File size: 8,362 Bytes
9efdaa1 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 | # Path Configuration
from tools.preprocess import *
# Processing context
trait = "Endometrioid_Cancer"
cohort = "GSE73637"
# Input paths
in_trait_dir = "../DATA/GEO/Endometrioid_Cancer"
in_cohort_dir = "../DATA/GEO/Endometrioid_Cancer/GSE73637"
# Output paths
out_data_file = "./output/z2/preprocess/Endometrioid_Cancer/GSE73637.csv"
out_gene_data_file = "./output/z2/preprocess/Endometrioid_Cancer/gene_data/GSE73637.csv"
out_clinical_data_file = "./output/z2/preprocess/Endometrioid_Cancer/clinical_data/GSE73637.csv"
json_path = "./output/z2/preprocess/Endometrioid_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 Data Availability
is_gene_available = True
# 2.1 Data Availability
trait_row = 3 # histopathology data contains endometrioid cancer information
age_row = None # Not available for cell lines
gender_row = None # Not available for cell lines
# 2.2 Data Type Conversion functions
def convert_trait(value):
"""Convert histopathology to binary endometrioid cancer status"""
if value is None:
return None
# Extract value after colon
if ':' in str(value):
histology = str(value).split(':')[1].strip()
else:
histology = str(value).strip()
# Check if it contains "Endometrioid" (including "Endometroid" variant)
if "Endometrioid" in histology or "Endometroid" in histology:
return 1
else:
return 0
def convert_age(value):
"""Age conversion function (not used since age_row is None)"""
return None
def convert_gender(value):
"""Gender conversion function (not used since gender_row is None)"""
return None
# 3. 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
if trait_row is not None:
selected_clinical_data = geo_select_clinical_features(clinical_data, trait, trait_row, convert_trait,
age_row, convert_age, gender_row, convert_gender)
print("Clinical data extracted:")
print(preview_df(selected_clinical_data))
# Save clinical data
os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)
selected_clinical_data.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
# Examine the gene identifiers from the previous step output
gene_identifiers = ['1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', '13', '14', '15', '16', '17', '18', '19', '20']
print("Sample gene identifiers:", gene_identifiers[:10])
print("These appear to be numeric probe/array identifiers, not human gene symbols")
print("Human gene symbols are typically alphanumeric like BRCA1, TP53, GAPDH, etc.")
print("These numeric identifiers will need to be mapped to actual gene symbols")
requires_gene_mapping = True
# Step 5: Gene Annotation
# First, let's examine the SOFT file structure to understand what we're dealing with
print("Examining SOFT file structure:")
with gzip.open(soft_file, 'rt') as f:
lines = []
for i, line in enumerate(f):
lines.append(line.strip())
if i >= 50: # Read first 50 lines
break
# Show first 20 lines to understand the structure
for i, line in enumerate(lines[:20]):
print(f"Line {i}: {line[:100]}...") # Show first 100 chars of each line
print("\n" + "="*50)
# Look for gene annotation section - typically starts after platform info
annotation_start = None
for i, line in enumerate(lines):
if line.startswith('!platform_table_begin'):
annotation_start = i + 1
print(f"Found annotation section starting at line {annotation_start}")
break
if annotation_start:
print("Sample annotation lines:")
for i in range(annotation_start, min(annotation_start + 10, len(lines))):
if i < len(lines):
print(f"Line {i}: {lines[i]}")
# Try alternative approach to get gene annotation
try:
with gzip.open(soft_file, 'rt') as f:
content = f.read()
# Find the platform table section
if '!platform_table_begin' in content and '!platform_table_end' in content:
start_marker = '!platform_table_begin'
end_marker = '!platform_table_end'
start_idx = content.find(start_marker) + len(start_marker)
end_idx = content.find(end_marker)
table_content = content[start_idx:end_idx].strip()
# Parse as CSV
gene_annotation = pd.read_csv(io.StringIO(table_content), delimiter='\t', low_memory=False)
print("\nSuccessfully extracted gene annotation data!")
print("Gene annotation preview:")
print(preview_df(gene_annotation))
else:
print("Could not find platform table markers in SOFT file")
except Exception as e:
print(f"Alternative parsing also failed: {e}")
# If all parsing fails, we may need to work without gene annotation
gene_annotation = None
# Step 6: Gene Identifier Mapping
# 1. Identify the columns for gene identifiers and gene symbols
# The 'ID' column matches the gene expression data identifiers (numeric: 1, 2, 3, ...)
# The 'GeneSymbol' column contains the actual gene symbols (PRPF8, CAPNS1, etc.)
# 2. Get gene mapping dataframe
gene_mapping = get_gene_mapping(gene_annotation, 'ID', 'GeneSymbol')
# 3. Apply gene mapping to convert probe-level measurements to gene expression data
gene_data = apply_gene_mapping(gene_data, gene_mapping)
print(f"Gene expression data shape after mapping: {gene_data.shape}")
print(f"Sample gene symbols: {list(gene_data.index[:10])}")
# Step 7: Data Normalization and Linking
# 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)
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.
linked_data = geo_link_clinical_genetic_data(selected_clinical_data, normalized_gene_data)
# 3. Handle missing values in the linked data
linked_data = handle_missing_values(linked_data, trait)
# 4. Determine whether the trait and some demographic features are severely biased, and remove biased features.
is_trait_biased, unbiased_linked_data = judge_and_remove_biased_features(linked_data, trait)
# 5. Conduct quality check and save the cohort information.
is_usable = validate_and_save_cohort_info(True, cohort, json_path, True, True, is_trait_biased, unbiased_linked_data)
# 6. If the linked data is usable, save it as a CSV file to 'out_data_file'.
if is_usable:
os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
unbiased_linked_data.to_csv(out_data_file)
print(f"Successfully saved processed data to {out_data_file}")
else:
print("Dataset is not usable for analysis - not saved") |