| # Path Configuration | |
| from tools.preprocess import * | |
| # Processing context | |
| trait = "Cystic_Fibrosis" | |
| cohort = "GSE142610" | |
| # Input paths | |
| in_trait_dir = "../DATA/GEO/Cystic_Fibrosis" | |
| in_cohort_dir = "../DATA/GEO/Cystic_Fibrosis/GSE142610" | |
| # Output paths | |
| out_data_file = "./output/preprocess/1/Cystic_Fibrosis/GSE142610.csv" | |
| out_gene_data_file = "./output/preprocess/1/Cystic_Fibrosis/gene_data/GSE142610.csv" | |
| out_clinical_data_file = "./output/preprocess/1/Cystic_Fibrosis/clinical_data/GSE142610.csv" | |
| json_path = "./output/preprocess/1/Cystic_Fibrosis/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. Determine if gene expression data is likely available | |
| is_gene_available = True # Based on the series summary describing transcriptomic (gene expression) analysis | |
| # 2. Determine data availability (trait, age, gender) | |
| # We see that all samples are from a CF cell line without variation. Hence, trait is constant. | |
| # No age or gender information is provided. Therefore: | |
| trait_row = None | |
| age_row = None | |
| gender_row = None | |
| # 2.2 Define conversion functions. | |
| # Since trait_row, age_row, and gender_row are None, these functions will not be used here, | |
| # but we provide them for completeness. | |
| def convert_trait(raw_value: str): | |
| return None | |
| def convert_age(raw_value: str): | |
| return None | |
| def convert_gender(raw_value: str): | |
| return None | |
| # 3. Save metadata with initial filtering | |
| is_trait_available = (trait_row is not None) | |
| is_usable = 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. Since trait_row is None (trait not available), we skip clinical feature extraction. | |
| # 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]) | |
| # Based on inspection, some identifiers (e.g., "7A5", "A2BP1") appear to be synonyms or outdated symbols | |
| # rather than standard HGNC gene symbols. Therefore, they may require mapping to unify them into | |
| # current official human gene symbols. | |
| print("Some gene identifiers are synonyms or aliases rather than current official 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)) | |
| # STEP: Gene Identifier Mapping | |
| # 1. Decide which key in the gene annotation DataFrame matches the gene expression data IDs | |
| # and which key contains the gene symbols. | |
| # From the preview, both "ID" and "ORF" columns appear to match the probe IDs in the expression data, | |
| # but "ORF" likely corresponds to the gene symbol we want. | |
| # 2. Get a gene mapping DataFrame using the library function get_gene_mapping | |
| mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='ORF') | |
| # 3. Convert (probe-level) gene expression data to (gene-level) data | |
| gene_data = apply_gene_mapping(gene_data, mapping_df) |