p1/preprocess/Chronic_kidney_disease/code/GSE180393.py

# Path Configuration
from tools.preprocess import *

# Processing context
trait = "Chronic_kidney_disease"
cohort = "GSE180393"

# Input paths
in_trait_dir = "../DATA/GEO/Chronic_kidney_disease"
in_cohort_dir = "../DATA/GEO/Chronic_kidney_disease/GSE180393"

# Output paths
out_data_file = "./output/preprocess/1/Chronic_kidney_disease/GSE180393.csv"
out_gene_data_file = "./output/preprocess/1/Chronic_kidney_disease/gene_data/GSE180393.csv"
out_clinical_data_file = "./output/preprocess/1/Chronic_kidney_disease/clinical_data/GSE180393.csv"
json_path = "./output/preprocess/1/Chronic_kidney_disease/cohort_info.json"

# STEP1
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)
# 1. Evaluate whether gene expression data is available
is_gene_available = True  # Based on microarray transcriptome data from the summary

# 2. Determine availability of trait, age, and gender data
#    and define corresponding conversion functions

# From the sample characteristics dictionary, row 0 provides multi-category info
# about disease status. We will code "Living donor" or "unaffected" as 0 (control)
# and all others as 1 (CKD). Rows for age/gender do not appear available.
trait_row = 0
age_row = None
gender_row = None

def convert_trait(value: str):
    # Extract the part after the colon
    parts = value.split(":", 1)
    if len(parts) < 2:
        return None  # Unknown format
    label = parts[1].strip().lower()
    # Living donor or unaffected
    if "living donor" in label or "unaffected" in label:
        return 0
    else:
        return 1

def convert_age(value: str):
    # No age data available
    return None

def convert_gender(value: str):
    # No gender data available
    return None

# 3. Conduct 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. If trait data is available, extract and preview clinical features
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 the resulting clinical dataframe
    preview = preview_df(selected_clinical_df, n=5)
    print("Preview of selected clinical features:", preview)
    selected_clinical_df.to_csv(out_clinical_data_file, index=False)
# STEP3
# 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])
# The listed identifiers (e.g., "100009613_at", "10000_at", etc.) are typical Affymetrix probe set IDs,
# not standard human gene symbols. Therefore, they require mapping to gene symbols.
print("requires_gene_mapping = True")
# STEP5
# 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))
# STEP6: Gene Identifier Mapping

# Observing the preview from step 5, the annotation DataFrame has columns 'ID' and 'ENTREZ_GENE_ID',
# but 'ENTREZ_GENE_ID' is purely numeric, which leads to an empty mapping when the library's
# apply_gene_mapping() tries to parse it as a “human gene symbol.” We will therefore implement
# a custom mapping function that distributes expression values to numeric Entrez IDs without
# filtering them out.

def apply_entrez_id_mapping(expression_df: pd.DataFrame, annotation_df: pd.DataFrame) -> pd.DataFrame:
    """
    Convert probe-level data to gene-level data using numeric Entrez IDs.
    If a probe maps to multiple Entrez IDs (split by '///'), each gene gets an equal split.
    Then we sum contributions from multiple probes associated with the same gene ID.
    """
    # Keep only the columns we need, renaming ENTREZ_GENE_ID to 'Gene'
    mapping_df = annotation_df[['ID', 'ENTREZ_GENE_ID']].copy()
    mapping_df.columns = ['ID', 'Gene']
    mapping_df.dropna(subset=['Gene'], inplace=True)

    # Filter to probes that exist in expression_df
    mapping_df = mapping_df[mapping_df['ID'].isin(expression_df.index)].copy()

    # A single probe might have multiple Entrez IDs separated by '///'
    def split_entrez_ids(gene_str):
        if '///' in gene_str:
            return [x.strip() for x in gene_str.split('///') if x.strip()]
        else:
            return [gene_str.strip()]

    mapping_df['Gene'] = mapping_df['Gene'].apply(split_entrez_ids)
    # Remove rows with no valid gene IDs
    mapping_df = mapping_df[mapping_df['Gene'].map(len) > 0]

    # Count how many genes per probe
    mapping_df['num_genes'] = mapping_df['Gene'].map(len)

    # Explode so each gene occupies its own row
    mapping_df.set_index('ID', inplace=True)
    mapping_df = mapping_df.explode('Gene')

    # Join expression data
    merged_df = mapping_df.join(expression_df, how='inner')
    expr_cols = [c for c in merged_df.columns if c not in ['Gene', 'num_genes']]

    # Divide the probe expression among mapped genes
    merged_df[expr_cols] = merged_df[expr_cols].div(merged_df['num_genes'], axis=0)

    # Sum expressions for each gene
    gene_df = merged_df.groupby('Gene')[expr_cols].sum()
    return gene_df

# 1 & 2. Identify columns for probe ID and gene ID, then map
gene_data = apply_entrez_id_mapping(gene_data, gene_annotation)

# 3. Print shape after mapping to confirm we have gene-level data
print("Gene expression data shape after mapping:", gene_data.shape)
# STEP7
# 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)
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_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.
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=linked_data
)

# 6. If the linked data is usable, save it as a CSV file to 'out_data_file'.
if is_usable:
    unbiased_linked_data.to_csv(out_data_file)