| # Path Configuration | |
| from tools.preprocess import * | |
| # Processing context | |
| trait = "Cervical_Cancer" | |
| cohort = "GSE163114" | |
| # Input paths | |
| in_trait_dir = "../DATA/GEO/Cervical_Cancer" | |
| in_cohort_dir = "../DATA/GEO/Cervical_Cancer/GSE163114" | |
| # Output paths | |
| out_data_file = "./output/preprocess/1/Cervical_Cancer/GSE163114.csv" | |
| out_gene_data_file = "./output/preprocess/1/Cervical_Cancer/gene_data/GSE163114.csv" | |
| out_clinical_data_file = "./output/preprocess/1/Cervical_Cancer/clinical_data/GSE163114.csv" | |
| json_path = "./output/preprocess/1/Cervical_Cancer/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) | |
| # Step 1: Gene Expression Data Availability | |
| # Based on the background information ("Ki-67 promotes carcinogenesis by enabling global transcriptional programmes") | |
| # and the use of the HeLa cell line, it is likely that this dataset contains gene expression data. | |
| is_gene_available = True | |
| # Step 2: Variable Availability and Data Type Conversion | |
| # From the sample characteristics dictionary: | |
| # {0: ['cell line: HeLa'], 1: ['lentivirus: shRNA control', 'lentivirus: shRNA Ki-67']} | |
| # - All samples come from the HeLa cell line, which is derived from cervical cancer, but this is a constant feature (no variation). | |
| # - There's no row providing age or gender information. | |
| # Hence, no variable has meaningful variation. We set rows to None. | |
| trait_row = None # No row captures a varying cervical cancer trait | |
| age_row = None # No row for age | |
| gender_row = None # No row for gender | |
| # Even though these functions won't be used (since trait_row, age_row, gender_row = None), | |
| # we provide them per instructions. | |
| def convert_trait(value: str): | |
| """ | |
| Convert the trait to the chosen type. | |
| Not applicable here, but defined for completeness. | |
| """ | |
| if not value or ':' not in value: | |
| return None | |
| val = value.split(':', 1)[-1].strip() | |
| return val if val else None | |
| def convert_age(value: str): | |
| """ | |
| Convert age data to a continuous type. | |
| Not applicable here, but defined for completeness. | |
| """ | |
| if not value or ':' not in value: | |
| return None | |
| val = value.split(':', 1)[-1].strip() | |
| # We do not actually have numeric values, so just return None. | |
| return None | |
| def convert_gender(value: str): | |
| """ | |
| Convert gender data to binary. | |
| Not applicable here, but defined for completeness. | |
| """ | |
| if not value or ':' not in value: | |
| return None | |
| val = value.split(':', 1)[-1].strip().lower() | |
| if val in ['male', 'm']: | |
| return 1 | |
| elif val in ['female', 'f']: | |
| return 0 | |
| return None | |
| # Step 3: Save Metadata | |
| # If trait_row is None, trait data is considered unavailable. | |
| 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 | |
| ) | |
| # Step 4: Since trait_row is None, we skip geo_select_clinical_features. | |
| # No clinical data extraction is performed because the trait is not available. | |
| # 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 the numeric IDs (1,2,3,...), they do not appear to be standard human gene symbols. | |
| # They seem like probe identifiers or some form of numeric reference that would require mapping. | |
| print("These numeric IDs likely need mapping to standard 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 stores the gene identifiers | |
| # matching the gene expression data. From the preview, the 'ID' column in gene_annotation | |
| # corresponds to the numeric probe IDs in gene_data. For gene symbols, we use 'GENE_SYMBOL'. | |
| # 2. Get a gene mapping dataframe | |
| mapping_df = get_gene_mapping(gene_annotation, prob_col='ID', gene_col='GENE_SYMBOL') | |
| # 3. Convert probe-level measurements to gene expression data | |
| gene_data = apply_gene_mapping(gene_data, mapping_df) | |
| # Print a quick check of the mapped dataframe | |
| print("Mapped gene_data shape:", gene_data.shape) | |
| print("Head of mapped gene_data:") | |
| print(gene_data.head()) | |
| # STEP 7 | |
| # 1. Normalize gene symbols in the gene_data, then save to CSV. | |
| normalized_gene_data = normalize_gene_symbols_in_index(gene_data) | |
| normalized_gene_data.to_csv(out_gene_data_file) | |
| # Since trait_row was None in earlier steps, there is no actual trait data available. | |
| # We cannot link clinical data or do trait-based QC. However, the library requires a final | |
| # validation with a DataFrame and a Boolean for is_biased. | |
| # Create an empty DataFrame as a placeholder, and declare is_biased=False by default. | |
| placeholder_df = pd.DataFrame() | |
| trait_biased = False | |
| # 2. Perform final validation, marking trait as unavailable but providing the required arguments. | |
| is_usable = validate_and_save_cohort_info( | |
| is_final=True, | |
| cohort=cohort, | |
| info_path=json_path, | |
| is_gene_available=True, # Gene data is present | |
| is_trait_available=False, # Trait is not available | |
| is_biased=trait_biased, | |
| df=placeholder_df, # Provide a placeholder DataFrame | |
| note="No trait data in this series. Final validation with placeholder DataFrame." | |
| ) | |
| # 3. If the dataset were usable (it won't be without trait), we would save final linked data. | |
| if is_usable: | |
| # Typically we would link data and save CSV, but trait is absent. Skipping. | |
| pass |