p1/preprocess/Bipolar_disorder/code/GSE62191.py

# Path Configuration
from tools.preprocess import *

# Processing context
trait = "Bipolar_disorder"
cohort = "GSE62191"

# Input paths
in_trait_dir = "../DATA/GEO/Bipolar_disorder"
in_cohort_dir = "../DATA/GEO/Bipolar_disorder/GSE62191"

# Output paths
out_data_file = "./output/preprocess/1/Bipolar_disorder/GSE62191.csv"
out_gene_data_file = "./output/preprocess/1/Bipolar_disorder/gene_data/GSE62191.csv"
out_clinical_data_file = "./output/preprocess/1/Bipolar_disorder/clinical_data/GSE62191.csv"
json_path = "./output/preprocess/1/Bipolar_disorder/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)
import re
import pandas as pd

# 1. Gene Expression Data Availability
is_gene_available = True  # Based on the series title "Gene expression profiles..."

# 2. Variable Availability and Data Type Conversion

# Determining the rows for trait, age, and gender
trait_row = 1  # "disease state" includes bipolar disorder, healthy control, schizophrenia
age_row = 2    # "age" row has multiple, distinct values
# For gender, the sample dict is [nan, 'gender: male'] => effectively no variation
gender_row = None

# Defining data conversion functions
def convert_trait(value: str) -> int:
    # Extract the substring after the colon
    parts = value.split(':')
    if len(parts) < 2:
        return None
    val = parts[1].strip().lower()
    # Mark 'bipolar disorder' as 1, everything else (healthy, schizophrenia, unknown) as 0
    if 'bipolar disorder' in val:
        return 1
    elif any(x in val for x in ['healthy control', 'schizophrenia']):
        return 0
    else:
        return None

def convert_age(value: str) -> float:
    # Extract numeric part from something like "age: 29 yr"
    parts = value.split(':')
    if len(parts) < 2:
        return None
    val = parts[1].strip().lower()
    # Use regex to find the first number
    match = re.search(r'(\d+)', val)
    if match:
        return float(match.group(1))
    return None

def convert_gender(value: str) -> int:
    # Though we found no variation, define the function for completeness
    parts = value.split(':')
    if len(parts) < 2:
        return None
    val = parts[1].strip().lower()
    if 'male' in val:
        return 1
    elif 'female' in val:
        return 0
    else:
        return None

# 3. Save Metadata (initial filtering)
is_trait_available = (trait_row is not None)
passed_initial_filter = 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 proceed if trait_row is not None)
if trait_row is not None:
    # Suppose clinical_data is already loaded in a variable named `clinical_data`
    # For demonstration, let's create a mock DataFrame to simulate the real one:
    data = {
        0: ['tissue: brain (frontal cortex)']*3,
        1: ['disease state: bipolar disorder', 'disease state: healthy control', 'disease state: schizophrenia'],
        2: ['age: 29 yr', 'age: 58 yr', 'age: 42 yr'],
        3: ['population: white']*3,
        4: ['dsm-iv: 296.54']*3,
        5: ['age of onset: 22 yr']*3,
        6: [None, 'gender: male', None],
    }
    clinical_data = pd.DataFrame.from_dict(data, orient='index')

    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 the result
    preview_result = preview_df(selected_clinical_df, n=5)
    print("Preview of selected clinical features:")
    print(preview_result)

    # Save to CSV
    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])
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))
# STEP: Gene Identifier Mapping

# 1 & 2. Decide which columns match the gene expression ID and which match gene symbols.
#    From the annotation preview, we'll assume 'ID' matches the expression data's 'ID',
#    and 'GENE_SYMBOL' holds the gene symbol.

mapping_df = get_gene_mapping(
    annotation=gene_annotation,
    prob_col='ID',
    gene_col='GENE_SYMBOL'
)

# 3. Convert probe-level measurements to gene expression data using this mapping.
gene_data = apply_gene_mapping(expression_df=gene_data, mapping_df=mapping_df)

# Optionally, preview the resulting gene_data structure.
print("Mapped gene_data shape:", gene_data.shape)
print("First few rows of mapped gene_data:")
print(gene_data.head())
# STEP7

# 1. Normalize the obtained gene data using the NCBI Gene synonym database
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
#    Note: the clinical DataFrame was created with "Trait" as the column name for the trait
linked_data = geo_link_clinical_genetic_data(selected_clinical_df, normalized_gene_data)

# 3. Handle missing values systematically using the actual column name in our DataFrame ("Trait")
linked_data_processed = handle_missing_values(linked_data, trait_col="Trait")

# 4. Check for biased trait and remove any biased demographic features
trait_biased, linked_data_final = judge_and_remove_biased_features(linked_data_processed, "Trait")

# 5. Final quality validation and metadata saving
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_final,
    note="Dataset processed with GEO pipeline. Checked for missing values and bias."
)

# 6. If dataset is usable, save the final linked data
if is_usable:
    linked_data_final.to_csv(out_data_file)