Liu-Hy's picture
Add files using upload-large-folder tool
a0747da verified
# Path Configuration
from tools.preprocess import *
# Processing context
trait = "Lung_Cancer"
cohort = "GSE244645"
# Input paths
in_trait_dir = "../DATA/GEO/Lung_Cancer"
in_cohort_dir = "../DATA/GEO/Lung_Cancer/GSE244645"
# Output paths
out_data_file = "./output/preprocess/3/Lung_Cancer/GSE244645.csv"
out_gene_data_file = "./output/preprocess/3/Lung_Cancer/gene_data/GSE244645.csv"
out_clinical_data_file = "./output/preprocess/3/Lung_Cancer/clinical_data/GSE244645.csv"
json_path = "./output/preprocess/3/Lung_Cancer/cohort_info.json"
# Get file paths
soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
# Extract background info and clinical data using specified prefixes
background_info, clinical_data = get_background_and_clinical_data(
matrix_file,
prefixes_a=['!Series_title', '!Series_summary', '!Series_overall_design'],
prefixes_b=['!Sample_geo_accession', '!Sample_characteristics_ch1']
)
# Get unique values per clinical feature
sample_characteristics = get_unique_values_by_row(clinical_data)
# Print background info
print("Dataset Background Information:")
print(f"{background_info}\n")
# Print sample characteristics
print("Sample Characteristics:")
for feature, values in sample_characteristics.items():
print(f"Feature: {feature}")
print(f"Values: {values}\n")
# 1. Gene Expression Data Availability
# Based on background info, this is platelet gene expression data from microarray
is_gene_available = True
# 2. Variable Availability and Row Detection
# trait (cancer state) is in Feature 1 - tumour presence/absence
trait_row = 1
# age is in Feature 5
age_row = 5
# gender is in Feature 4
gender_row = 4
# 2. Data Type Conversion Functions
def convert_trait(value: str) -> int:
"""Convert tumor status to binary: 1 for tumor presence, 0 for tumor free"""
if not value or value == '-':
return None
value = value.split(': ')[1].lower()
if 'tumour presence' in value:
return 1
elif 'tumour free' in value:
return 0
return None
def convert_age(value: str) -> float:
"""Convert age string to float"""
if not value or value == '-':
return None
try:
return float(value.split(': ')[1])
except:
return None
def convert_gender(value: str) -> int:
"""Convert gender to binary: 1 for male, 0 for female"""
if not value or value == '-':
return None
value = value.split(': ')[1].lower()
if value == 'male':
return 1
elif value == 'female':
return 0
return None
# 3. Save metadata
validate_and_save_cohort_info(is_final=False,
cohort=cohort,
info_path=json_path,
is_gene_available=is_gene_available,
is_trait_available=trait_row is not None)
# 4. Extract clinical features
if trait_row is not None:
clinical_features = 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 of extracted clinical features:")
print(preview_df(clinical_features))
# Save clinical data
clinical_features.to_csv(out_clinical_data_file)
# Get file paths
soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
# Extract gene expression data from matrix file
gene_data = get_genetic_data(matrix_file)
# Print first 20 row IDs and shape of data to help debug
print("Shape of gene expression data:", gene_data.shape)
print("\nFirst few rows of data:")
print(gene_data.head())
print("\nFirst 20 gene/probe identifiers:")
print(gene_data.index[:20])
# Inspect a snippet of raw file to verify identifier format
import gzip
with gzip.open(matrix_file, 'rt', encoding='utf-8') as f:
lines = []
for i, line in enumerate(f):
if "!series_matrix_table_begin" in line:
# Get the next 5 lines after the marker
for _ in range(5):
lines.append(next(f).strip())
break
print("\nFirst few lines after matrix marker in raw file:")
for line in lines:
print(line)
# The identifiers (e.g. TC0100006437.hg.1) appear to be probe IDs from a microarray platform
# rather than standard human gene symbols like BRCA1, TP53 etc.
# They will need to be mapped to official gene symbols.
requires_gene_mapping = True
# Extract gene annotation data
gene_metadata = get_gene_annotation(soft_file)
# Try searching for ID patterns in all columns
print("All column names:", gene_metadata.columns.tolist())
print("\nPreview first few rows of each column to locate numeric IDs:")
for col in gene_metadata.columns:
sample_values = gene_metadata[col].dropna().head().tolist()
print(f"\n{col}:")
print(sample_values)
# Inspect raw file to see unfiltered annotation format
import gzip
print("\nRaw SOFT file preview:")
with gzip.open(soft_file, 'rt', encoding='utf-8') as f:
header = []
for i, line in enumerate(f):
header.append(line.strip())
if i >= 10: # Preview first 10 lines
break
print('\n'.join(header))
# Create function to extract gene symbols from annotation text
def extract_gene_symbols(text):
if not isinstance(text, str):
return []
symbols = []
# Get symbols from parentheses after "Homo sapiens"
matches = re.findall(r'Homo sapiens.*?\((\w+)\)', text)
symbols.extend(matches)
# Get symbols from HGNC tags
hgnc_matches = re.findall(r'\[Source:HGNC Symbol;Acc:HGNC:\d+\].*?(\w+)', text)
symbols.extend(hgnc_matches)
return list(set(symbols))
# Create mapping dataframe by extracting gene symbols from SPOT_ID.1 column
gene_metadata['Gene'] = gene_metadata['SPOT_ID.1'].apply(extract_gene_symbols)
mapping_data = gene_metadata[['ID', 'Gene']].copy()
# Apply mapping to convert probe-level data to gene-level data
gene_data = apply_gene_mapping(gene_data, mapping_data)
# Save genetic data
gene_data.to_csv(out_gene_data_file)
print("Shape of gene expression data after mapping:", gene_data.shape)
print("\nFirst few rows and columns of mapped data:")
print(gene_data.head().iloc[:, :5])
# Get file paths
soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)
# Extract gene annotation data and load gene data from file
gene_metadata = get_gene_annotation(soft_file)
# Refine extraction of gene symbols
def extract_gene_symbols_from_annotation(text):
if not isinstance(text, str):
return []
# Focus on RefSeq entries which typically have cleaner gene names
refseq_match = re.search(r'NM_\d+ // RefSeq // Homo sapiens .*? \((\w+)\)', text)
if refseq_match:
return [refseq_match.group(1)] # Return the symbol in parentheses
return []
# Create mapping dataframe
gene_metadata['Gene'] = gene_metadata['SPOT_ID.1'].apply(extract_gene_symbols_from_annotation)
mapping_data = gene_metadata[['ID', 'Gene']].copy()
# Re-apply mapping with refined gene symbol extraction
gene_data = apply_gene_mapping(gene_data, mapping_data)
# Normalize gene symbols
gene_data = normalize_gene_symbols_in_index(gene_data)
# Save normalized gene expression data
gene_data.to_csv(out_gene_data_file)
# Link clinical and genetic data
selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
linked_data = geo_link_clinical_genetic_data(selected_clinical_df, gene_data)
# Handle missing values
linked_data = handle_missing_values(linked_data, trait)
# Evaluate bias in features
is_biased, linked_data = judge_and_remove_biased_features(linked_data, trait)
# Record cohort information
is_usable = validate_and_save_cohort_info(
is_final=True,
cohort=cohort,
info_path=json_path,
is_gene_available=True,
is_trait_available=True,
is_biased=is_biased,
df=linked_data,
note="Gene expression and clinical data processed and linked using refined gene symbol extraction."
)
# Save data if usable
if is_usable:
linked_data.to_csv(out_data_file)