744905,GSM6744906,GSM6744907,GSM6744908,GSM6744909,GSM6744910,GSM6744911,GSM6744912,GSM6744913,GSM6744914,GSM6744915,GSM6744916,GSM6744917,GSM6744918,GSM6744920,GSM6744924,GSM6744928,GSM6744932,GSM6744936,GSM6744938,GSM6744941,GSM6744944,GSM6744945,GSM6744946,GSM6744947,GSM6744948,GSM6744949,GSM6744950,GSM6744951,GSM6744952,GSM6744953,GSM6744954,GSM6744955,GSM6744956,GSM6744957,GSM6744958,GSM6744959,GSM6744960,GSM6744961,GSM6744962,GSM6744963,GSM6744964,GSM6744965,GSM6744966,GSM6744967,GSM6744968,GSM6744969,GSM6744970,GSM6744971,GSM6744972,GSM6744973,GSM6744974,GSM6744975,GSM6744976,GSM6744977,GSM6744978,GSM6744979,GSM6744980,GSM6744981,GSM6744982,GSM6744983,GSM6744984,GSM6744985,GSM6744986,GSM6744987,GSM6744988,GSM6744989,GSM6744990,GSM6744991,GSM6744992,GSM6744993,GSM6744994,GSM6744995,GSM6744996,GSM6744997,GSM6744998,GSM6744999,GSM6745000,GSM6745001,GSM6745002,GSM6745003,GSM6745004,GSM6745005,GSM6745006,GSM6745007,GSM6745008,GSM6745009,GSM6745010,GSM6745011,GSM6745012,GSM6745013,GSM6745014,GSM6745015,GSM6745016,GSM6745017,GSM6745018,GSM6745019,GSM6745020,GSM6745021,GSM6745022,GSM6745023,GSM6745024,GSM6745025,GSM6745026,GSM6745027,GSM6745028,GSM6745029,GSM6745030,GSM6745031,GSM6745032,GSM6745033,GSM6745034,GSM6745035,GSM6745036,GSM6745037,GSM6745038,GSM6745039,GSM6745040,GSM6745041,GSM6745042,GSM6745043,GSM6745044,GSM6745045,GSM6745046,GSM6745047,GSM6745048,GSM6745049,GSM6745050,GSM6745051,GSM6745052,GSM6745053,GSM6745054,GSM6745055,GSM6745056,GSM6745057,GSM6745058,GSM6745059,GSM6745060,GSM6745061,GSM6745062,GSM6745063,GSM6745064,GSM6745065,GSM6745066,GSM6745067,GSM6745068,GSM6745069,GSM6745070,GSM6745071,GSM6745072,GSM6745073,GSM6745074,GSM6745075,GSM6745076,GSM6745077,GSM6745078,GSM6745079,GSM6745080,GSM6745081,GSM6745082,GSM6745083,GSM6745084,GSM6745085,GSM6745086,GSM6745087,GSM6745088,GSM6745089,GSM6745090,GSM6745091,GSM6745092,GSM6745093,GSM6745094,GSM6745095,GSM6745096,GSM6745097,GSM6745098,GSM6745099,GSM6745100,GSM6745101,GSM6745102,GSM6745103,GSM6745104,GSM6745105,GSM6745106,GSM6745107,GSM6745108,GSM6745109,GSM6745110,GSM6745111,GSM6745112,GSM6745113,GSM6745114,GSM6745115,GSM6745116,GSM6745117,GSM6745118,GSM6745119,GSM6745120,GSM6745121,GSM6745122,GSM6745123,GSM6745124,GSM6745125,GSM6745126,GSM6745127,GSM6745128,GSM6745129,GSM6745130,GSM6745131,GSM6745132,GSM6745133,GSM6745134,GSM6745135,GSM6745136,GSM6745137,GSM6745138,GSM6745139,GSM6745140,GSM6745141,GSM6745142,GSM6745143,GSM6745144,GSM6745145,GSM6745146,GSM6745147,GSM6745148,GSM6745149,GSM6745150,GSM6745151,GSM6745152,GSM6745153,GSM6745154,GSM6745155,GSM6745156,GSM6745157,GSM6745158,GSM6745159,GSM6745160,GSM6745161,GSM6745162,GSM6745163,GSM6745164,GSM6745165,GSM6745166,GSM6745167,GSM6745168,GSM6745169,GSM6745170,GSM6745171,GSM6745172,GSM6745173,GSM6745174,GSM6745175,GSM6745176,GSM6745177,GSM6745178,GSM6745179,GSM6745180,GSM6745181,GSM6745182,GSM6745183,GSM6745184,GSM6745185,GSM6745186,GSM6745187,GSM6745188,GSM6745189,GSM6745190,GSM6745191,GSM6745192,GSM6745193,GSM6745194,GSM6745195,GSM6745196,GSM6745197,GSM6745198,GSM6745199,GSM6745200,GSM6745201,GSM6745202,GSM6745203,GSM6745204,GSM6745205,GSM6745206,GSM6745207,GSM6745208,GSM6745209,GSM6745210,GSM6745211,GSM6745212,GSM6745213,GSM6745214,GSM6745215,GSM6745216,GSM6745217,GSM6745218,GSM6745219,GSM6745220,GSM6745221,GSM6745222,GSM6745223,GSM6745224,GSM6745225,GSM6745226,GSM6745227,GSM6745228,GSM6745229,GSM6745230,GSM6745231,GSM6745232,GSM6745233,GSM6745234,GSM6745235,GSM6745236,GSM6745237,GSM6745238,GSM6745239,GSM6745240,GSM6745241,GSM6745242,GSM6745243,GSM6745244,GSM6745245,GSM6745246,GSM6745247,GSM6745248,GSM6745249,GSM6745250,GSM6745251,GSM6745252,GSM6745253,GSM6745254,GSM6745255,GSM6745256,GSM6745257,GSM6745258,GSM6745259,GSM6745260,GSM6745261,GSM6745262,GSM6745263,GSM6745264,GSM6745265,GSM6745266,GSM6745267,GSM6745268,GSM6745269,GSM6745270,GSM6745271,GSM6745272,GSM6745273,GSM6745274,GSM6745275,GSM6745276,GSM6745277,GSM6745278,GSM6745279,GSM6745280,GSM6745281,GSM6745282,GSM6745283,GSM6745284,GSM6745285,GSM6745286,GSM6745287,GSM6745288,GSM6745289,GSM6745290,GSM6745291,GSM6745292,GSM6745293,GSM6745294,GSM6745295,GSM6745296,GSM6745297,GSM6745298,GSM6745299,GSM6745300,GSM6745301,GSM6745302,GSM6745303,GSM6745304,GSM6745305,GSM6745306,GSM6745307,GSM6745308,GSM6745309,GSM6745310,GSM6745311,GSM6745312,GSM6745313,GSM6745314,GSM6745315,GSM6745316,GSM6745317,GSM6745318
+Liver_Cancer,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0

output/preprocess/Liver_Cancer/clinical_data/GSE228782.csv

@@ -1,2 +1,2 @@
-,2
-Liver_Cancer,1.0
+GSM7136390,GSM7136391,GSM7136392,GSM7136393,GSM7136394,GSM7136395,GSM7136396,GSM7136397,GSM7136398,GSM7136399,GSM7136400,GSM7136401,GSM7136402,GSM7136403,GSM7136404,GSM7136405,GSM7136406,GSM7136407,GSM7136408,GSM7136409,GSM7136410,GSM7136411,GSM7136412,GSM7136413,GSM7136414,GSM7136415,GSM7136416,GSM7136417,GSM7136418,GSM7136419,GSM7136420,GSM7136421,GSM7136422,GSM7136423,GSM7136424,GSM7136425,GSM7136426,GSM7136427,GSM7136428,GSM7136429,GSM7136430,GSM7136431,GSM7136432,GSM7136433,GSM7136434,GSM7136435,GSM7136436,GSM7136437,GSM7136438,GSM7136439,GSM7136440,GSM7136441,GSM7136442,GSM7136443,GSM7136444,GSM7136445,GSM7136446,GSM7136447,GSM7136448,GSM7136449,GSM7136450,GSM7136451,GSM7136452,GSM7136453,GSM7136454,GSM7136455,GSM7136456,GSM7136457,GSM7136458,GSM7136460,GSM7136462,GSM7136465,GSM7136468,GSM7136471,GSM7136472,GSM7136473,GSM7136474,GSM7136475,GSM7136476,GSM7136477,GSM7136478,GSM7136479,GSM7136480
+1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0

output/preprocess/Liver_Cancer/clinical_data/GSE228783.csv

@@ -1,2 +1,2 @@
-GSM7136321,GSM7136322,GSM7136323,GSM7136324,GSM7136325,GSM7136326,GSM7136327,GSM7136328,GSM7136329,GSM7136330,GSM7136331,GSM7136332,GSM7136333,GSM7136334,GSM7136335,GSM7136336,GSM7136337,GSM7136338,GSM7136339,GSM7136340,GSM7136341,GSM7136342,GSM7136343,GSM7136344,GSM7136345,GSM7136346,GSM7136347,GSM7136348,GSM7136349,GSM7136350,GSM7136351,GSM7136352,GSM7136353,GSM7136354,GSM7136355,GSM7136356,GSM7136357,GSM7136358,GSM7136359,GSM7136360,GSM7136361,GSM7136362,GSM7136363,GSM7136364,GSM7136365,GSM7136366,GSM7136367,GSM7136368,GSM7136369,GSM7136370,GSM7136371,GSM7136372,GSM7136373,GSM7136374,GSM7136375,GSM7136376,GSM7136377,GSM7136378,GSM7136379,GSM7136380,GSM7136381,GSM7136382,GSM7136383,GSM7136384,GSM7136385,GSM7136386,GSM7136387,GSM7136388,GSM7136389,GSM7136390,GSM7136391,GSM7136392,GSM7136393,GSM7136394,GSM7136395,GSM7136396,GSM7136397,GSM7136398,GSM7136399,GSM7136400,GSM7136401,GSM7136402,GSM7136403,GSM7136404,GSM7136405,GSM7136406,GSM7136407,GSM7136408,GSM7136409,GSM7136410,GSM7136411,GSM7136412,GSM7136413,GSM7136414,GSM7136415,GSM7136416,GSM7136417,GSM7136418,GSM7136419,GSM7136420,GSM7136421,GSM7136422,GSM7136423,GSM7136424,GSM7136425,GSM7136426,GSM7136427,GSM7136428,GSM7136429,GSM7136430,GSM7136431,GSM7136432,GSM7136433,GSM7136434,GSM7136435,GSM7136436,GSM7136437,GSM7136438,GSM7136439,GSM7136440,GSM7136441,GSM7136442,GSM7136443,GSM7136444,GSM7136445,GSM7136446,GSM7136447,GSM7136448,GSM7136449,GSM7136450,GSM7136451,GSM7136452,GSM7136453,GSM7136454,GSM7136455,GSM7136456,GSM7136457,GSM7136458,GSM7136460,GSM7136462,GSM7136465,GSM7136468,GSM7136471,GSM7136472,GSM7136473,GSM7136474,GSM7136475,GSM7136476,GSM7136477,GSM7136478,GSM7136479,GSM7136480
-0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0
+,GSM7136321,GSM7136322,GSM7136323,GSM7136324,GSM7136325,GSM7136326,GSM7136327,GSM7136328,GSM7136329,GSM7136330,GSM7136331,GSM7136332,GSM7136333,GSM7136334,GSM7136335,GSM7136336,GSM7136337,GSM7136338,GSM7136339,GSM7136340,GSM7136341,GSM7136342,GSM7136343,GSM7136344,GSM7136345,GSM7136346,GSM7136347,GSM7136348,GSM7136349,GSM7136350,GSM7136351,GSM7136352,GSM7136353,GSM7136354,GSM7136355,GSM7136356,GSM7136357,GSM7136358,GSM7136359,GSM7136360,GSM7136361,GSM7136362,GSM7136363,GSM7136364,GSM7136365,GSM7136366,GSM7136367,GSM7136368,GSM7136369,GSM7136370,GSM7136371,GSM7136372,GSM7136373,GSM7136374,GSM7136375,GSM7136376,GSM7136377,GSM7136378,GSM7136379,GSM7136380,GSM7136381,GSM7136382,GSM7136383,GSM7136384,GSM7136385,GSM7136386,GSM7136387,GSM7136388,GSM7136389,GSM7136390,GSM7136391,GSM7136392,GSM7136393,GSM7136394,GSM7136395,GSM7136396,GSM7136397,GSM7136398,GSM7136399,GSM7136400,GSM7136401,GSM7136402,GSM7136403,GSM7136404,GSM7136405,GSM7136406,GSM7136407,GSM7136408,GSM7136409,GSM7136410,GSM7136411,GSM7136412,GSM7136413,GSM7136414,GSM7136415,GSM7136416,GSM7136417,GSM7136418,GSM7136419,GSM7136420,GSM7136421,GSM7136422,GSM7136423,GSM7136424,GSM7136425,GSM7136426,GSM7136427,GSM7136428,GSM7136429,GSM7136430,GSM7136431,GSM7136432,GSM7136433,GSM7136434,GSM7136435,GSM7136436,GSM7136437,GSM7136438,GSM7136439,GSM7136440,GSM7136441,GSM7136442,GSM7136443,GSM7136444,GSM7136445,GSM7136446,GSM7136447,GSM7136448,GSM7136449,GSM7136450,GSM7136451,GSM7136452,GSM7136453,GSM7136454,GSM7136455,GSM7136456,GSM7136457,GSM7136458,GSM7136460,GSM7136462,GSM7136465,GSM7136468,GSM7136471,GSM7136472,GSM7136473,GSM7136474,GSM7136475,GSM7136476,GSM7136477,GSM7136478,GSM7136479,GSM7136480
+Liver_Cancer,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,,,,,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,,,,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,,,,,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0

output/preprocess/Liver_Cancer/clinical_data/GSE45032.csv

@@ -1,4 +1,4 @@
 ,GSM1096016,GSM1096017,GSM1096018,GSM1096019,GSM1096020,GSM1096021,GSM1096022,GSM1096023,GSM1096024,GSM1096025,GSM1096026,GSM1096027,GSM1096028,GSM1096029,GSM1096030,GSM1096031,GSM1096032,GSM1096033,GSM1096034,GSM1096035,GSM1096036,GSM1096037,GSM1096038,GSM1096039,GSM1096040,GSM1096041,GSM1096042,GSM1096043,GSM1096044,GSM1096045,GSM1096046,GSM1096047,GSM1096048,GSM1096049,GSM1096050,GSM1096051,GSM1096052,GSM1096053,GSM1096054,GSM1096055,GSM1096056,GSM1096057,GSM1096058,GSM1096059,GSM1096060,GSM1096061,GSM1096062,GSM1096063
 Liver_Cancer,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0
 Age,67.0,56.0,76.0,79.0,66.0,70.0,68.0,72.0,62.0,66.0,55.0,62.0,71.0,73.0,74.0,61.0,54.0,64.0,68.0,59.0,79.0,69.0,59.0,71.0,64.0,55.0,66.0,56.0,66.0,68.0,25.0,41.0,50.0,56.0,66.0,58.0,67.0,49.0,63.0,70.0,60.0,50.0,58.0,61.0,60.0,59.0,52.0,51.0
-Gender,1.0,1.0,0.0,1.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,1.0,0.0,1.0,1.0,1.0,0.0,1.0,1.0,1.0,0.0,0.0,1.0,1.0,1.0,0.0,1.0,0.0,0.0,0.0,1.0,0.0,1.0,1.0,1.0,0.0,0.0,1.0,1.0,0.0,1.0,1.0,1.0
+Gender,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0

output/preprocess/Liver_Cancer/code/GSE148346.py

@@ -0,0 +1,110 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE148346"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE148346"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE148346.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE148346.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE148346.csv"
+json_path = "./output/z4/preprocess/Liver_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 pandas as pd
+import numpy as np
+
+# 1. Gene Expression Data Availability
+# Given the study context (biopsy sub-study with molecular responses), gene expression data is likely available.
+is_gene_available = True
+
+# 2. Variable Availability and Data Type Conversion
+
+# 2.1 Data Availability
+# The provided sample characteristics do not include liver cancer status, age, or gender.
+trait_row = None
+age_row = None
+gender_row = None
+
+# 2.2 Data Type Conversion
+def _after_colon(x):
+    if x is None:
+        return None
+    if isinstance(x, str):
+        parts = x.split(":", 1)
+        return parts[1].strip() if len(parts) > 1 else x.strip()
+    return x
+
+def convert_trait(x):
+    # Trait is Liver_Cancer; no related field present in this dataset.
+    return None
+
+def convert_age(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    v = v.strip().lower()
+    if v in {"na", "n/a", "unknown", "", "nan"}:
+        return None
+    # Extract numeric age
+    try:
+        # Remove non-digit/non-dot characters
+        num = "".join(ch for ch in v if (ch.isdigit() or ch == '.' or ch == '-'))
+        if num == "" or num == "-" or num == ".":
+            return None
+        val = float(num)
+        # Age must be within a reasonable human range
+        if 0 <= val <= 120:
+            return val
+        return None
+    except Exception:
+        return None
+
+def convert_gender(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    v = v.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
+# Skipped because trait_row is None (no clinical trait data for Liver_Cancer in this dataset).

output/preprocess/Liver_Cancer/code/GSE164760.py

@@ -0,0 +1,201 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE164760"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE164760"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE164760.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE164760.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE164760.csv"
+json_path = "./output/z4/preprocess/Liver_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
+# 1) Gene expression availability (expression array mentioned in series summary)
+is_gene_available = True
+
+# 2) Variable availability and converters based on the provided sample characteristics
+# Sample Characteristics Dictionary shows only key 0 with tissue types
+trait_row = 0  # Use tissue to infer Liver_Cancer status
+age_row = None
+gender_row = None
+
+def _extract_value(cell):
+    if cell is None:
+        return None
+    if isinstance(cell, str):
+        parts = cell.split(":", 1)
+        val = parts[1].strip() if len(parts) == 2 else cell.strip()
+        return val if val != "" else None
+    return None
+
+def convert_trait(cell):
+    # Map tissue categories to Liver_Cancer: 1 for NASH-HCC tumor, else 0
+    v = _extract_value(cell)
+    if v is None:
+        return None
+    vl = v.lower()
+    # Explicit handling of known categories in this dataset
+    if vl == "nash-hcc tumor":
+        return 1
+    if vl in {"nash liver", "cirrhotic liver", "healthy liver"}:
+        return 0
+    if "non-tumoral" in vl or "adjacent" in vl:
+        return 0
+    # Conservative default: treat other non-matching tissues as non-cancer
+    return 0
+
+def convert_age(cell):
+    v = _extract_value(cell)
+    if v is None:
+        return None
+    # Try to extract a number from the string
+    import re
+    m = re.search(r"[-+]?\d*\.?\d+", v)
+    if m:
+        try:
+            age_val = float(m.group())
+            # Reasonable human age bounds
+            if 0 <= age_val <= 120:
+                return age_val
+        except Exception:
+            return None
+    return None
+
+def convert_gender(cell):
+    v = _extract_value(cell)
+    if v is None:
+        return None
+    vl = v.strip().lower()
+    if vl in {"female", "f", "woman", "women"}:
+        return 0
+    if vl in {"male", "m", "man", "men"}:
+        return 1
+    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 (only if trait data 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)
+
+    # Save clinical data
+    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 appropriate columns for mapping based on annotation preview:
+# Probe ID column: 'ID'
+# Gene symbol column: 'Gene Symbol'
+
+# Ensure we have the annotation dataframe (if not already in scope)
+try:
+    gene_annotation
+except NameError:
+    gene_annotation = get_gene_annotation(soft_file)
+
+# Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Gene Symbol')
+
+# Preserve the original probe-level data and map to gene-level
+probe_level_df = gene_data  # from Step 3
+gene_data = apply_gene_mapping(expression_df=probe_level_df, mapping_df=mapping_df)
+
+# Step 7: Data Normalization and Linking
+# 1. Normalize gene symbols and save gene-level 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 (use the correct clinical variable name)
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Handle missing values in the linked data
+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)
+is_trait_biased = bool(is_trait_biased)  # ensure Python-native bool
+
+# 5. Final validation and save cohort info (force Python bools to avoid JSON issues)
+is_gene_available = bool((normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0))
+is_trait_available = bool((trait in selected_clinical_df.index) and bool(selected_clinical_df.loc[trait].notna().any()))
+note = "INFO: Trait inferred from tissue; age and gender not available in matrix."
+
+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/Liver_Cancer/code/GSE174570.py

@@ -0,0 +1,187 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE174570"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE174570"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE174570.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE174570.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE174570.csv"
+json_path = "./output/z4/preprocess/Liver_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) Determine gene expression availability
+# Affymetrix Human Genome U219 Array => mRNA expression data available
+is_gene_available = True
+
+# 2) Variable availability based on the Sample Characteristics Dictionary shown
+# Keys: {0: ['disease state: HCC'], 1: ['tissue: Tumour (liver)', 'tissue: Non-tumour adjacent (liver)']}
+# disease state is constant; tissue distinguishes tumour vs non-tumour adjacent -> can infer trait
+trait_row = 1
+age_row = None
+gender_row = None
+
+# 2.2) Converters
+def _after_colon(value):
+    if value is None:
+        return ""
+    s = str(value)
+    if ":" in s:
+        s = s.split(":", 1)[1]
+    return s.strip()
+
+def convert_trait(value):
+    s = _after_colon(value).lower()
+    if s in {"", "na", "n/a", "none", "unknown"}:
+        return None
+    # Map tumour/tumor as 1 (Liver_Cancer), non-tumour/adjacent/normal as 0
+    # Prioritize non-tumor indicators to avoid false positives
+    if any(k in s for k in ["non-tumour", "non-tumor", "non tumor", "adjacent", "non-neoplastic", "normal"]):
+        return 0
+    if any(k in s for k in ["tumour", "tumor", "cancer", "malignant", "hcc"]):
+        return 1
+    return None
+
+def convert_age(value):
+    s = _after_colon(value).lower()
+    if s in {"", "na", "n/a", "none", "unknown"}:
+        return None
+    m = re.search(r"[-+]?\d*\.?\d+", s)
+    if m:
+        try:
+            return float(m.group())
+        except Exception:
+            return None
+    return None
+
+def convert_gender(value):
+    s = _after_colon(value).lower()
+    if s in {"", "na", "n/a", "none", "unknown"}:
+        return None
+    if "female" in s or s == "f":
+        return 0
+    if "male" in s or s == "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,
+        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)
+
+# 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 (e.g., 11715100_at, 11715101_s_at) are 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
+# Map probe IDs to gene symbols and aggregate to gene-level expression
+
+# 1-2) Identify appropriate columns and build mapping dataframe
+# Expression data index matches annotation column 'ID'; gene symbols are in 'Gene Symbol'
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Gene Symbol')
+
+# 3) Apply mapping: distribute multi-gene probe signal and sum to gene level
+probe_data = gene_data  # preserve original probe-level data from previous step
+gene_data = apply_gene_mapping(probe_data, 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)
+
+# Ensure clinical features dataframe is available (fallback to reload if needed)
+if 'selected_clinical_df' not in locals():
+    selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+
+# 2. Link clinical and genetic data
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)
+
+# 3. Missing value handling
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Bias assessment 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: Trait inferred from tissue labels (tumour=1, non-tumour adjacent=0); "
+        "age and gender unavailable in this series. Platform HG-U219; probe-to-gene mapping and "
+        "gene symbol normalization applied.")
+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/Liver_Cancer/code/GSE178201.py

@@ -0,0 +1,196 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE178201"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE178201"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE178201.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE178201.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE178201.csv"
+json_path = "./output/z4/preprocess/Liver_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 availability
+# L1000 is a gene expression platform (not miRNA or methylation).
+is_gene_available = True
+
+# 2) Variable availability and converters
+
+# From the sample characteristics, tissue is at key 0 and includes:
+# - 'tissue: HEPG2' (liver cancer cell line)
+# - 'tissue: MCF10A.*' (breast epithelial cell line)
+# We infer the trait "Liver_Cancer" as binary: 1 for HepG2 (liver cancer), 0 otherwise.
+trait_row = 0
+
+# Age and Gender are not present
+age_row = None
+gender_row = None
+
+def _extract_value(x):
+    if pd.isna(x):
+        return None
+    s = str(x)
+    if ":" in s:
+        s = s.split(":", 1)[1]
+    s = s.strip()
+    if s.lower() in {"", "na", "n/a", "none", "unknown", "null"}:
+        return None
+    return s
+
+def convert_trait(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    lv = v.lower()
+    # HepG2 is a hepatocellular carcinoma (liver cancer) cell line
+    if "hepg2" in lv:
+        return 1
+    # MCF10A cell lines are not liver cancer
+    if "mcf10a" in lv:
+        return 0
+    # Fallback heuristics in case of other labels
+    if any(k in lv for k in ["liver", "hepatocellular", "hcc"]):
+        return 1
+    return 0
+
+def convert_age(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    # Try to parse a number from the string
+    try:
+        # Remove common units if present
+        v_clean = "".join(ch for ch in v if (ch.isdigit() or ch == "." or ch == "-"))
+        return float(v_clean) if v_clean not in {"", "-"} else None
+    except Exception:
+        return None
+
+def convert_gender(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    lv = v.lower()
+    if lv in {"male", "m"}:
+        return 1
+    if lv 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 (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=None,
+        gender_row=gender_row,
+        convert_gender=None
+    )
+    # Preview and save
+    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
+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 observation:
+# - gene_data row IDs are numeric strings like '16', '23', etc. These match the 'pr_analyte_num' in annotation.
+# - Gene symbols are in 'pr_gene_symbol'.
+mapping_df = get_gene_mapping(gene_annotation, prob_col='pr_analyte_num', gene_col='pr_gene_symbol')
+
+# Normalize 'ID' in mapping to match gene_data index (e.g., convert '16.0' -> '16')
+mapping_df['ID'] = pd.to_numeric(mapping_df['ID'], errors='coerce')
+mapping_df = mapping_df.dropna(subset=['ID'])
+mapping_df['ID'] = mapping_df['ID'].astype(int).astype(str)
+
+# Apply mapping to convert probe/analyte-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
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+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 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.
+note_msg = "INFO: Trait is constant (all HepG2) in this matrix; dataset marked biased."
+is_usable = validate_and_save_cohort_info(
+    True, cohort, json_path, True, True, is_trait_biased, unbiased_linked_data, note=note_msg
+)
+
+# 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)

output/preprocess/Liver_Cancer/code/GSE209875.py

@@ -0,0 +1,190 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE209875"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE209875"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE209875.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE209875.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE209875.csv"
+json_path = "./output/z4/preprocess/Liver_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+import os
+import re
+from tools.preprocess import *
+
+# 1) Identify paths to the series matrix file and SOFT file (SOFT optional at this step)
+files = os.listdir(in_cohort_dir)
+
+# Prefer canonical GEO naming for series matrix files
+matrix_candidates = [
+    f for f in files
+    if re.search(r'series_matrix.*\.txt(\.gz)?$', f, re.IGNORECASE) or 'matrix' in f.lower()
+]
+
+if not matrix_candidates:
+    print("Files in cohort dir:", files)
+    raise FileNotFoundError("Series matrix file not found (e.g., *series_matrix.txt.gz).")
+
+matrix_file = os.path.join(in_cohort_dir, matrix_candidates[0])
+
+# Try to find a SOFT file; if absent, set to None and proceed
+soft_candidates = [
+    f for f in files
+    if re.search(r'\.soft(\.gz)?$', f, re.IGNORECASE) or 'family.soft' in f.lower()
+]
+soft_file = os.path.join(in_cohort_dir, soft_candidates[0]) if soft_candidates else None
+
+print("Files in cohort dir:", files)
+print(f"Selected series matrix file: {matrix_file}")
+print(f"Selected SOFT file: {soft_file if soft_file is not None else 'None (not found)'}")
+
+# 2) Obtain background information and clinical dataframe from the matrix file by prefix matching
+background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']
+clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']
+
+try:
+    if matrix_file.lower().endswith('.gz'):
+        background_info, clinical_data = get_background_and_clinical_data(
+            matrix_file, background_prefixes, clinical_prefixes
+        )
+    else:
+        # Fallback for plain text files
+        with open(matrix_file, 'rt', encoding='utf-8', errors='ignore') as fh:
+            matrix_text = fh.read()
+        background_info, clinical_data = filter_content_by_prefix(
+            matrix_text,
+            prefixes_a=background_prefixes,
+            prefixes_b=clinical_prefixes,
+            unselect=False,
+            source_type='string',
+            return_df_a=False,
+            return_df_b=True
+        )
+except Exception as e:
+    print(f"Error while reading matrix file: {matrix_file}")
+    print(f"Available files: {files}")
+    raise
+
+# 3) Create a dictionary of unique values for each clinical feature row
+if clinical_data is not None and not clinical_data.empty:
+    sample_characteristics_dict = get_unique_values_by_row(clinical_data, max_len=30)
+else:
+    sample_characteristics_dict = {}
+    print("Warning: clinical_data is empty or None.")
+
+# 4) Print 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
+# The selected platform indicates miRNA-only data in this subseries.
+is_gene_available = False
+
+# 2) Variable availability (from provided Sample Characteristics Dictionary)
+trait_row = 0    # 'histology' field
+age_row = 1      # 'age' field
+gender_row = 2   # 'Sex' field
+
+# 2) Conversion functions
+def _after_colon(x):
+    if x is None:
+        return None
+    try:
+        s = str(x)
+    except Exception:
+        return None
+    if ':' in s:
+        return s.split(':', 1)[1].strip()
+    return s.strip()
+
+def convert_trait(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    vl = v.lower()
+    # Non-tumor and benign considered controls
+    if 'non-tumor' in vl or 'non tumor' in vl or 'nontumor' in vl:
+        return 0
+    if 'benign' in vl:
+        return 0
+    # Carcinomas considered cases
+    if 'hepatocellular carcinoma' in vl or 'cholangiocarcinoma' in vl or 'carcinoma' in vl:
+        return 1
+    # Heuristic: "tumor part" without benign mentioned implies tumor
+    if 'tumor part' in vl:
+        return 1
+    return None
+
+def convert_age(x):
+    v = _after_colon(x)
+    if v is None or v == '':
+        return None
+    # Remove any non-numeric trailing characters if present
+    try:
+        return float(str(v).strip())
+    except Exception:
+        # Try extracting leading numeric
+        import re
+        m = re.search(r'[-+]?\d+\.?\d*', str(v))
+        if m:
+            try:
+                return float(m.group(0))
+            except Exception:
+                return None
+        return None
+
+def convert_gender(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    vl = v.strip().lower()
+    if vl in ['m', 'male']:
+        return 1
+    if vl in ['f', 'female']:
+        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_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,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    # Preview
+    preview = preview_df(selected_clinical_df)
+    print("Preview of selected clinical features:", 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)

output/preprocess/Liver_Cancer/code/GSE212047.py

@@ -0,0 +1,210 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE212047"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE212047"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE212047.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE212047.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE212047.csv"
+json_path = "./output/z4/preprocess/Liver_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 data availability based on provided background and sample characteristics.
+# Assumptions:
+# - This cohort appears to be a mouse HSC dataset (non-human), with no explicit human trait (Liver_Cancer), age, or gender fields.
+# - It likely contains gene expression (bulk RNA-seq or microarray), not miRNA-only or methylation-only.
+
+is_gene_available = True  # GEO series description and sample characteristics suggest RNA-seq/microarray data.
+trait_row = None          # No human Liver_Cancer case/control status in sample characteristics.
+age_row = None            # No age information.
+gender_row = None         # No gender information.
+
+# Conversion functions (robust to typical GEO "key: value" formatting).
+def _after_colon(val):
+    if val is None:
+        return None
+    s = str(val)
+    parts = s.split(":", 1)
+    return parts[1].strip() if len(parts) == 2 else s.strip()
+
+def convert_trait(x):
+    v = _after_colon(x)
+    if v is None or v == "":
+        return None
+    # Heuristic mapping for liver cancer status if ever encountered:
+    vl = v.lower()
+    # Positive indicators
+    pos_terms = ["hcc", "hepatocellular carcinoma", "liver cancer", "tumor", "tumour", "cancer", "neoplasm", "malignant"]
+    # Negative indicators
+    neg_terms = ["normal", "non-tumor", "non tumour", "noncancer", "control", "nt", "healthy", "adjacent non-tumor"]
+    if any(t in vl for t in pos_terms):
+        return 1
+    if any(t in vl for t in neg_terms):
+        return 0
+    return None
+
+def convert_age(x):
+    v = _after_colon(x)
+    if v is None or v == "":
+        return None
+    vl = v.lower().replace("years", "").replace("year", "").replace("yrs", "").replace("yr", "").replace("old", "").strip()
+    # Try to parse as float
+    try:
+        return float(vl)
+    except:
+        return None
+
+def convert_gender(x):
+    v = _after_colon(x)
+    if v is None or v == "":
+        return None
+    vl = v.lower()
+    if vl in ["male", "m", "man"]:
+        return 1
+    if vl in ["female", "f", "woman"]:
+        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 only if trait data is available (not the case here)
+if trait_row is not None:
+    selected = 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, n=5)
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected.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
+# The observed identifiers are numeric probe-like IDs (e.g., '10338001'), 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
+# Decide identifier and gene symbol columns based on annotation preview:
+# - Probe/feature identifiers: 'ID' (matches numeric IDs in expression data)
+# - Gene symbols/info: 'gene_assignment' (contains gene symbols within mixed annotations)
+
+# 1-2. Build mapping dataframe
+mapping_df = get_gene_mapping(annotation=gene_annotation, prob_col='ID', gene_col='gene_assignment')
+
+# Improve symbol extraction for mouse-style mixed-case symbols by uppercasing before regex extraction
+mapping_df['Gene'] = mapping_df['Gene'].astype(str).str.upper()
+
+# 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 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)
+
+# Determine trait availability from earlier step context
+is_trait_available = (trait_row is not None)
+
+# If trait is unavailable, record metadata and skip linking/QC
+if not is_trait_available:
+    _ = validate_and_save_cohort_info(
+        is_final=False,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=True,
+        is_trait_available=False
+    )
+else:
+    # 2. Link clinical and genetic data
+    selected_clinical_data = 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
+    )
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, 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
+    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 final validation since clinical trait was available."
+    )
+
+    # 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/Liver_Cancer/code/GSE218438.py

@@ -0,0 +1,288 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE218438"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE218438"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE218438.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE218438.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE218438.csv"
+json_path = "./output/z4/preprocess/Liver_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: Determine gene expression availability
+is_gene_available = True  # L1000 transcriptomics indicates gene expression data
+
+# Step 2: Determine availability of trait, age, and gender, and define converters
+trait_row = 0   # 'cell type' row
+age_row = None
+gender_row = None
+
+def _extract_value(x):
+    if x is None:
+        return None
+    try:
+        s = str(x)
+    except Exception:
+        return None
+    parts = s.split(":", 1)
+    v = parts[1] if len(parts) == 2 else parts[0]
+    v = v.strip()
+    return v if v != "" else None
+
+def convert_trait(x):
+    v = _extract_value(x)
+    if v is None:
+        return None
+    v_low = v.lower()
+    # Positive if hepatocellular carcinoma or clearly liver-related
+    if ("hepatocellular" in v_low) or ("hepat" in v_low) or ("liver" in v_low):
+        return 1
+    # Otherwise, considered non-liver cancer related
+    return 0
+
+def convert_age(x):
+    # Not available in this dataset
+    return None
+
+def convert_gender(x):
+    # Not available in this dataset
+    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 (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,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    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)
+else:
+    print("No clinical trait data available; skipping clinical feature extraction.")
+
+# 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 (e.g., 1007_s_at) are not human gene symbols; mapping to gene symbols is required.
+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
+# Robust probe->gene symbol mapping with safeguards
+
+# 1) Identify the probe ID column by overlap with expression IDs
+id_overlap_counts = {}
+for col in gene_annotation.columns:
+    try:
+        overlap = gene_annotation[col].astype(str).str.strip().isin(gene_data.index).sum()
+        id_overlap_counts[col] = int(overlap)
+    except Exception:
+        continue
+
+if not id_overlap_counts or max(id_overlap_counts.values()) == 0:
+    possible_id_cols = [c for c in ['ID', 'SPOT_ID'] if c in gene_annotation.columns]
+    probe_id_col = possible_id_cols[0] if possible_id_cols else None
+else:
+    probe_id_col = max(id_overlap_counts, key=id_overlap_counts.get)
+
+# 2) Identify a likely gene symbol column with stricter rules
+symbol_priority = [
+    'Gene Symbol', 'GENE_SYMBOL', 'GeneSymbol', 'Symbol', 'SYMBOL', 'GENE_SYMBOLS',
+    'gene_assignment', 'Gene Assignment', 'GENE_ASSIGNMENT', 'Associated Gene Name',
+    'ENTREZ_GENE_ID', 'ENTREZ_GENE', 'ORF', 'ORF_NAME', 'Gene Title', 'GENE_TITLE', 'Gene'
+]
+
+lower_to_original = {c.lower(): c for c in gene_annotation.columns}
+gene_symbol_col = None
+for name in symbol_priority:
+    if name.lower() in lower_to_original:
+        gene_symbol_col = lower_to_original[name.lower()]
+        break
+
+# If no priority column found, do content-based detection restricted to plausible columns
+if gene_symbol_col is None:
+    exclude_cols = set()
+    if probe_id_col is not None:
+        exclude_cols.add(probe_id_col)
+    # Explicitly exclude known non-symbol columns
+    for c in ['FLAG', 'SEQUENCE', 'SPOT_ID']:
+        if c in gene_annotation.columns:
+            exclude_cols.add(c)
+
+    # Restrict to columns whose names suggest gene info
+    def is_plausible_symbol_col(colname: str) -> bool:
+        cl = colname.lower()
+        keys = ['gene', 'symbol', 'assign', 'title', 'entrez', 'orf', 'locus']
+        return any(k in cl for k in keys)
+
+    candidates = [c for c in gene_annotation.columns if c not in exclude_cols and is_plausible_symbol_col(c)]
+
+    best_col = None
+    best_hits = -1
+    for col in candidates:
+        s = gene_annotation[col].dropna().astype(str)
+        hits = s.map(lambda x: len(extract_human_gene_symbols(x)) > 0).sum()
+        if hits > best_hits:
+            best_hits = int(hits)
+            best_col = col
+    gene_symbol_col = best_col if best_hits > 0 else None
+
+print(f"Selected probe_id_col: {probe_id_col}")
+print(f"Selected gene_symbol_col: {gene_symbol_col}")
+
+mapped_gene_data = None
+mapping_valid = False
+
+if (probe_id_col is not None) and (gene_symbol_col is not None):
+    # Pre-check: ensure mapping yields a reasonable number of unique symbols
+    try:
+        tmp_map = gene_annotation[[probe_id_col, gene_symbol_col]].dropna()
+        token_lists = tmp_map[gene_symbol_col].astype(str).map(extract_human_gene_symbols)
+        unique_genes = set()
+        for lst in token_lists:
+            unique_genes.update(lst)
+        n_unique_genes = len(unique_genes)
+        print(f"Pre-check: unique extracted gene symbols = {n_unique_genes}")
+
+        mapping_valid = n_unique_genes >= 100  # sanity threshold
+    except Exception:
+        mapping_valid = False
+
+    if mapping_valid:
+        mapping_df = get_gene_mapping(gene_annotation, prob_col=probe_id_col, gene_col=gene_symbol_col)
+        try:
+            mapped_gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)
+        except Exception:
+            mapped_gene_data = None
+
+# Post-check: ensure resulting gene-level data is reasonable
+if isinstance(mapped_gene_data, pd.DataFrame):
+    n_rows = mapped_gene_data.shape[0]
+    if n_rows < 100:
+        print(f"Post-check failed: mapped gene data has too few genes ({n_rows}). Falling back to probe-level.")
+        mapped_gene_data = None
+
+# Apply mapping if successful
+if isinstance(mapped_gene_data, pd.DataFrame) and mapped_gene_data.shape[0] > 0:
+    gene_data = mapped_gene_data
+else:
+    # Mapping not applied; keeping probe-level data
+    pass
+
+# Step 7: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1. Normalize gene symbols; if normalization fails (e.g., probe IDs), fall back to probe-level data
+normalization_note = "INFO: Gene symbols normalized using NCBI synonym mapping."
+try:
+    normalized_gene_data = normalize_gene_symbols_in_index(gene_data.copy())
+except Exception as e:
+    normalized_gene_data = pd.DataFrame()
+
+# Heuristic: if normalization produced empty data, likely probe-level IDs; fall back to original
+if not isinstance(normalized_gene_data, pd.DataFrame) or normalized_gene_data.shape[0] == 0:
+    final_gene_data = gene_data.copy()
+    normalization_note = ("WARNING: Probe-to-gene symbol mapping was unavailable; normalization produced empty gene data. "
+                          "Proceeding with probe-level Affymetrix IDs as features.")
+else:
+    final_gene_data = normalized_gene_data
+
+# Ensure output directory exists and save gene data
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+final_gene_data.to_csv(out_gene_data_file)
+
+# 2. Link the clinical and genetic data
+# Use the variable from Step 2 if available; otherwise reload from disk
+if 'selected_clinical_df' not in globals():
+    if os.path.exists(out_clinical_data_file):
+        selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
+    else:
+        raise RuntimeError("Clinical data is not available in memory or on disk.")
+
+linked_data = geo_link_clinical_genetic_data(selected_clinical_df, final_gene_data)
+
+# 3. Handle missing values in the linked data
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Determine whether the trait and 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 final quality validation and save cohort info
+has_gene = isinstance(final_gene_data, pd.DataFrame) and final_gene_data.shape[0] > 0 and final_gene_data.shape[1] > 0
+has_trait = isinstance(selected_clinical_df, pd.DataFrame) and (trait in selected_clinical_df.index) and selected_clinical_df.shape[1] > 0
+
+note = f"{normalization_note} INFO: Gene feature count={final_gene_data.shape[0]}, sample count={final_gene_data.shape[1]}."
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort=cohort,
+    info_path=json_path,
+    is_gene_available=has_gene,
+    is_trait_available=has_trait,
+    is_biased=is_trait_biased,
+    df=unbiased_linked_data,
+    note=note
+)
+
+# 6. Save the 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/Liver_Cancer/code/GSE228782.py

@@ -0,0 +1,210 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE228782"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE228782"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE228782.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE228782.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE228782.csv"
+json_path = "./output/z4/preprocess/Liver_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 availability
+is_gene_available = True  # Affymetrix transcriptome profiles indicate gene expression data
+
+# 2) Variable availability
+trait_row = 2   # 'disease' field
+age_row = None  # Not available
+gender_row = None  # Not available
+
+is_trait_available = trait_row is not None
+
+# 2.2) Conversion functions
+def _after_colon(val):
+    if val is None or (isinstance(val, float) and pd.isna(val)):
+        return None
+    s = str(val)
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    s = s.strip()
+    if s.lower() in {'', 'na', 'n/a', 'null', 'none', 'nan'}:
+        return None
+    return s
+
+def convert_trait(v):
+    """
+    Define Liver_Cancer as primary liver cancer: HCC or CCC -> 1
+    CRC metastasis and 'other' -> 0
+    """
+    x = _after_colon(v)
+    if x is None:
+        return None
+    xl = x.lower()
+    # Primary liver cancers
+    if xl in {'hcc', 'ccc'} or 'cholangiocarcinoma' in xl:
+        return 1
+    # Non-primary liver cancers or other conditions
+    if 'crc' in xl or 'met' in xl or 'metast' in xl:
+        return 0
+    if xl == 'other':
+        return 0
+    # Fallback heuristic
+    if 'hepatocellular' in xl:
+        return 1
+    return None
+
+def convert_age(v):
+    x = _after_colon(v)
+    if x is None:
+        return None
+    try:
+        num = ''.join(ch for ch in x if (ch.isdigit() or ch in {'.', '-', '+'}))
+        if num in {'', '.', '-', '+', '+.', '-.'}:
+            return None
+        return float(num)
+    except Exception:
+        return None
+
+def convert_gender(v):
+    x = _after_colon(v)
+    if x is None:
+        return None
+    xl = x.lower()
+    if xl in {'female', 'f', 'woman', 'women'}:
+        return 0
+    if xl in {'male', 'm', 'man', 'men'}:
+        return 1
+    return 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 (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=None,
+        gender_row=gender_row,
+        convert_gender=None
+    )
+    clinical_preview = preview_df(selected_clinical_df)
+    print(clinical_preview)
+    # 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=False)
+
+# 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 previous previews:
+# Probe/ID column: 'ID'
+# Gene symbol column: 'Gene Symbol'
+
+# 1-2) Build mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Gene Symbol')
+
+# 3) Apply mapping to convert probe-level 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 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 clinical and genetic data
+# Ensure clinical data is available in memory; if not, reload from CSV saved in Step 2
+try:
+    selected_clinical_df
+except NameError:
+    tmp = pd.read_csv(out_clinical_data_file)
+    # The saved clinical CSV had index=False; restore the single row index as the trait name
+    tmp.index = [trait]
+    selected_clinical_df = tmp
+
+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 if any
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5) Final validation and save cohort info
+note = ("INFO: Gene mapping performed from Affymetrix HG-U219 probes to symbols; "
+        "clinical trait derived from 'disease' field; no age/gender available.")
+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/Liver_Cancer/code/GSE228783.py

@@ -0,0 +1,171 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE228783"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE228783"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE228783.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE228783.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE228783.csv"
+json_path = "./output/z4/preprocess/Liver_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
+
+# 1) Determine gene expression availability
+is_gene_available = True  # Transcriptome study; likely gene expression microarray/RNA-seq
+
+# 2) Variable availability and converters
+trait_row = 2  # 'disease' field with multiple categories
+age_row = None
+gender_row = None
+
+def _after_colon(x):
+    if x is None:
+        return None
+    if isinstance(x, str):
+        parts = x.split(":", 1)
+        val = parts[1] if len(parts) > 1 else parts[0]
+        return val.strip()
+    return x
+
+def convert_trait(x):
+    v = _after_colon(x)
+    if v is None:
+        return None
+    v_low = v.lower()
+    # Binary: primary liver cancer (HCC, CCC) = 1; CRC liver metastasis = 0; other/unknown = None
+    if v_low in {"hcc", "hepatocellular carcinoma", "ccc", "cholangiocarcinoma"}:
+        return 1
+    if v_low in {"crc met", "colorectal cancer metastasis", "crc metastasis"}:
+        return 0
+    if v_low in {"other", "na", "n/a", "unknown"}:
+        return None
+    return None
+
+def convert_age(x):
+    # No age information available in this dataset
+    return None
+
+def convert_gender(x):
+    # No gender information available in this dataset
+    return None
+
+# 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 (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
+    )
+    print(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
+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'
+gene_symbol_col = 'Gene Symbol'
+
+# 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(gene_data, mapping_df)
+
+# Step 7: Data Normalization and Linking
+import os
+
+# 1. Normalize gene symbols and save gene-level 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. 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_gene_available = True
+is_trait_available = True
+note = "INFO: Affymetrix HG-U219 probes mapped to symbols via platform annotation; no age/gender fields provided."
+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/Liver_Cancer/code/GSE45032.py

@@ -0,0 +1,185 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE45032"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE45032"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE45032.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE45032.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE45032.csv"
+json_path = "./output/z4/preprocess/Liver_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: Gene Expression Data Availability
+is_gene_available = True  # Microarray mRNA expression per background info
+
+# Step 2: Variable Availability and Conversion Functions
+# Identified keys from Sample Characteristics Dictionary:
+# 0: 'cell type: hepatocallular carcinoma' vs 'cell type: chronic hepatitis type C' -> trait
+# 2: 'gender: male/female' -> gender
+# 3: 'age(yrs): <number>' -> age
+trait_row = 0
+age_row = 3
+gender_row = 2
+
+def _extract_value_after_colon(x: str) -> str:
+    if x is None:
+        return ""
+    parts = str(x).split(":", 1)
+    val = parts[1] if len(parts) > 1 else parts[0]
+    return val.strip()
+
+def convert_trait(x):
+    # Map Liver_Cancer status: HCC -> 1, CHC -> 0
+    val = _extract_value_after_colon(x).lower()
+    # Normalize common typos and abbreviations
+    val_norm = val.replace("-", " ").replace("_", " ").strip()
+    # Positive case: hepatocellular carcinoma (HCC)
+    if "hcc" in val_norm:
+        return 1
+    if ("hepatoc" in val_norm and "carcinoma" in val_norm) or "hepatocellular carcinoma" in val_norm:
+        return 1
+    # Negative case: chronic hepatitis C (CHC)
+    if "chc" in val_norm:
+        return 0
+    if "chronic hepatitis" in val_norm or "hepatitis" in val_norm:
+        return 0
+    # If it's explicitly "normal" or similar (not expected here), treat as control
+    if "normal" in val_norm or "control" in val_norm:
+        return 0
+    return None
+
+def convert_age(x):
+    import re
+    val = _extract_value_after_colon(x).lower()
+    # Extract first integer number as age
+    m = re.search(r"\d+", val)
+    if m:
+        return float(m.group(0))
+    return None
+
+def convert_gender(x):
+    val = _extract_value_after_colon(x).lower()
+    if "male" in val:
+        return 1
+    if "female" in val:
+        return 0
+    return None
+
+# Determine trait availability based on trait_row presence
+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 (since trait_row is 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 and save
+clinical_preview = preview_df(selected_clinical_df)
+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 the appropriate columns for mapping based on earlier previews:
+# - Gene expression data uses numeric 'ID' as identifiers.
+# - Gene symbols are stored in the 'GeneName' column of the annotation.
+
+# 1-2. Build mapping dataframe from annotation
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='GeneName')
+
+# 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
+# 1. Normalize gene symbols and save
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+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. 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: Probe->gene mapping used annotation columns ID->GeneName; "
+        "split multi-gene probe signal equally and summed per gene; "
+        "normalized symbols using NCBI synonym map; "
+        "dropped genes with >20% missing and samples with >5% missing genes; "
+        "imputed Gender with mode and others with mean.")
+is_usable = validate_and_save_cohort_info(
+    True,
+    cohort,
+    json_path,
+    is_gene_available,
+    is_trait_available,
+    is_trait_biased,
+    unbiased_linked_data,
+    note=note
+)
+
+# 6. Save linked data if usable
+if is_usable:
+    unbiased_linked_data.to_csv(out_data_file)

output/preprocess/Liver_Cancer/code/GSE66843.py

@@ -0,0 +1,229 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+cohort = "GSE66843"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_Cancer"
+in_cohort_dir = "../DATA/GEO/Liver_Cancer/GSE66843"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/GSE66843.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/GSE66843.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/GSE66843.csv"
+json_path = "./output/z4/preprocess/Liver_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 Data Availability
+is_gene_available = True  # Based on background info, this series likely contains mRNA gene expression (not pure miRNA/methylation)
+
+# 2. Variable Availability and Data Type Conversion
+
+# 2.1 Data Availability
+# Sample Characteristics Dictionary indicates:
+# 0: time post infection, 1: infection status, 2: cell line (Huh7.5.1)
+# None of these provide human-level Liver_Cancer trait, age, or gender info; also constant cell line metadata wouldn't be useful.
+trait_row = None
+age_row = None
+gender_row = None
+
+# 2.2 Data Type Conversion
+
+def _after_colon(x):
+    if x is None:
+        return None
+    if isinstance(x, (int, float)):
+        return str(x)
+    s = str(x)
+    parts = s.split(":", 1)
+    return parts[1].strip() if len(parts) == 2 else s.strip()
+
+def convert_trait(x):
+    """
+    Binary: 1 = Liver cancer/HCC/tumor case; 0 = non-cancer/normal/control.
+    Unknown -> None.
+    """
+    val = _after_colon(x).lower() if _after_colon(x) is not None else None
+    if val is None or val == "":
+        return None
+    # Positive indicators
+    pos_keywords = [
+        "liver cancer", "hepatocellular carcinoma", "hcc", "tumor", "tumour", "carcinoma", "cancer"
+    ]
+    # Negative indicators
+    neg_keywords = [
+        "normal", "non-tumor", "non tumour", "adjacent normal", "healthy", "control", "mock"
+    ]
+    if any(k in val for k in pos_keywords):
+        return 1
+    if any(k in val for k in neg_keywords):
+        return 0
+    return None
+
+def convert_age(x):
+    """
+    Continuous: extract numeric age in years. Unknown -> None.
+    """
+    val = _after_colon(x)
+    if val is None or val == "":
+        return None
+    s = val.lower()
+    # common patterns like "57", "57 years", "age 57", ">=50"
+    m = re.search(r'(\d+(\.\d+)?)', s)
+    if m:
+        try:
+            return float(m.group(1))
+        except:
+            return None
+    return None
+
+def convert_gender(x):
+    """
+    Binary: female=0, male=1. Unknown -> None.
+    """
+    val = _after_colon(x)
+    if val is None or val == "":
+        return None
+    s = val.strip().lower()
+    if s in {"male", "m", "man"}:
+        return 1
+    if s in {"female", "f", "woman"}:
+        return 0
+    # Handle encoded gender if present
+    if s in {"0", "1"}:
+        # Ambiguous without mapping; return None to avoid wrong assignment
+        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 (skip because trait_row is None)
+# If clinical data 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 = preview_df(selected_df)
+#     os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+#     selected_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
+# Illumina probe IDs (e.g., ILMN_1343291) are not human gene symbols and require mapping.
+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
+# 1-2. Determine appropriate columns for mapping and create mapping dataframe
+# Probe identifiers in expression data are Illumina IDs like 'ILMN_1343291', which correspond to 'ID' in annotation.
+# Gene symbols are in the 'Symbol' column.
+mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='Symbol')
+
+# 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
+import pandas as pd
+
+# 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)
+
+# 2-6. Link clinical and genetic data only if processed clinical features exist; otherwise finalize as unusable
+if os.path.exists(out_clinical_data_file):
+    # Load previously saved clinical features
+    selected_clinical_data = pd.read_csv(out_clinical_data_file, index_col=0)
+
+    # Link clinical and genetic data
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, 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)
+
+    # 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 loaded from file and linked successfully."
+    )
+
+    # 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)
+else:
+    # Trait/clinical data not available; perform final validation and ensure gene availability is correctly recorded.
+    # Pass a non-empty, wide dataframe to avoid abnormality override.
+    _ = 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/clinical data not available for this cohort; linking skipped. Only gene data saved."
+    )

output/preprocess/Liver_Cancer/code/TCGA.py

@@ -0,0 +1,353 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_Cancer"
+
+# Input paths
+tcga_root_dir = "../DATA/TCGA"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_Cancer/TCGA.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_Cancer/gene_data/TCGA.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_Cancer/clinical_data/TCGA.csv"
+json_path = "./output/z4/preprocess/Liver_Cancer/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+import os
+import pandas as pd
+
+# Find candidate cohort directories
+subdirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+
+# Select the most specific cohort matching Liver Cancer (prefer LIHC)
+def score_dir(name: str) -> int:
+    lname = name.lower()
+    score = 0
+    if "liver_cancer_(lihc)" in lname:
+        score += 100
+    if "lihc" in lname:
+        score += 50
+    if "liver" in lname:
+        score += 10
+    return score
+
+scored = [(score_dir(d), d) for d in subdirs]
+scored.sort(reverse=True)
+best_score, selected_dir = scored[0] if scored else (0, None)
+
+if best_score <= 0 or selected_dir is None:
+    # No suitable directory found; mark task as skipped for 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 = selected_dir
+    cohort_dir = os.path.join(tcga_root_dir, cohort)
+
+    # Identify clinical and genetic file paths
+    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)
+
+    # Load 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)
+
+    # Print clinical column names
+    print(list(clinical_df.columns))
+
+# Step 2: Find Candidate Demographic Features
+import os
+import pandas as pd
+
+# Use the provided column list from the previous step as the basis for detection
+available_cols = ['_INTEGRATION', '_PATIENT', '_cohort', '_primary_disease', '_primary_site', 'additional_pharmaceutical_therapy', 'additional_radiation_therapy', 'adjacent_hepatic_tissue_inflammation_extent_type', 'age_at_initial_pathologic_diagnosis', 'albumin_result_lower_limit', 'albumin_result_specified_value', 'albumin_result_upper_limit', 'bcr_followup_barcode', 'bcr_patient_barcode', 'bcr_sample_barcode', 'bilirubin_lower_limit', 'bilirubin_upper_limit', 'cancer_first_degree_relative', 'child_pugh_classification_grade', 'creatinine_lower_level', 'creatinine_upper_limit', 'creatinine_value_in_mg_dl', 'days_to_birth', 'days_to_collection', 'days_to_death', 'days_to_initial_pathologic_diagnosis', 'days_to_last_followup', 'days_to_new_tumor_event_additional_surgery_procedure', 'days_to_new_tumor_event_after_initial_treatment', 'eastern_cancer_oncology_group', 'fetoprotein_outcome_lower_limit', 'fetoprotein_outcome_upper_limit', 'fetoprotein_outcome_value', 'fibrosis_ishak_score', 'followup_case_report_form_submission_reason', 'form_completion_date', 'gender', 'height', 'hist_hepato_carc_fact', 'hist_hepato_carcinoma_risk', 'histological_type', 'history_of_neoadjuvant_treatment', 'icd_10', 'icd_o_3_histology', 'icd_o_3_site', 'informed_consent_verified', 'initial_weight', 'inter_norm_ratio_lower_limit', 'intern_norm_ratio_upper_limit', 'is_ffpe', 'lost_follow_up', 'neoplasm_histologic_grade', 'new_neoplasm_event_occurrence_anatomic_site', 'new_neoplasm_event_type', 'new_neoplasm_occurrence_anatomic_site_text', 'new_tumor_event_ablation_embo_tx', 'new_tumor_event_additional_surgery_procedure', 'new_tumor_event_after_initial_treatment', 'new_tumor_event_liver_transplant', 'oct_embedded', 'other_dx', 'pathologic_M', 'pathologic_N', 'pathologic_T', 'pathologic_stage', 'pathology_report_file_name', 'patient_id', 'person_neoplasm_cancer_status', 'platelet_result_count', 'platelet_result_lower_limit', 'platelet_result_upper_limit', 'post_op_ablation_embolization_tx', 'postoperative_rx_tx', 'prothrombin_time_result_value', 'radiation_therapy', 'relative_family_cancer_history', 'residual_disease_post_new_tumor_event_margin_status', 'residual_tumor', 'sample_type', 'sample_type_id', 'specimen_collection_method_name', 'system_version', 'tissue_prospective_collection_indicator', 'tissue_retrospective_collection_indicator', 'tissue_source_site', 'total_bilirubin_upper_limit', 'tumor_tissue_site', 'vascular_tumor_cell_type', 'vial_number', 'viral_hepatitis_serology', 'vital_status', 'weight', 'year_of_initial_pathologic_diagnosis', '_GENOMIC_ID_TCGA_LIHC_gistic2', '_GENOMIC_ID_TCGA_LIHC_gistic2thd', '_GENOMIC_ID_TCGA_LIHC_mutation_bcm_gene', '_GENOMIC_ID_TCGA_LIHC_miRNA_HiSeq', '_GENOMIC_ID_TCGA_LIHC_PDMRNAseq', '_GENOMIC_ID_TCGA_LIHC_RPPA', '_GENOMIC_ID_TCGA_LIHC_exp_HiSeqV2_percentile', '_GENOMIC_ID_TCGA_LIHC_mutation_bcgsc_gene', '_GENOMIC_ID_data/public/TCGA/LIHC/miRNA_HiSeq_gene', '_GENOMIC_ID_TCGA_LIHC_PDMRNAseqCNV', '_GENOMIC_ID_TCGA_LIHC_exp_HiSeqV2_exon', '_GENOMIC_ID_TCGA_LIHC_mutation_ucsc_maf_gene', '_GENOMIC_ID_TCGA_LIHC_exp_HiSeqV2', '_GENOMIC_ID_TCGA_LIHC_exp_HiSeqV2_PANCAN', '_GENOMIC_ID_TCGA_LIHC_mutation_broad_gene', '_GENOMIC_ID_TCGA_LIHC_hMethyl450']
+
+# Identify candidate columns
+age_candidates = []
+gender_candidates = []
+for col in available_cols:
+    cl = col.lower()
+    if ('age' in cl and 'stage' not in cl) or ('birth' in cl):
+        age_candidates.append(col)
+    if ('gender' in cl) or (cl == 'sex') or ('sex_' in cl) or (cl.endswith('_sex')):
+        gender_candidates.append(col)
+
+candidate_age_cols = age_candidates
+candidate_gender_cols = gender_candidates
+
+# Print in the strictly required format
+print(f"candidate_age_cols = {candidate_age_cols}")
+print(f"candidate_gender_cols = {candidate_gender_cols}")
+
+# Try to load clinical data to preview candidate columns (if available)
+clinical_df = None
+try:
+    # Find a cohort dir likely corresponding to LIHC
+    cohort_dirs = [os.path.join(tcga_root_dir, d) for d in os.listdir(tcga_root_dir)
+                   if os.path.isdir(os.path.join(tcga_root_dir, d))]
+    preferred = [d for d in cohort_dirs if 'lihc' in os.path.basename(d).lower() or 'liver' in os.path.basename(d).lower()]
+    cohort_dir = preferred[0] if preferred else (cohort_dirs[0] if cohort_dirs else None)
+
+    if cohort_dir:
+        clinical_file_path, _ = tcga_get_relevant_filepaths(cohort_dir)
+        # Clinical matrix files from Xena are tab-delimited with sample IDs as the first column
+        clinical_df = pd.read_csv(clinical_file_path, sep='\t', index_col=0, dtype=str)
+except Exception:
+    clinical_df = None
+
+# Prepare previews if clinical data is available
+age_preview = {}
+gender_preview = {}
+if clinical_df is not None:
+    age_cols_present = [c for c in candidate_age_cols if c in clinical_df.columns]
+    gender_cols_present = [c for c in candidate_gender_cols if c in clinical_df.columns]
+
+    if len(age_cols_present) > 0:
+        age_preview = preview_df(clinical_df[age_cols_present], n=5)
+    if len(gender_cols_present) > 0:
+        gender_preview = preview_df(clinical_df[gender_cols_present], n=5)
+
+# Display previews as Python dictionaries
+print(age_preview)
+print(gender_preview)
+
+# Step 3: Select Demographic Features
+import re
+import pandas as pd
+import numpy as np
+
+# Helper to safely convert to int if possible
+def _to_int(x):
+    if pd.isna(x):
+        return None
+    if isinstance(x, (int, np.integer)):
+        return int(x)
+    if isinstance(x, float):
+        if np.isnan(x):
+            return None
+        return int(x)
+    m = re.search(r'-?\d+', str(x))
+    return int(m.group()) if m else None
+
+def _find_value_dict(candidate_cols):
+    # Try to find a dict in globals whose keys overlap with candidate columns and values are lists (preview samples)
+    for name, val in globals().items():
+        if isinstance(val, dict):
+            keys = set(val.keys())
+            overlap = [k for k in candidate_cols if k in keys]
+            if overlap and all(isinstance(val[k], list) for k in overlap):
+                return val
+    return None
+
+# Ensure candidate lists exist
+candidate_age_cols = globals().get('candidate_age_cols', [])
+candidate_gender_cols = globals().get('candidate_gender_cols', [])
+
+age_col = None
+gender_col = None
+
+# Attempt to locate the preview dictionaries
+age_values_dict = _find_value_dict(candidate_age_cols) if candidate_age_cols else None
+gender_values_dict = _find_value_dict(candidate_gender_cols) if candidate_gender_cols else None
+
+# Select age column
+if age_values_dict:
+    best = None
+    for col in candidate_age_cols:
+        if col not in age_values_dict:
+            continue
+        vals = age_values_dict[col]
+        non_missing = [v for v in vals if not pd.isna(v)]
+        if len(vals) == 0:
+            continue
+        ints = [_to_int(v) for v in non_missing]
+        ints = [v for v in ints if v is not None]
+        if not ints:
+            continue
+        age_like_frac = sum(0 <= v <= 120 for v in ints) / len(ints)
+        dob_like_frac = sum((v is not None) and (v <= -365) for v in ints) / len(ints)
+        missing_frac = 1 - (len(non_missing) / len(vals))
+        # Eligibility threshold: at least 60% look valid and missingness <= 60%
+        score = None
+        kind = None
+        if age_like_frac >= 0.6 and missing_frac <= 0.6:
+            score = (1, age_like_frac, -missing_frac, len(ints))
+            kind = 'age'
+        elif dob_like_frac >= 0.6 and missing_frac <= 0.6:
+            score = (0, dob_like_frac, -missing_frac, len(ints))
+            kind = 'dob'
+        if score is not None:
+            # Prefer age-like (score[0]=1) over dob-like (score[0]=0), then higher fraction, lower missing, more ints
+            if best is None or score > best[0]:
+                best = (score, col, kind)
+            # Prefer explicit age_at_initial_pathologic_diagnosis if tie
+            elif score == best[0] and col == 'age_at_initial_pathologic_diagnosis':
+                best = (score, col, kind)
+    if best:
+        age_col = best[1]
+    else:
+        # Fallback heuristic
+        if 'age_at_initial_pathologic_diagnosis' in candidate_age_cols:
+            age_col = 'age_at_initial_pathologic_diagnosis'
+        elif candidate_age_cols:
+            age_col = candidate_age_cols[0]
+else:
+    # No preview dict found; use domain heuristic
+    if 'age_at_initial_pathologic_diagnosis' in candidate_age_cols:
+        age_col = 'age_at_initial_pathologic_diagnosis'
+    elif candidate_age_cols:
+        age_col = candidate_age_cols[0]
+    else:
+        age_col = None
+
+# Select gender column
+if gender_values_dict:
+    best = None
+    for col in candidate_gender_cols:
+        if col not in gender_values_dict:
+            continue
+        vals = gender_values_dict[col]
+        non_missing = [v for v in vals if not pd.isna(v)]
+        if len(vals) == 0:
+            continue
+        normalized = [str(v).strip().lower() for v in non_missing]
+        mf_frac = sum(v in ('male', 'female') for v in normalized) / len(normalized)
+        missing_frac = 1 - (len(non_missing) / len(vals))
+        if mf_frac >= 0.6 and missing_frac <= 0.6:
+            score = (mf_frac, -missing_frac, len(non_missing))
+            if best is None or score > best[0]:
+                best = (score, col)
+            elif score == best[0] and col == 'gender':
+                best = (score, col)
+    if best:
+        gender_col = best[1]
+    else:
+        if 'gender' in candidate_gender_cols:
+            gender_col = 'gender'
+        elif candidate_gender_cols:
+            gender_col = candidate_gender_cols[0]
+        else:
+            gender_col = None
+else:
+    if 'gender' in candidate_gender_cols:
+        gender_col = 'gender'
+    elif candidate_gender_cols:
+        gender_col = candidate_gender_cols[0]
+    else:
+        gender_col = None
+
+# Explicitly print out the chosen columns and their preview values if available
+print("Selected age_col:", age_col)
+if age_col and age_values_dict and age_col in age_values_dict:
+    print("age_col preview values:", age_values_dict[age_col])
+
+print("Selected gender_col:", gender_col)
+if gender_col and gender_values_dict and gender_col in gender_values_dict:
+    print("gender_col preview values:", gender_values_dict[gender_col])
+
+# Step 4: Feature Engineering and Validation
+import os
+import pandas as pd
+
+# Ensure clinical and genetic data are loaded (reuse from previous steps if available; otherwise load)
+if 'clinical_df' not in globals() or 'genetic_df' not in globals():
+    # Locate LIHC cohort directory
+    subdirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+    def score_dir(name: str) -> int:
+        lname = name.lower()
+        score = 0
+        if "liver_cancer_(lihc)" in lname:
+            score += 100
+        if "lihc" in lname:
+            score += 50
+        if "liver" in lname:
+            score += 10
+        return score
+    scored = sorted([(score_dir(d), d) for d in subdirs], reverse=True)
+    if not scored or scored[0][0] <= 0:
+        # No suitable directory; record and exit gracefully for downstream steps
+        validate_and_save_cohort_info(
+            is_final=False,
+            cohort="TCGA_LIHC",
+            info_path=json_path,
+            is_gene_available=False,
+            is_trait_available=False
+        )
+        # Create empty placeholders to avoid NameError in following code blocks
+        clinical_df = pd.DataFrame()
+        genetic_df = pd.DataFrame()
+    else:
+        selected_dir = scored[0][1]
+        cohort_dir = os.path.join(tcga_root_dir, selected_dir)
+        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, low_memory=False, dtype=str)
+        genetic_df = pd.read_csv(genetic_file_path, sep='\t', index_col=0, low_memory=False)
+
+# 1) Extract and standardize clinical features (trait + optional age and gender)
+selected_clinical_df = tcga_select_clinical_features(
+    clinical_df=clinical_df,
+    trait=trait,
+    age_col=globals().get('age_col', None),
+    gender_col=globals().get('gender_col', None)
+)
+
+# 2) Normalize gene symbols and save normalized gene data
+# Subset gene expression to samples present in our clinical matrix
+common_sample_cols = [c for c in genetic_df.columns if c in selected_clinical_df.index]
+genetic_df_subset = genetic_df.loc[:, common_sample_cols].copy()
+
+# Coerce to numeric
+genetic_df_subset = genetic_df_subset.apply(pd.to_numeric, errors='coerce')
+
+# Normalize gene symbols using NCBI synonym dictionary and aggregate duplicates by mean
+normalized_gene_df = normalize_gene_symbols_in_index(genetic_df_subset)
+
+# Save normalized gene data
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_df.to_csv(out_gene_data_file)
+
+# 3) Link clinical and genetic data on sample IDs
+common_samples = selected_clinical_df.index.intersection(normalized_gene_df.columns)
+gene_data_T = normalized_gene_df.T.loc[common_samples]
+clinical_aligned = selected_clinical_df.loc[common_samples]
+linked_data = clinical_aligned.join(gene_data_T, how='inner')
+
+# 4) Handle missing values systematically
+processed_df = handle_missing_values(linked_data, trait_col=trait)
+
+# 5) Determine bias in trait and remove biased demographic features (if any)
+trait_biased, processed_df = judge_and_remove_biased_features(processed_df, trait)
+
+# 6) Final validation and save cohort info
+# Cast flags explicitly to native Python bool to avoid any np.bool_ leakage
+is_gene_available = bool(normalized_gene_df.shape[0] > 0)
+is_trait_available = bool((trait in linked_data.columns) and linked_data[trait].notna().any())
+trait_biased_flag = bool(trait_biased)
+
+note = str(
+    "INFO: Trait derived from TCGA sample type codes (01-09 tumor=1, 10-19 normal=0). "
+    "Gene expression subset to LIHC samples and gene symbols normalized via NCBI synonyms. "
+    "Missing values handled per pipeline (gene>20% NA removed; sample>5% NA removed; "
+    "imputation by mean/mode)."
+)
+
+is_usable = validate_and_save_cohort_info(
+    is_final=True,
+    cohort="TCGA_LIHC",
+    info_path=json_path,
+    is_gene_available=is_gene_available,
+    is_trait_available=is_trait_available,
+    is_biased=trait_biased_flag,
+    df=processed_df,
+    note=note
+)
+
+# 7) Save linked dataset only if usable
+if is_usable:
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    processed_df.to_csv(out_data_file)

output/preprocess/Liver_Cancer/cohort_info.json

@@ -1,122 +1 @@
-{
-  "GSE66843": {
-    "is_usable": false,
-    "is_gene_available": true,
-    "is_trait_available": true,
-    "is_available": true,
-    "is_biased": true,
-    "has_age": false,
-    "has_gender": false,
-    "sample_size": 17
-  },
-  "GSE45032": {
-    "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": 48
-  },
-  "GSE228783": {
-    "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": 152
-  },
-  "GSE228782": {
-    "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": 83
-  },
-  "GSE218438": {
-    "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
-  },
-  "GSE212047": {
-    "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
-  },
-  "GSE209875": {
-    "is_usable": false,
-    "is_gene_available": false,
-    "is_trait_available": true,
-    "is_available": false,
-    "is_biased": null,
-    "has_age": null,
-    "has_gender": null,
-    "sample_size": null
-  },
-  "GSE178201": {
-    "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
-  },
-  "GSE174570": {
-    "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
-  },
-  "GSE164760": {
-    "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": 170
-  },
-  "GSE148346": {
-    "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": 129
-  },
-  "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": 423
-  }
-}
+{"GSE66843": {"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/clinical data not available for this cohort; linking skipped. Only gene data saved."}, "GSE45032": {"is_usable": true, "is_gene_available": true, "is_trait_available": true, "is_available": true, "is_biased": false, "has_age": true, "has_gender": false, "sample_size": 48, "note": "INFO: Probe->gene mapping used annotation columns ID->GeneName; split multi-gene probe signal equally and summed per gene; normalized symbols using NCBI synonym map; dropped genes with >20% missing and samples with >5% missing genes; imputed Gender with mode and others with mean."}, "GSE228783": {"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": 140, "note": "INFO: Affymetrix HG-U219 probes mapped to symbols via platform annotation; no age/gender fields provided."}, "GSE228782": {"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": 83, "note": "INFO: Gene mapping performed from Affymetrix HG-U219 probes to symbols; clinical trait derived from 'disease' field; no age/gender available."}, "GSE218438": {"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": 636, "note": "WARNING: Probe-to-gene symbol mapping was unavailable; normalization produced empty gene data. Proceeding with probe-level Affymetrix IDs as features. INFO: Gene feature count=22268, sample count=636."}, "GSE212047": {"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}, "GSE209875": {"is_usable": false, "is_gene_available": false, "is_trait_available": true, "is_available": false, "is_biased": null, "has_age": null, "has_gender": null, "sample_size": null, "note": null}, "GSE178201": {"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": 3984, "note": "INFO: Trait is constant (all HepG2) in this matrix; dataset marked biased."}, "GSE174570": {"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": 114, "note": "INFO: Trait inferred from tissue labels (tumour=1, non-tumour adjacent=0); age and gender unavailable in this series. Platform HG-U219; probe-to-gene mapping and gene symbol normalization applied."}, "GSE164760": {"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": 170, "note": "INFO: Trait inferred from tissue; age and gender not available in matrix."}, "GSE148346": {"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_LIHC": {"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": 423, "note": "INFO: Trait derived from TCGA sample type codes (01-09 tumor=1, 10-19 normal=0). Gene expression subset to LIHC samples and gene symbols normalized via NCBI synonyms. Missing values handled per pipeline (gene>20% NA removed; sample>5% NA removed; imputation by mean/mode)."}}

output/preprocess/Liver_cirrhosis/clinical_data/GSE139602.csv

@@ -1,2 +1,2 @@
-0
-0.0
+,GSM4144550,GSM4144551,GSM4144552,GSM4144553,GSM4144554,GSM4144555,GSM4144556,GSM4144557,GSM4144558,GSM4144559,GSM4144560,GSM4144561,GSM4144562,GSM4144563,GSM4144564,GSM4144565,GSM4144566,GSM4144567,GSM4144568,GSM4144569,GSM4144570,GSM4144571,GSM4144572,GSM4144573,GSM4144574,GSM4144575,GSM4144576,GSM4144577,GSM4144578,GSM4144579,GSM4144580,GSM4144581,GSM4144582,GSM4144583,GSM4144584,GSM4144585,GSM4144586,GSM4144587,GSM4144588
+Liver_cirrhosis,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0

output/preprocess/Liver_cirrhosis/clinical_data/GSE285291.csv

@@ -0,0 +1,2 @@
+,GSM8700031,GSM8700032,GSM8700033,GSM8700034,GSM8700035,GSM8700036,GSM8700037,GSM8700038,GSM8700039,GSM8700040,GSM8700041,GSM8700042,GSM8700043,GSM8700044,GSM8700045,GSM8700046,GSM8700047,GSM8700048,GSM8700049,GSM8700050,GSM8700051,GSM8700052,GSM8700053,GSM8700054,GSM8700055,GSM8700056,GSM8700057,GSM8700058,GSM8700059,GSM8700060,GSM8700061,GSM8700062,GSM8700063,GSM8700064,GSM8700065,GSM8700066,GSM8700067,GSM8700068,GSM8700069,GSM8700070,GSM8700071,GSM8700072,GSM8700073,GSM8700074,GSM8700075,GSM8700076,GSM8700077,GSM8700078,GSM8700079,GSM8700080,GSM8700081,GSM8700082,GSM8700083
+Liver_cirrhosis,1.0,1.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0,1.0

output/preprocess/Liver_cirrhosis/code/GSE139602.py

@@ -0,0 +1,199 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE139602"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE139602"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE139602.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE139602.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE139602.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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 gene expression data availability
+is_gene_available = True  # Transcriptome analysis on liver biopsies indicates gene expression data
+
+# Step 2: Variable availability and converters
+
+# 2.1 Availability
+trait_row = 0  # disease state
+age_row = None
+gender_row = None
+
+# 2.2 Converters
+def _extract_value(x):
+    if x is None:
+        return None
+    s = str(x).strip()
+    if not s or s.lower() in {"na", "n/a", "none", "nan"}:
+        return None
+    if ":" in s:
+        s = s.split(":", 1)[1].strip()
+    return s if s else None
+
+def convert_trait(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    low = val.lower()
+    # Positive for cirrhosis: compensated/decompensated cirrhosis, ACLF (typically arises in cirrhosis)
+    if "cirrhosis" in low:
+        return 1
+    if "acute-on-chronic liver failure" in low or "aclf" in low:
+        return 1
+    # Negative for cirrhosis: healthy, early CLD/fibrosis (eCLD)
+    if low in {"healthy", "control"}:
+        return 0
+    if low in {"ecld", "fibrosis"} or "fibrosis" in low:
+        return 0
+    return None
+
+def convert_age(x):
+    # Not available in this dataset
+    return None
+
+def convert_gender(x):
+    # Not available in this dataset
+    return None
+
+# Step 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
+)
+
+# Step 4: Clinical feature extraction (only if clinical 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
+    )
+    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
+# Identify appropriate columns for mapping: probe ID ('ID') and gene symbol ('Gene Symbol')
+prob_col = 'ID'
+gene_col = 'Gene Symbol'
+
+# Build mapping dataframe from annotation
+mapping_df = get_gene_mapping(gene_annotation, prob_col=prob_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
+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) Ensure clinical data is available (reload if necessary) and link clinical/genetic data
+if 'selected_clinical_df' not in locals():
+    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 checking 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 metadata
+#    Coerce to native Python bool to avoid numpy.bool_ serialization issues.
+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 bool(linked_data[trait].notna().any()))
+is_trait_biased = bool(is_trait_biased)
+
+note = (
+    "INFO: Trait derived from 'disease state' (cirrhosis/ACLF=1; healthy/eCLD/fibrosis=0). "
+    "Age and Gender not available in this series."
+)
+
+try:
+    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
+    )
+except TypeError:
+    # Fallback in case of JSON serialization issues from a corrupted file or non-serializable entries
+    if os.path.exists(json_path):
+        os.remove(json_path)
+    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/Liver_cirrhosis/code/GSE150734.py

@@ -0,0 +1,221 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE150734"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE150734"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE150734.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE150734.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE150734.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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
+is_gene_available = True  # Gene expression profiling of liver biopsy tissues (not miRNA-only or methylation-only)
+
+# 2) Variable availability and conversion functions
+
+# Based on the sample characteristics dictionary:
+# 0: ['fibrosis stage: 0', 'fibrosis stage: 1'] -> No cirrhosis present (cirrhosis is fibrosis stage 4). Constant "no" for Liver_cirrhosis.
+# 1: ['pls risk prediction: Intermediate', 'Low', 'High'] -> Not the target trait; also no age/gender keys provided.
+trait_row = None
+age_row = None
+gender_row = None
+
+def _after_colon(val):
+    if val is None:
+        return None
+    s = str(val)
+    parts = s.split(":", 1)
+    return parts[1].strip() if len(parts) > 1 else s.strip()
+
+def convert_trait(val):
+    """
+    Convert fibrosis-related or cirrhosis-indicative strings to binary: cirrhosis -> 1, non-cirrhosis -> 0.
+    Unknown -> None.
+    This function is robust but will not be used here since trait_row is None (no informative variation).
+    """
+    v = _after_colon(val)
+    if v is None or v == "":
+        return None
+    s = v.strip().lower()
+
+    # Direct cirrhosis indications
+    if s in {"cirrhosis", "yes", "y"}:
+        return 1
+    if s in {"no", "n", "non-cirrhosis", "non cirrhosis"}:
+        return 0
+
+    # Handle fibrosis stage patterns (e.g., "0", "1", "F3", "F4", "stage 4")
+    # Extract first number present
+    m = re.search(r'([fF]?\s*([0-9]+))', s)
+    if m:
+        try:
+            num = int(re.search(r'([0-9]+)', m.group(0)).group(1))
+            return 1 if num >= 4 else 0
+        except Exception:
+            pass
+
+    # Heuristic keywords
+    if "f4" in s or "stage 4" in s or "advanced cirrhosis" in s:
+        return 1
+    if "f0" in s or "stage 0" in s:
+        return 0
+
+    return None
+
+def convert_age(val):
+    """
+    Extract numeric age from strings like 'age: 54', 'Age: 54 years'.
+    Returns float; unknown -> None.
+    """
+    v = _after_colon(val)
+    if v is None or v == "":
+        return None
+    m = re.search(r'(\d+(\.\d+)*)', v)
+    if m:
+        try:
+            return float(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def convert_gender(val):
+    """
+    Convert gender to binary: female -> 0, male -> 1. Unknown -> None.
+    """
+    v = _after_colon(val)
+    if v is None or v == "":
+        return None
+    s = v.strip().lower()
+    if s in {"female", "f", "woman", "women"}:
+        return 0
+    if s 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
+    )
+    _preview = preview_df(selected_clinical_df, n=5)
+    # 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)
+
+# 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., AARS, ABLIM1, ACOT2), these appear to be standard human gene symbols.
+requires_gene_mapping = False
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: 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. Guarded linking and validation depending on trait availability
+linked_data = None
+proceed_linking = (
+    'selected_clinical_data' in locals()
+    and isinstance(selected_clinical_data, pd.DataFrame)
+    and (trait in selected_clinical_data.index)
+)
+
+if proceed_linking:
+    # 2. Link clinical and genetic data
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
+
+    # 3. Handle missing values
+    linked_data = handle_missing_values(linked_data, trait)
+
+    # 4. Bias check 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: Proceeded with full linking, missing-value handling, and bias checks."
+    )
+
+    # 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 not available in this cohort (as determined in Step 2); skip linking and mark as unavailable for analysis.
+    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,
+        note="INFO: Trait not available for this cohort. Sample characteristics show fibrosis stage 0-1 only; no cirrhosis label present."
+    )
+    # Do not save out_data_file when not usable

output/preprocess/Liver_cirrhosis/code/GSE163211.py

@@ -0,0 +1,139 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE163211"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE163211"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE163211.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE163211.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE163211.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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
+# Nanostring nCounter assay measuring 800 genes => gene expression data available
+is_gene_available = True
+
+# 2) Variable availability and converters
+
+# Sample Characteristics Dictionary assessment (from previous step):
+# 8: 'nafld stage: Steatosis', 'NASH_F1_F4', 'Normal', 'NASH_F0'
+# Trait is Liver_cirrhosis; cirrhosis corresponds to fibrosis stage F4.
+# Here, fibrosis is aggregated as NASH_F1_F4, so we cannot isolate F4 specifically.
+trait_row = None  # Not derivable from provided categories
+age_row = 3       # 'age: <number>'
+gender_row = 4    # 'Sex: Female' / 'Sex: Male'
+
+def _extract_value(x):
+    if x is None:
+        return None
+    if isinstance(x, str):
+        parts = x.split(":", 1)
+        val = parts[1] if len(parts) > 1 else parts[0]
+        return val.strip()
+    return x
+
+def convert_trait(x):
+    # Cirrhosis status is not determinable from available categories ('NASH_F1_F4' lumps F1-F4)
+    # Return None to mark as unavailable/unknown
+    return None
+
+def convert_age(x):
+    val = _extract_value(x)
+    if val is None or val == "" or str(val).lower() in {"na", "nan", "none", "unknown"}:
+        return None
+    try:
+        v = float(val)
+        # Sanity check for human adult ages
+        if v <= 0 or v > 120:
+            return None
+        return v
+    except Exception:
+        return None
+
+def convert_gender(x):
+    val = _extract_value(x)
+    if val is None:
+        return None
+    v = str(val).strip().lower()
+    if v in {"female", "f", "woman", "girl"}:
+        return 0
+    if v in {"male", "m", "man", "boy"}:
+        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
+    )
+    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)
+else:
+    print({
+        "is_gene_available": is_gene_available,
+        "trait_row": trait_row,
+        "age_row": age_row,
+        "gender_row": gender_row
+    })
+
+# 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
+# The provided identifiers (e.g., A1BG, A2M, ABCG5) are standard human gene symbols (HGNC).
+requires_gene_mapping = False
+print(f"requires_gene_mapping = {requires_gene_mapping}")

output/preprocess/Liver_cirrhosis/code/GSE182060.py

@@ -0,0 +1,148 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE182060"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE182060"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE182060.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE182060.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE182060.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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 data availability
+is_gene_available = True  # Gene expression profiling of liver biopsy tissues
+
+# 2) Variable availability based on provided Sample Characteristics Dictionary
+# Keys present: 0: patient ID, 1: tissue (constant), 2: time_point (Baseline/Follow-up)
+trait_row = None   # Liver cirrhosis status not provided or inferable
+age_row = None     # No age field
+gender_row = None  # No gender field
+
+# 2.2) Converters
+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]
+    return val.strip()
+
+def convert_trait(x):
+    """
+    Convert trait to binary: Liver cirrhosis -> 1, non-cirrhosis -> 0.
+    Only map when explicit; otherwise return None.
+    """
+    val = _after_colon(x)
+    if val is None or val == "":
+        return None
+    s = val.strip().lower()
+
+    # Positive indicators
+    pos_terms = [
+        "cirrhosis", "liver cirrhosis", "cirrhotic", "cirrhose", "hepatic cirrhosis",
+        "lc"
+    ]
+    if any(t in s for t in pos_terms):
+        return 1
+    if "f4" in s:  # Metavir/Brunt F4 indicates cirrhosis
+        return 1
+
+    # Negative indicators
+    neg_terms = [
+        "no cirrhosis", "non-cirrhosis", "non cirrhosis", "control", "healthy", "normal"
+    ]
+    if any(t in s for t in neg_terms):
+        return 0
+    # Fibrosis stages F0-F3 imply non-cirrhosis
+    if re.search(r"\bf[0-3]\b", s):
+        return 0
+
+    # If dataset disease is NAFLD/NASH without explicit cirrhosis info, do not force 0.
+    return None
+
+def convert_age(x):
+    """Convert age to continuous (float years)."""
+    val = _after_colon(x)
+    if val is None:
+        return None
+    s = str(val).strip().lower()
+    # Extract first number (integer or float)
+    m = re.search(r"[-+]?\d*\.?\d+", s)
+    if not m:
+        return None
+    try:
+        return float(m.group())
+    except Exception:
+        return None
+
+def convert_gender(x):
+    """Convert gender to binary: female->0, male->1."""
+    val = _after_colon(x)
+    if val is None:
+        return None
+    s = str(val).strip().lower()
+    if s in {"f", "female", "woman", "women"}:
+        return 0
+    if s in {"m", "male", "man", "men"}:
+        return 1
+    # Handle phrases like "sex: Female"
+    if "female" in s:
+        return 0
+    if "male" in s:
+        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 since trait_row is None)
+# If in another dataset trait_row is not None, the following would be used:
+# 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)

output/preprocess/Liver_cirrhosis/code/GSE182065.py

@@ -0,0 +1,176 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE182065"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE182065"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE182065.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE182065.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE182065.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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) Determine gene expression availability based on provided background info
+is_gene_available = True  # Gene expression profiling of liver tissues (not miRNA/methylation)
+
+# 2) Variable availability assessment from Sample Characteristics Dictionary
+# Observed keys:
+# 0: tissue (constant: Liver)
+# 1: sample group (treatment/control/baseline - experimental condition, not trait/age/gender)
+# 2: compound (treatments)
+# 3: concentration (treatment doses)
+trait_row = None   # No cirrhosis status available
+age_row = None     # No age information available
+gender_row = None  # No gender information available
+
+# 2.2) Converters
+def _after_colon(x: str) -> str:
+    if x is None:
+        return ""
+    parts = str(x).split(":", 1)
+    return parts[1].strip() if len(parts) == 2 else str(x).strip()
+
+def convert_trait(x):
+    s = _after_colon(x).strip().lower()
+    if not s or s in {"na", "n/a", "nan", "none"}:
+        return None
+
+    # Direct cirrhosis indications
+    if "cirrhosis" in s or "cirrhotic" in s:
+        # Non-cirrhosis phrases
+        if any(neg in s for neg in ["non-cirrhosis", "no cirrhosis", "noncirrhotic", "non cirrhotic"]):
+            return 0
+        return 1
+
+    # F-stage mapping if present
+    m = re.search(r'\bf\s*([0-4])\b', s, flags=re.IGNORECASE)
+    if m:
+        stage = int(m.group(1))
+        return 1 if stage == 4 else 0
+
+    # Common control/healthy indicators
+    if any(k in s for k in ["healthy", "control", "normal liver", "non-diseased", "non diseased"]):
+        return 0
+
+    # Abbreviation 'LC' for liver cirrhosis as a standalone token
+    tokens = re.findall(r'\b[a-zA-Z]+\b', s)
+    if "lc" in tokens:
+        return 1
+
+    return None
+
+def convert_age(x):
+    s = _after_colon(x).lower()
+    if not s or s in {"na", "n/a", "nan", "none"}:
+        return None
+    nums = re.findall(r'\d+\.?\d*', s)
+    if not nums:
+        return None
+    try:
+        return float(nums[0])
+    except Exception:
+        return None
+
+def convert_gender(x):
+    s = _after_colon(x).strip().lower()
+    if not s or s in {"na", "n/a", "nan", "none"}:
+        return None
+    if any(k == s or k in s.split() for k in ["female", "f", "woman", "girl", "women", "ladies"]):
+        return 0
+    if any(k == s or k in s.split() for k in ["male", "m", "man", "boy", "men", "gentleman"]):
+        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 (skip since trait_row is None)
+# If in future steps trait_row becomes available, the following snippet can be used:
+# 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 = preview_df(selected_df)
+#     os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+#     selected_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 = False
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Data Normalization and Linking
+import os
+
+# 1. Normalize the obtained gene data and save it
+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 data is unavailable (from Step 2), skip linking and downstream steps.
+is_gene_available = True
+is_trait_available = False
+
+# 5. Conduct final quality validation and save cohort information.
+note = ("INFO: Trait data unavailable in this GEO series; no age or gender annotations either. "
+        "Only normalized gene expression data were saved; clinical-genetic linking skipped.")
+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=False,                 # Placeholder; not applicable when trait is unavailable
+    df=normalized_gene_data.T,       # Provide a dataframe for validation; samples as rows
+    note=note
+)
+
+# 6. Do not save linked data since trait is unavailable (is_usable will be False).

output/preprocess/Liver_cirrhosis/code/GSE185529.py

@@ -0,0 +1,368 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE185529"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE185529"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE185529.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE185529.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE185529.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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
+# 1) Determine data availability based on provided background and sample characteristics
+# From the sample characteristics: only "treatment: siCTRL/siBNC2" is available.
+# No human trait (Liver_cirrhosis), age, or gender information is present.
+is_gene_available = True  # Title suggests gene expression data (not miRNA/methylation)
+trait_row = None
+age_row = None
+gender_row = None
+
+# 2) Define conversion functions
+
+def _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 != "" else None
+    return None
+
+def convert_trait(x):
+    # Binary: 1 = Liver cirrhosis/case, 0 = control/non-cirrhosis. Unknown -> None
+    val = _after_colon(x)
+    if val is None:
+        return None
+    v = val.lower()
+    # Positive indicators
+    positives = ["cirrhosis", "liver cirrhosis", "cirrhotic", "lc"]
+    if any(p in v for p in positives):
+        return 1
+    # Common fibrosis stage mapping (heuristic): F4 -> cirrhosis
+    if "f4" in v or "fibrosis stage 4" in v or "ishak 5" in v or "ishak 6" in v:
+        return 1
+    # Negative indicators
+    negatives = ["control", "normal", "healthy", "non-cirrhotic", "no cirrhosis"]
+    if any(n in v for n in negatives):
+        return 0
+    # If it's treatment labels or unrelated fields, return None
+    if "sictrl" in v or "sibnc2" in v or "treatment" in v:
+        return None
+    return None
+
+def convert_age(x):
+    # Continuous: age in years as float. Unknown -> None
+    import re
+    val = _after_colon(x)
+    if val is None:
+        return None
+    m = re.search(r"(\d+(?:\.\d+)?)", val)
+    if not m:
+        return None
+    try:
+        age = float(m.group(1))
+        if 0 <= age <= 120:
+            return age
+    except Exception:
+        pass
+    return None
+
+def convert_gender(x):
+    # Binary: female=0, male=1. Unknown -> None
+    val = _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
+    if v in ["na", "n/a", "unknown", "not available", "not provided", ""]:
+        return 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 (skip because trait_row is None)
+# If trait_row were available:
+# selected_clinical_df = geo_select_clinical_features(
+#     clinical_df=clinical_data,  # assumed to be available from previous step
+#     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
+# Gene Identifier Mapping with species-agnostic symbol extraction and explicit overlap validation
+
+import re
+import pandas as pd
+
+ga = gene_annotation.copy()
+
+# Ensure string dtype
+ga = ga.applymap(lambda x: str(x) if pd.notnull(x) else x)
+
+# Candidate ID columns likely present in Affymetrix platform SOFT
+id_base_candidates = [
+    'probeset_id', 'transcript_cluster_id', 'ID', 'ID_REF',
+    'Probe Set ID', 'probe_id', 'probesetid', 'AFFYID'
+]
+id_base_candidates = [c for c in id_base_candidates if c in ga.columns]
+# Add other columns as fallbacks
+id_base_candidates += [c for c in ga.columns if c not in id_base_candidates]
+
+# Prepare expression index variants
+expr_variants = {
+    "direct": gene_data,
+    "strip_suffix": gene_data.rename(index=lambda x: str(x).split('_')[0] if isinstance(x, str) else x)
+}
+
+# Compute overlaps to pick best ID column and index variant
+best = ("direct", id_base_candidates[0] if id_base_candidates else None, 0)
+for expr_name, expr_df in expr_variants.items():
+    expr_ids = set(expr_df.index.astype(str))
+    for id_col in id_base_candidates:
+        annot_ids = set(ga[id_col].astype(str))
+        overlap = len(expr_ids & annot_ids)
+        if overlap > best[2]:
+            best = (expr_name, id_col, overlap)
+
+best_expr_name, best_id_col, best_overlap = best
+expr_df_best = expr_variants[best_expr_name]
+
+print(f"Chosen identifier column: {best_id_col}")
+print(f"Expression ID handling: {best_expr_name}")
+print(f"Identifier overlap count: {best_overlap}")
+
+if best_overlap == 0:
+    # As a last fallback, try adding common suffixes to annotation IDs and recompute against direct expr IDs
+    if 'probeset_id' in ga.columns:
+        for suf in ['_st', '_at']:
+            tmp_col = f"probeset_id{suf}"
+            ga[tmp_col] = ga['probeset_id'].astype(str) + suf
+            overlap = len(set(gene_data.index.astype(str)) & set(ga[tmp_col].astype(str)))
+            if overlap > best_overlap:
+                best_overlap = overlap
+                best_id_col = tmp_col
+                best_expr_name = 'direct'
+                expr_df_best = expr_variants[best_expr_name]
+    print(f"Fallback overlap after adding suffixes: {best_overlap}")
+
+if best_overlap == 0:
+    print("WARNING: No overlap between expression probe IDs and annotation IDs. Mapping may result in empty gene data.")
+
+# Choose gene symbol column
+symbol_candidates_preferred = [
+    'gene_symbol', 'Gene Symbol', 'GENE_SYMBOL', 'Symbol', 'SYMBOL', 'Gene symbol', 'Gene symbols'
+]
+if 'gene_assignment' in ga.columns:
+    gene_col = 'gene_assignment'
+else:
+    gene_col = None
+    for c in symbol_candidates_preferred:
+        if c in ga.columns:
+            gene_col = c
+            break
+    if gene_col is None:
+        # Any column with 'symbol' substring
+        for c in ga.columns:
+            if 'symbol' in c.lower():
+                gene_col = c
+                break
+# Final fallback to a verbose text column
+if gene_col is None:
+    avg_lens = {c: ga[c].astype(str).map(len).mean() for c in ga.columns}
+    gene_col = max(avg_lens, key=avg_lens.get)
+
+print(f"Chosen gene symbol source column: {gene_col}")
+
+def parse_gene_symbols_species_agnostic(text: str):
+    if text is None:
+        return []
+    s = str(text).strip()
+    if s == '' or s == '---':
+        return []
+    # Split entries separated by '///'
+    parts = re.split(r'\s*///\s*', s)
+    symbols = []
+    for part in parts:
+        tokens = re.split(r'\s*//\s*', part)
+        if len(tokens) >= 2:
+            sym = tokens[1].strip()
+            # Filter out placeholders
+            if sym and sym != '---':
+                symbols.append(sym)
+    # If no symbols were captured via the structured pattern, fallback:
+    if not symbols:
+        # Heuristic: capture capitalized words with letters/digits/-_. This is lenient and species-agnostic.
+        candidates = re.findall(r'\b[A-Za-z][A-Za-z0-9._-]{1,24}\b', s)
+        # Remove common database tokens
+        blacklist = {'RefSeq', 'ENSEMBL', 'GenBank', 'ENSEMBL', 'ENSEMBL', 'ENSEMBLE', 'chr', 'chromosome', 'cdna',
+                     'ncrna', 'transcript', 'gene', 'biotype', 'protein_coding', 'antisense', 'mRNA', 'mus', 'musculus'}
+        symbols = [c for c in candidates if c not in blacklist]
+    # Deduplicate preserving order
+    seen = set()
+    ordered = []
+    for sym in symbols:
+        if sym not in seen:
+            seen.add(sym)
+            ordered.append(sym)
+    return ordered
+
+# Build mapping DataFrame
+mapping_df = ga[[best_id_col, gene_col]].dropna().copy()
+mapping_df = mapping_df.rename(columns={best_id_col: 'ID', gene_col: 'Gene'})
+mapping_df['ID'] = mapping_df['ID'].astype(str)
+
+# Align mapping to expression IDs in the chosen variant
+expr_ids_set = set(expr_df_best.index.astype(str))
+mapping_df = mapping_df[mapping_df['ID'].isin(expr_ids_set)]
+
+# Parse symbols to lists
+mapping_df['Gene'] = mapping_df['Gene'].apply(parse_gene_symbols_species_agnostic)
+
+# Expand multi-gene mappings and distribute expression
+mapping_df['num_genes'] = mapping_df['Gene'].apply(lambda x: len(x))
+mapping_df = mapping_df.explode('Gene')
+mapping_df = mapping_df.dropna(subset=['Gene'])
+mapping_df = mapping_df[mapping_df['num_genes'] > 0]
+
+# Join with expression and distribute by num_genes
+mapping_df = mapping_df.set_index('ID')
+merged = mapping_df.join(expr_df_best, how='inner')
+expr_cols = [c for c in merged.columns if c not in ['Gene', 'num_genes']]
+if len(merged) == 0 or len(expr_cols) == 0:
+    print("WARNING: No mappings joined with expression data. Resulting gene_data may be empty.")
+else:
+    merged[expr_cols] = merged[expr_cols].div(merged['num_genes'], axis=0)
+
+# Sum to gene level
+gene_data = merged.groupby('Gene')[expr_cols].sum()
+
+print(f"Mapped probes to genes: {gene_data.shape[0]} genes, {gene_data.shape[1] if gene_data.shape[0]>0 else 0} samples.")
+
+# 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) Link clinical and genetic data only if clinical features exist
+try:
+    _ = selected_clinical_data  # probe existence without NameError
+    has_clinical = isinstance(selected_clinical_data, pd.DataFrame) and not selected_clinical_data.empty
+except NameError:
+    has_clinical = False
+
+if has_clinical:
+    # Link
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
+
+    # 3) Handle missing values
+    linked_data = handle_missing_values(linked_data, trait)
+
+    # 4) Judge 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_gene_available = (normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0)
+    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=True,
+        is_biased=is_trait_biased,
+        df=unbiased_linked_data,
+        note="INFO: Clinical features present and linked."
+    )
+
+    # 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:
+    # No clinical data available; cannot link
+    linked_data = pd.DataFrame()
+
+    # Final validation reflecting true state without fabricating data
+    df_for_validation = normalized_gene_data.T
+    is_gene_available = (normalized_gene_data.shape[0] > 0) and (normalized_gene_data.shape[1] > 0)
+
+    note = ("WARNING: No human trait/clinical data available; sample characteristics only show treatment labels "
+            "(siCTRL/siBNC2). Not suitable for trait analysis. Linked dataset not saved.")
+    _ = validate_and_save_cohort_info(
+        is_final=True,
+        cohort=cohort,
+        info_path=json_path,
+        is_gene_available=is_gene_available,
+        is_trait_available=False,
+        is_biased=False,
+        df=df_for_validation,
+        note=note
+    )

output/preprocess/Liver_cirrhosis/code/GSE212047.py

@@ -0,0 +1,136 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE212047"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE212047"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE212047.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE212047.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE212047.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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
+# This cohort appears to be mouse HSC samples with genetic/treatment manipulation; no human clinical variables present.
+
+# 1) Gene expression availability (RNA-seq/microarray indicated; not miRNA/methylation-only)
+is_gene_available = True
+
+# 2) Variable availability (no human trait/age/gender in the sample characteristics; all mouse-related)
+trait_row = None
+age_row = None
+gender_row = None
+
+# 2.2) Conversion functions
+
+def _extract_value(x):
+    if x is None:
+        return None
+    if isinstance(x, (int, float)):
+        return x
+    s = str(x).strip()
+    # Take text after the first colon if present
+    if ':' in s:
+        s = s.split(':', 1)[1].strip()
+    # Normalize common NAs
+    if s.lower() in {'na', 'n/a', 'none', 'null', ''}:
+        return None
+    return s
+
+def convert_trait(x):
+    # Binary: presence (1) vs absence (0) of Liver cirrhosis
+    v = _extract_value(x)
+    if v is None:
+        return None
+    val = str(v).lower()
+    # Positive (cirrhosis present)
+    positives = ['cirrhosis', 'cirrhotic', 'nt cirrhotic', 'liver cirrhosis', 'alcoholic cirrhosis', 'hcv cirrhosis', 'hbv cirrhosis']
+    # Negative (no cirrhosis)
+    negatives = ['normal', 'healthy', 'control', 'non-cirrhotic', 'no cirrhosis', 'fibrosis stage 0', 'non cirrhotic']
+    if any(p in val for p in positives):
+        return 1
+    if any(n in val for n in negatives):
+        return 0
+    # Heuristic: advanced fibrosis often implies cirrhosis depending on context, but avoid overcalling here
+    if 'fibrosis' in val and any(k in val for k in ['f4', 'stage 4']):
+        return 1
+    return None
+
+def convert_age(x):
+    # Continuous age in years
+    v = _extract_value(x)
+    if v is None:
+        return None
+    s = str(v).lower()
+    # Extract the first number in the string
+    import re
+    m = re.search(r'[-+]?\d*\.?\d+', s)
+    if not m:
+        return None
+    try:
+        age_val = float(m.group())
+    except Exception:
+        return None
+    # Convert from months if explicitly indicated
+    if 'month' in s:
+        age_val = age_val / 12.0
+    return age_val
+
+def convert_gender(x):
+    # Binary: female -> 0, male -> 1
+    v = _extract_value(x)
+    if v is None:
+        return None
+    s = str(v).strip().lower()
+    if s in {'f', 'female', 'woman', 'girl'}:
+        return 0
+    if s in {'m', 'male', 'man', 'boy'}:
+        return 1
+    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: skipped because trait_row is None (no human clinical data available)
+# If clinical data were available, we would use:
+# 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 = preview_df(selected_df)
+# os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+# selected_df.to_csv(out_clinical_data_file)

output/preprocess/Liver_cirrhosis/code/GSE285291.py

@@ -0,0 +1,178 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE285291"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE285291"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE285291.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE285291.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE285291.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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
+import numpy as np
+import re
+
+# 1) Gene expression data availability
+is_gene_available = True  # mRNA gene expression from intestinal biopsies (not miRNA/methylation)
+
+# 2) Variable availability and conversion functions
+
+# Trait: Liver cirrhosis presence inferred from "status": Control vs Compensated/Decompensated
+trait_row = 1  # 'status' field
+age_row = None  # Not provided; "age-matched" implies no per-sample ages
+gender_row = None  # All men per background => constant => not usable
+
+def _after_colon(value):
+    if value is None or (isinstance(value, float) and np.isnan(value)):
+        return None
+    s = str(value)
+    if ':' in s:
+        s = s.split(':', 1)[1]
+    return s.strip() if s is not None else None
+
+def convert_trait(value):
+    v = _after_colon(value)
+    if v is None:
+        return None
+    vlow = v.lower()
+    if vlow in {"compensated", "decompensated"}:
+        return 1  # cirrhosis present
+    if vlow in {"control", "healthy control", "healthy"}:
+        return 0  # no cirrhosis
+    return None
+
+def convert_age(value):
+    # Not available in this dataset; keep as None if ever called
+    v = _after_colon(value)
+    if v is None:
+        return None
+    # Try to extract a number if any
+    m = re.search(r'(\d+(\.\d+)?)', v)
+    if m:
+        try:
+            return float(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def convert_gender(value):
+    # Not available here, but implement standard mapping if ever called
+    v = _after_colon(value)
+    if v is None:
+        return None
+    vlow = v.lower()
+    if vlow in {"male", "m"}:
+        return 1
+    if vlow 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 (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=convert_age,
+        gender_row=gender_row,
+        convert_gender=convert_gender
+    )
+    preview = preview_df(selected_clinical_df)
+    print("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)
+
+# 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., A2M, AADAT, AANAT, ABCB7, ABL1), these are standard human gene symbols (HGNC).
+requires_gene_mapping = False
+print(f"requires_gene_mapping = {requires_gene_mapping}")
+
+# Step 5: Data Normalization and Linking
+import os
+import pandas as pd
+
+# 1. Normalize gene symbols and save
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# Ensure clinical features are available in-memory (fallback to disk if needed)
+if 'selected_clinical_df' not in locals():
+    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 in the linked data
+linked_data = handle_missing_values(linked_data, trait)
+
+# 4. Determine whether the trait and demographic features are severely biased; remove biased demographics
+is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+# 5. Final validation and save cohort information
+note = ("INFO: All subjects are male per series description; per-sample age not provided. "
+        "No gender/age covariates included. Gene IDs were already HGNC symbols; normalized via synonyms.")
+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 the 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/Liver_cirrhosis/code/GSE66843.py

@@ -0,0 +1,156 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE66843"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE66843"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE66843.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE66843.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE66843.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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
+# This is a cell-line SuperSeries (Huh7.5.1) with infection/time-post-infection metadata; no human clinical traits.
+is_gene_available = True  # Likely gene expression profiling (not pure miRNA/methylation based on context)
+trait_row = None          # Liver_cirrhosis not recorded in this cell-line dataset
+age_row = None            # No human age
+gender_row = None         # No human gender
+
+# Conversion functions (kept robust though not used since rows 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"} else None
+    return x
+
+def convert_trait(x):
+    # No Liver_cirrhosis information in this dataset
+    return None
+
+def convert_age(x):
+    val = _extract_after_colon(x)
+    if val is None:
+        return None
+    # Attempt to parse numeric age if ever encountered
+    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", "woman", "women"}:
+        return 0
+    if v in {"male", "m", "man", "men"}:
+        return 1
+    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_row is None
+
+# 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 appropriate columns in the annotation for probe IDs and gene symbols
+prob_col = 'ID'       # Matches ILMN_* probe IDs in gene_data
+gene_col = 'Symbol'   # Contains gene symbols (may include multiple/complex entries)
+
+# Build the mapping dataframe
+mapping_df = get_gene_mapping(gene_annotation, prob_col=prob_col, gene_col=gene_col)
+
+# 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
+# 1. Normalize gene symbols and save gene-level data
+normalized_gene_data = normalize_gene_symbols_in_index(gene_data)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# Guard: this cohort lacks human clinical trait/covariates (cell-line dataset in Step 2)
+has_selected_clinical = ('selected_clinical_data' in globals()) or ('selected_clinical_data' in locals())
+has_trait_row = ('trait_row' in globals()) or ('trait_row' in locals())
+trait_available = has_trait_row and (trait_row is not None)
+
+if trait_available and has_selected_clinical:
+    # 2-4. Link, handle missing values, and bias checks
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
+    linked_data = handle_missing_values(linked_data, trait)
+    is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
+
+    # 5. Final validation and save cohort info
+    note = "INFO: Trait and covariates available; linked data processed with QC."
+    is_usable = validate_and_save_cohort_info(
+        True, cohort, json_path, True, True, is_trait_biased, unbiased_linked_data, note
+    )
+
+    # 6. Save linked data if usable
+    if is_usable:
+        unbiased_linked_data.to_csv(out_data_file)
+else:
+    # No clinical trait available; skip linking and downstream trait-dependent processing.
+    linked_data = None
+    note = "INFO: Cell-line dataset without human clinical trait/covariates; only gene expression saved."
+    # Provide a non-empty placeholder df for final validation to avoid abnormality override
+    _ = validate_and_save_cohort_info(
+        True, cohort, json_path, True, False, False, normalized_gene_data.T, note
+    )

output/preprocess/Liver_cirrhosis/code/GSE85550.py

@@ -0,0 +1,220 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+cohort = "GSE85550"
+
+# Input paths
+in_trait_dir = "../DATA/GEO/Liver_cirrhosis"
+in_cohort_dir = "../DATA/GEO/Liver_cirrhosis/GSE85550"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/GSE85550.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/GSE85550.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/GSE85550.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/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  # Matrix file parsed; dataset likely contains gene expression (not miRNA-only or methylation-only)
+
+# 2) Variable availability based on Sample Characteristics Dictionary
+# Keys: 0 -> patient IDs (unique per sample), 1 -> tissue (constant), 2 -> time_point (Baseline/Follow-up)
+# No disease status (cirrhosis), age, or gender fields available.
+trait_row = None
+age_row = None
+gender_row = None
+
+# 2.2) Conversion functions
+def _extract_value(cell):
+    if cell is None:
+        return None
+    s = str(cell)
+    return s.split(":", 1)[1].strip() if ":" in s else s.strip()
+
+def convert_trait(cell):
+    v = _extract_value(cell)
+    if v is None:
+        return None
+    v_low = v.lower()
+
+    # Map explicit disease/control mentions
+    positive_terms = [
+        'cirrhosis', 'cirrhotic', 'liver cirrhosis'
+    ]
+    negative_terms = [
+        'control', 'healthy', 'normal', 'non-cirrhotic', 'non cirrhotic', 'no cirrhosis'
+    ]
+    if any(t in v_low for t in positive_terms):
+        return 1
+    if any(t in v_low for t in negative_terms):
+        return 0
+
+    # Time-point or unrelated fields should not be used as trait
+    if v_low in {'baseline', 'follow-up', 'follow up', 'liver biopsy', 'biopsy'}:
+        return None
+
+    return None
+
+def convert_age(cell):
+    v = _extract_value(cell)
+    if v is None:
+        return None
+    v_low = v.lower()
+    if v_low in {'na', 'n/a', 'unknown', ''}:
+        return None
+    # Extract a numeric age; accept integers or decimals
+    m = re.search(r'(\d+(\.\d+)?)', v_low)
+    if m:
+        try:
+            return float(m.group(1))
+        except Exception:
+            return None
+    return None
+
+def convert_gender(cell):
+    v = _extract_value(cell)
+    if v is None:
+        return None
+    v_low = v.lower()
+    # Standard mappings
+    if v_low in {'male', 'm'}:
+        return 1
+    if v_low in {'female', 'f'}:
+        return 0
+    # Unknowns
+    if v_low in {'na', 'n/a', 'unknown', ''}:
+        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 (skip because trait_row is None)
+# If clinical features were available, we would call:
+# 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)
+# 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 = False")
+
+# Step 5: 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)
+
+# Ensure output directory exists before saving
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+normalized_gene_data.to_csv(out_gene_data_file)
+
+# Robust detection of trait availability
+trait_row = None if 'trait_row' not in globals() else trait_row
+age_row = None if 'age_row' not in globals() else age_row
+gender_row = None if 'gender_row' not in globals() else gender_row
+is_trait_available = (trait_row is not None)
+
+if is_trait_available:
+    # 2. Extract clinical features
+    selected_clinical_data = 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
+    )
+    # Save clinical data
+    os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
+    selected_clinical_data.to_csv(out_clinical_data_file)
+
+    # 2. Link the clinical and genetic data
+    linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
+
+    # 3. Handle missing values
+    linked_data = handle_missing_values(linked_data, trait)
+
+    # 4. Bias checks 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 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: Linked clinical-genetic data processed with missing value handling and bias checks."
+    )
+
+    # 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 not available: skip linking and downstream steps; record final metadata
+    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,  # Not applicable without trait; placeholder
+        df=normalized_gene_data,
+        note=("INFO: Trait data not available in clinical annotations for this cohort "
+              "(only patient ID, tissue=liver biopsy, time_point). Gene expression saved; "
+              "linked data not generated.")
+    )
+    # Do not save out_data_file when trait is unavailable

output/preprocess/Liver_cirrhosis/code/TCGA.py

@@ -0,0 +1,202 @@
+# Path Configuration
+from tools.preprocess import *
+
+# Processing context
+trait = "Liver_cirrhosis"
+
+# Input paths
+tcga_root_dir = "../DATA/TCGA"
+
+# Output paths
+out_data_file = "./output/z4/preprocess/Liver_cirrhosis/TCGA.csv"
+out_gene_data_file = "./output/z4/preprocess/Liver_cirrhosis/gene_data/TCGA.csv"
+out_clinical_data_file = "./output/z4/preprocess/Liver_cirrhosis/clinical_data/TCGA.csv"
+json_path = "./output/z4/preprocess/Liver_cirrhosis/cohort_info.json"
+
+
+# Step 1: Initial Data Loading
+import os
+import pandas as pd
+
+# Identify the most relevant TCGA cohort directory for Liver_cirrhosis
+all_dirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+selected_dir = None
+# Prioritize explicit liver cancer cohort (LIHC), then general liver mentions
+priority_keys = ['lihc', 'liver']
+candidates = [d for d in all_dirs if any(k in d.lower() for k in priority_keys)]
+if candidates:
+    lihc = [d for d in candidates if 'lihc' in d.lower()]
+    selected_dir = lihc[0] if lihc else candidates[0]
+
+if selected_dir is None:
+    # No suitable cohort found; mark as unavailable and stop further loading
+    validate_and_save_cohort_info(
+        is_final=False,
+        cohort="TCGA",
+        info_path=json_path,
+        is_gene_available=False,
+        is_trait_available=False
+    )
+    print("No suitable TCGA cohort directory found for the trait. Skipping.")
+    clinical_df = pd.DataFrame()
+    genetic_df = pd.DataFrame()
+else:
+    cohort_dir = os.path.join(tcga_root_dir, selected_dir)
+    # Find clinical and genetic file paths
+    clinical_file_path, genetic_file_path = tcga_get_relevant_filepaths(cohort_dir)
+
+    # Load dataframes
+    clinical_df = pd.read_csv(clinical_file_path, sep='\t', index_col=0, compression='infer', low_memory=False)
+    genetic_df = pd.read_csv(genetic_file_path, sep='\t', index_col=0, compression='infer', low_memory=False)
+
+    # Print clinical column names for inspection
+    print(clinical_df.columns.tolist())
+
+# Step 2: Find Candidate Demographic Features
+import os
+import re
+import pandas as pd
+
+# Identify cohort directory (prefer LIHC for liver cancer)
+cohort_dirs = [d for d in os.listdir(tcga_root_dir) if os.path.isdir(os.path.join(tcga_root_dir, d))]
+preferred = [d for d in cohort_dirs if d.upper() == "LIHC"]
+if preferred:
+    cohort_dir = os.path.join(tcga_root_dir, preferred[0])
+else:
+    # fallback: pick first dir containing 'LIHC', else the first available dir
+    contains = [d for d in cohort_dirs if "LIHC" in d.upper()]
+    cohort_dir = os.path.join(tcga_root_dir, (contains[0] if contains else cohort_dirs[0]))
+
+clinical_file_path, _ = tcga_get_relevant_filepaths(cohort_dir)
+
+# Load clinical data (TCGA Xena clinicalMatrix is tab-delimited with samples as index)
+clinical_df = pd.read_csv(clinical_file_path, sep='\t', header=0, index_col=0)
+
+# Find candidate demographic columns with token-level matching to avoid false positives (e.g., "stage")
+def tokenize(col_name: str):
+    return [tok for tok in re.split(r'[^a-z]+', col_name.lower()) if tok]
+
+tokens_map = {col: tokenize(col) for col in clinical_df.columns}
+
+candidate_age_cols = []
+candidate_gender_cols = []
+
+for col, toks in tokens_map.items():
+    if ('age' in toks) or ('birth' in toks):
+        candidate_age_cols.append(col)
+    if ('gender' in toks) or ('sex' in toks):
+        candidate_gender_cols.append(col)
+
+# Deduplicate while preserving order
+def dedup(seq):
+    seen = set()
+    out = []
+    for x in seq:
+        if x not in seen:
+            out.append(x)
+            seen.add(x)
+    return out
+
+candidate_age_cols = dedup(candidate_age_cols)
+candidate_gender_cols = dedup(candidate_gender_cols)
+
+# Print required lists in strict format
+print(f"candidate_age_cols = {candidate_age_cols}")
+print(f"candidate_gender_cols = {candidate_gender_cols}")
+
+# Preview extracted candidate columns
+age_cols_present = [c for c in candidate_age_cols if c in clinical_df.columns]
+gender_cols_present = [c for c in candidate_gender_cols if c in clinical_df.columns]
+
+age_preview = preview_df(clinical_df[age_cols_present]) if age_cols_present else {}
+gender_preview = preview_df(clinical_df[gender_cols_present]) if gender_cols_present else {}
+
+print(age_preview)
+print(gender_preview)
+
+# Step 3: Select Demographic Features
+# Select the most appropriate columns for age and gender based on candidate previews
+age_col = None
+gender_col = None
+
+# Prefer age in years over days_to_birth
+if 'age_at_initial_pathologic_diagnosis' in candidate_age_cols:
+    age_col = 'age_at_initial_pathologic_diagnosis'
+elif candidate_age_cols:
+    # Fallback to the first available candidate if needed
+    age_col = candidate_age_cols[0]
+
+# Gender column appears clean and meaningful
+if 'gender' in candidate_gender_cols:
+    gender_col = 'gender'
+elif candidate_gender_cols:
+    gender_col = candidate_gender_cols[0]
+
+print(f"Selected age_col: {age_col}")
+print(f"Selected gender_col: {gender_col}")
+
+# Step 4: Feature Engineering and Validation
+import os
+
+# 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 using NCBI synonym mapping and save
+gene_df_norm = normalize_gene_symbols_in_index(genetic_df.copy())
+os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)
+gene_df_norm.to_csv(out_gene_data_file)
+
+# 3) Link clinical and genetic data on sample IDs
+linked_data = selected_clinical_df.join(gene_df_norm.T, how='inner')
+
+# 4) Handle missing values systematically
+linked_data_clean = handle_missing_values(linked_data, trait_col=trait)
+
+# 5) Determine bias in trait and demographics; drop biased demographics
+trait_biased, linked_data_clean = judge_and_remove_biased_features(linked_data_clean, trait)
+
+# 6) Final validation and save cohort info
+# Cast to native Python bool to avoid potential JSON serialization issues
+is_gene_available = bool((gene_df_norm.shape[0] > 0) and (gene_df_norm.shape[1] > 0))
+is_trait_available = bool((trait in linked_data_clean.columns) and linked_data_clean[trait].notna().any())
+
+note = (
+    "INFO: TCGA LIHC cohort processed. Trait derived from TCGA sample barcode (tumor=01-09, normal=10-19). "
+    "Gene symbols normalized via NCBI synonyms; unmapped symbols removed; duplicates averaged."
+)
+
+try:
+    is_usable = validate_and_save_cohort_info(
+        is_final=True,
+        cohort="TCGA",
+        info_path=json_path,
+        is_gene_available=is_gene_available,
+        is_trait_available=is_trait_available,
+        is_biased=bool(trait_biased),
+        df=linked_data_clean,
+        note=note
+    )
+except TypeError:
+    # If existing JSON contains non-serializable types from previous runs, reset and retry
+    if os.path.exists(json_path):
+        os.remove(json_path)
+    is_usable = validate_and_save_cohort_info(
+        is_final=True,
+        cohort="TCGA",
+        info_path=json_path,
+        is_gene_available=is_gene_available,
+        is_trait_available=is_trait_available,
+        is_biased=bool(trait_biased),
+        df=linked_data_clean,
+        note=note
+    )
+
+# 7) Save linked data only if usable
+if bool(is_usable):
+    os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
+    linked_data_clean.to_csv(out_data_file)