p3/preprocess/Pheochromocytoma_and_Paraganglioma/code/GSE64957.py

# Path Configuration
from tools.preprocess import *

# Processing context
trait = "Pheochromocytoma_and_Paraganglioma"
cohort = "GSE64957"

# Input paths
in_trait_dir = "../DATA/GEO/Pheochromocytoma_and_Paraganglioma"
in_cohort_dir = "../DATA/GEO/Pheochromocytoma_and_Paraganglioma/GSE64957"

# Output paths
out_data_file = "./output/preprocess/3/Pheochromocytoma_and_Paraganglioma/GSE64957.csv"
out_gene_data_file = "./output/preprocess/3/Pheochromocytoma_and_Paraganglioma/gene_data/GSE64957.csv"
out_clinical_data_file = "./output/preprocess/3/Pheochromocytoma_and_Paraganglioma/clinical_data/GSE64957.csv"
json_path = "./output/preprocess/3/Pheochromocytoma_and_Paraganglioma/cohort_info.json"

# Get file paths
soft_file_path, matrix_file_path = geo_get_relevant_filepaths(in_cohort_dir)

# Get background info and clinical data
background_info, clinical_data = get_background_and_clinical_data(matrix_file_path)
print("Background Information:")
print(background_info)
print("\nSample Characteristics:")

# Get dictionary of unique values per row 
unique_values_dict = get_unique_values_by_row(clinical_data)
for row, values in unique_values_dict.items():
    print(f"\n{row}:")
    print(values)
# 1. Gene Expression Data Availability
# Yes, this dataset contains gene expression data from Affymetrix Human Genome U133 Plus 2.0 Array
is_gene_available = True

# 2.1 Variable Availability
# Trait (pheo vs non-pheo) can be inferred from row 0 (disease field)
trait_row = 0

# Age not available 
age_row = None

# Gender not available
gender_row = None

# 2.2 Data Type Conversion Functions
def convert_trait(value):
    """Convert disease value to binary: Pheochromocytoma (1) vs non-Pheochromocytoma (0)"""
    if not isinstance(value, str):
        return None
    value = value.lower().split(': ')[-1]
    if 'pheochromocytoma' in value:
        return 1
    elif "conn's syndrome" in value:
        return 0
    return None

def convert_age(value):
    return None

def convert_gender(value): 
    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. 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
    )
    
    # Preview the extracted features
    preview = preview_df(clinical_features)
    
    # Save clinical features
    clinical_features.to_csv(out_clinical_data_file)
# Get gene expression data from matrix file
genetic_data = get_genetic_data(matrix_file_path)

# Examine data structure
print("Data structure and head:")
print(genetic_data.head())

print("\nShape:", genetic_data.shape)

print("\nFirst 20 row IDs (gene/probe identifiers):")
print(list(genetic_data.index)[:20])

# Get a few column names to verify sample IDs
print("\nFirst 5 column names:")
print(list(genetic_data.columns)[:5])
# The IDs appear to be numerical probe identifiers (e.g. 7892501, 7892502)
# rather than human gene symbols (e.g. TP53, BRCA1)
# These are likely probe IDs from a microarray platform that need to be mapped to gene symbols
requires_gene_mapping = True
# Extract gene annotation data
gene_annotation = get_gene_annotation(soft_file_path)

# Display column names and preview data
print("Column names:")
print(gene_annotation.columns)

print("\nPreview of gene annotation data:")
print(preview_df(gene_annotation))
# In gene_annotation, 'ID' column stores probe identifiers matching genetic_data indices
# 'gene_assignment' column stores gene symbol information in format "gene symbol // gene title // ..."

# Create mapping dataframe from probe ID to gene symbol
mapping_data = get_gene_mapping(gene_annotation, 'ID', 'gene_assignment')

# Extract gene symbols from gene assignment string
mapping_data['Gene'] = mapping_data['Gene'].str.split(' // ').str[0]

# Apply gene mapping to convert probe measurements to gene expression
gene_data = apply_gene_mapping(genetic_data, mapping_data)

# Preview results
print("Gene expression data shape:", gene_data.shape)
print("\nFirst few gene symbols:")
print(list(gene_data.index)[:10])
print("\nFirst few values:")
print(gene_data.head())
# First check the clinical data processing
print("Clinical Data Preview:")
selected_clinical_df = pd.read_csv(out_clinical_data_file, index_col=0)
print(selected_clinical_df.head())
print("\nClinical Data Shape:", selected_clinical_df.shape)
print("\nClinical Data Column Names:", selected_clinical_df.columns)
print("\nClinical Data Info:")
print(selected_clinical_df.info())

# If clinical data is valid, proceed with processing
if not selected_clinical_df.empty and not selected_clinical_df.isna().all().all():
    # 1. Normalize gene symbols
    genetic_data = normalize_gene_symbols_in_index(gene_data)
    genetic_data.to_csv(out_gene_data_file)
    print("\nGenetic Data Shape after normalization:", genetic_data.shape)

    # 2. Link clinical and genetic data 
    linked_data = geo_link_clinical_genetic_data(selected_clinical_df, genetic_data)
    print("\nLinked Data Shape:", linked_data.shape)

    # 3. Handle missing values
    linked_data = handle_missing_values(linked_data, trait)

    # 4. Check for bias
    trait_biased, linked_data = judge_and_remove_biased_features(linked_data, trait)

    # 5. Final validation and saving metadata
    note = "Gene expression data from Affymetrix array with disease status (pheochromocytoma vs Conn's syndrome)"
    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=trait_biased,
        df=linked_data,
        note=note
    )

    # 6. Save if usable
    if is_usable:
        os.makedirs(os.path.dirname(out_data_file), exist_ok=True)
        linked_data.to_csv(out_data_file)
else:
    print("Error: Clinical data processing failed - empty or invalid data")
    validate_and_save_cohort_info(
        is_final=True,
        cohort=cohort,
        info_path=json_path,
        is_gene_available=True,
        is_trait_available=True,
        is_biased=True,
        df=pd.DataFrame(),
        note="Clinical data processing failed - empty or invalid data"
    )