{ "cells": [ { "cell_type": "code", "execution_count": null, "id": "e8c1a62a", "metadata": {}, "outputs": [], "source": [ "import sys\n", "import os\n", "sys.path.append(os.path.abspath(os.path.join(os.getcwd(), '../..')))\n", "\n", "# Path Configuration\n", "from tools.preprocess import *\n", "\n", "# Processing context\n", "trait = \"Chronic_kidney_disease\"\n", "cohort = \"GSE180394\"\n", "\n", "# Input paths\n", "in_trait_dir = \"../../input/GEO/Chronic_kidney_disease\"\n", "in_cohort_dir = \"../../input/GEO/Chronic_kidney_disease/GSE180394\"\n", "\n", "# Output paths\n", "out_data_file = \"../../output/preprocess/Chronic_kidney_disease/GSE180394.csv\"\n", "out_gene_data_file = \"../../output/preprocess/Chronic_kidney_disease/gene_data/GSE180394.csv\"\n", "out_clinical_data_file = \"../../output/preprocess/Chronic_kidney_disease/clinical_data/GSE180394.csv\"\n", "json_path = \"../../output/preprocess/Chronic_kidney_disease/cohort_info.json\"\n" ] }, { "cell_type": "markdown", "id": "6874114c", "metadata": {}, "source": [ "### Step 1: Initial Data Loading" ] }, { "cell_type": "code", "execution_count": null, "id": "b9717358", "metadata": {}, "outputs": [], "source": [ "from tools.preprocess import *\n", "# 1. Identify the paths to the SOFT file and the matrix file\n", "soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)\n", "\n", "# 2. Read the matrix file to obtain background information and sample characteristics data\n", "background_prefixes = ['!Series_title', '!Series_summary', '!Series_overall_design']\n", "clinical_prefixes = ['!Sample_geo_accession', '!Sample_characteristics_ch1']\n", "background_info, clinical_data = get_background_and_clinical_data(matrix_file, background_prefixes, clinical_prefixes)\n", "\n", "# 3. Obtain the sample characteristics dictionary from the clinical dataframe\n", "sample_characteristics_dict = get_unique_values_by_row(clinical_data)\n", "\n", "# 4. Explicitly print out all the background information and the sample characteristics dictionary\n", "print(\"Background Information:\")\n", "print(background_info)\n", "print(\"Sample Characteristics Dictionary:\")\n", "print(sample_characteristics_dict)\n" ] }, { "cell_type": "markdown", "id": "42edf409", "metadata": {}, "source": [ "### Step 2: Dataset Analysis and Clinical Feature Extraction" ] }, { "cell_type": "code", "execution_count": null, "id": "e6415012", "metadata": {}, "outputs": [], "source": [ "import pandas as pd\n", "import numpy as np\n", "import os\n", "import re\n", "from typing import Dict, List, Optional, Callable, Any, Tuple\n", "\n", "# 1. Gene Expression Data Availability\n", "# From the background information, this is gene expression data from Affymetrix microarray\n", "# \"Profiling was performed on Affymetrix ST2.1 microarray platform\"\n", "is_gene_available = True\n", "\n", "# 2.1 Data Availability\n", "# For trait:\n", "# The sample group includes different kidney diseases and living donors\n", "trait_row = 0 # This corresponds to 'sample group' in the sample characteristics\n", "\n", "# For age:\n", "# No age information is available in the characteristics\n", "age_row = None\n", "\n", "# For gender:\n", "# No gender information is available in the characteristics\n", "gender_row = None\n", "\n", "# 2.2 Data Type Conversion Functions\n", "def convert_trait(value_str):\n", " if not isinstance(value_str, str):\n", " return None\n", " \n", " # Extract the value after colon\n", " if \":\" in value_str:\n", " value = value_str.split(\":\", 1)[1].strip()\n", " else:\n", " value = value_str.strip()\n", " \n", " # Binary classification: Living donor (0) vs CKD (1)\n", " if \"Living donor\" in value:\n", " return 0 # Control\n", " elif any(term in value for term in [\"DN\", \"FSGS\", \"GN\", \"IgAN\", \"Nephritis\", \"Hypertensive Nephrosclerosis\", \n", " \"Light-Chain Deposit Disease\", \"LN-WHO\", \"MCD\", \"MN\", \"CKD\", \n", " \"Interstitial fibrosis\", \"Thin-BMD\"]):\n", " return 1 # CKD patient\n", " elif \"Tumor Nephrectomy\" in value:\n", " # These are unaffected parts from tumor nephrectomy, likely normal kidney tissue\n", " return 0\n", " \n", " return None # Unknown or undefined\n", "\n", "# The following functions are defined as placeholders since the data is not available\n", "def convert_age(value_str):\n", " return None\n", "\n", "def convert_gender(value_str):\n", " return None\n", "\n", "# 3. Save Metadata\n", "# Determine trait data availability\n", "is_trait_available = trait_row is not None\n", "\n", "# Initial validation\n", "validate_and_save_cohort_info(\n", " is_final=False,\n", " cohort=cohort,\n", " info_path=json_path,\n", " is_gene_available=is_gene_available,\n", " is_trait_available=is_trait_available\n", ")\n", "\n", "# 4. Clinical Feature Extraction\n", "# We only process this step if clinical data is available\n", "if trait_row is not None:\n", " # Convert the sample characteristics dictionary to a DataFrame\n", " # The dictionary is in the format {row_index: [values_for_samples]}\n", " # We need to create a DataFrame where each row is a feature and each column is a sample\n", " \n", " # Sample characteristics from the previous output\n", " sample_char_dict = {\n", " 0: ['sample group: Living donor', \"sample group: 2' FSGS\", 'sample group: chronic Glomerulonephritis (GN) with infiltration by CLL', \n", " 'sample group: DN', 'sample group: FGGS', 'sample group: FSGS', 'sample group: Hydronephrosis', 'sample group: IgAN', \n", " 'sample group: Interstitial nephritis', 'sample group: Hypertensive Nephrosclerosis', \n", " 'sample group: Light-Chain Deposit Disease (IgG lambda)', 'sample group: LN-WHO III', 'sample group: LN-WHO III+V', \n", " 'sample group: LN-WHO IV', 'sample group: LN-WHO IV+V', 'sample group: LN-WHO V', 'sample group: LN-WHO-I/II', \n", " 'sample group: MCD', 'sample group: MN', 'sample group: CKD with mod-severe Interstitial fibrosis', \n", " 'sample group: Thin-BMD', 'sample group: Unaffected parts of Tumor Nephrectomy'],\n", " 1: ['tissue: Tubuli from kidney biopsy']\n", " }\n", " \n", " # Create a DataFrame from the dictionary\n", " # Each key in the dictionary becomes a row in the DataFrame\n", " clinical_data = pd.DataFrame.from_dict(sample_char_dict, orient='index')\n", " \n", " # Extract clinical features\n", " selected_clinical_df = geo_select_clinical_features(\n", " clinical_df=clinical_data,\n", " trait=trait,\n", " trait_row=trait_row,\n", " convert_trait=convert_trait,\n", " age_row=age_row,\n", " convert_age=convert_age,\n", " gender_row=gender_row,\n", " convert_gender=convert_gender\n", " )\n", " \n", " # Preview the data\n", " preview = preview_df(selected_clinical_df)\n", " print(\"Preview of selected clinical features:\")\n", " for key, value in preview.items():\n", " print(f\"{key}: {value}\")\n", " \n", " # Ensure output directory exists\n", " os.makedirs(os.path.dirname(out_clinical_data_file), exist_ok=True)\n", " \n", " # Save the clinical data\n", " selected_clinical_df.to_csv(out_clinical_data_file, index=False)\n", " print(f\"Clinical data saved to {out_clinical_data_file}\")\n" ] }, { "cell_type": "markdown", "id": "58d3a7ce", "metadata": {}, "source": [ "### Step 3: Gene Data Extraction" ] }, { "cell_type": "code", "execution_count": null, "id": "df6d0124", "metadata": {}, "outputs": [], "source": [ "# 1. Identify the paths to the SOFT file and the matrix file\n", "soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)\n", "print(f\"SOFT file: {soft_file}\")\n", "print(f\"Matrix file: {matrix_file}\")\n", "\n", "# Set gene availability flag\n", "is_gene_available = True # Assume gene data is available\n", "\n", "# Extract gene data\n", "try:\n", " # Extract gene data from the matrix file\n", " gene_data = get_genetic_data(matrix_file)\n", " print(f\"Gene data shape: {gene_data.shape}\")\n", " \n", " # Print the first 20 gene/probe identifiers\n", " print(\"First 20 gene/probe identifiers:\")\n", " print(gene_data.index[:20].tolist())\n", "except Exception as e:\n", " print(f\"Error extracting gene data: {e}\")\n", " print(f\"File path: {matrix_file}\")\n", " print(\"Please check if the file exists and contains the expected markers.\")\n" ] }, { "cell_type": "markdown", "id": "65b70031", "metadata": {}, "source": [ "### Step 4: Gene Identifier Review" ] }, { "cell_type": "code", "execution_count": null, "id": "93ad297a", "metadata": {}, "outputs": [], "source": [ "# Reviewing the gene identifiers\n", "# The identifiers follow the pattern \"number_at\" which is characteristic of Affymetrix probe IDs\n", "# These are not standard human gene symbols and need to be mapped\n", "# For example, '100009613_at' is an Affymetrix probe ID, not a standard gene symbol like \"BRCA1\"\n", "\n", "requires_gene_mapping = True\n" ] }, { "cell_type": "markdown", "id": "3c5252c1", "metadata": {}, "source": [ "### Step 5: Gene Annotation" ] }, { "cell_type": "code", "execution_count": null, "id": "7c2face4", "metadata": {}, "outputs": [], "source": [ "# 1. Use the 'get_gene_annotation' function from the library to get gene annotation data from the SOFT file.\n", "soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)\n", "gene_annotation = get_gene_annotation(soft_file)\n", "\n", "# 2. Analyze the gene annotation dataframe to identify which columns contain the gene identifiers and gene symbols\n", "print(\"\\nGene annotation preview:\")\n", "print(f\"Columns in gene annotation: {gene_annotation.columns.tolist()}\")\n", "print(preview_df(gene_annotation, n=5))\n", "\n", "# Get a more complete view to understand the annotation structure\n", "print(\"\\nComplete sample of a few rows:\")\n", "print(gene_annotation.iloc[:3].to_string())\n", "\n", "# Check if there are any columns that might contain gene information beyond what we've seen\n", "potential_gene_columns = [col for col in gene_annotation.columns if \n", " any(term in col.upper() for term in [\"GENE\", \"SYMBOL\", \"NAME\", \"ID\"])]\n", "print(f\"\\nPotential gene-related columns: {potential_gene_columns}\")\n", "\n", "# Look for additional columns that might contain gene symbols\n", "# Since we only have 'ID' and 'ENTREZ_GENE_ID', check if we need to use Entrez IDs for mapping\n", "gene_id_col = 'ID'\n", "gene_symbol_col = None\n", "\n", "# Check various potential column names for gene symbols\n", "for col_name in ['GENE_SYMBOL', 'SYMBOL', 'GENE', 'GENE_NAME', 'GB_ACC']:\n", " if col_name in gene_annotation.columns:\n", " gene_symbol_col = col_name\n", " break\n", "\n", "# If no dedicated symbol column is found, we'll need to use ENTREZ_GENE_ID\n", "if gene_symbol_col is None and 'ENTREZ_GENE_ID' in gene_annotation.columns:\n", " gene_symbol_col = 'ENTREZ_GENE_ID'\n", " print(\"\\nNo direct gene symbol column found. Will use Entrez Gene IDs for mapping.\")\n", "\n", "if gene_id_col in gene_annotation.columns and gene_symbol_col is not None:\n", " print(f\"\\nSample mappings from '{gene_id_col}' to '{gene_symbol_col}':\")\n", " sample_mappings = gene_annotation[[gene_id_col, gene_symbol_col]].head(10)\n", " print(sample_mappings)\n", " \n", " # Check for non-null mappings to confirm data quality\n", " non_null_mappings = gene_annotation[[gene_id_col, gene_symbol_col]].dropna(subset=[gene_symbol_col])\n", " print(f\"\\nNumber of probes with gene ID mappings: {len(non_null_mappings)}\")\n", " print(f\"Sample of valid mappings:\")\n", " print(non_null_mappings.head(5))\n", "else:\n", " print(\"Required mapping columns not found in the annotation data. Will need to explore alternative mapping approaches.\")\n" ] }, { "cell_type": "markdown", "id": "9e599958", "metadata": {}, "source": [ "### Step 6: Gene Identifier Mapping" ] }, { "cell_type": "code", "execution_count": null, "id": "5ef38c81", "metadata": {}, "outputs": [], "source": [ "# 1. Determine which column in gene annotation corresponds to gene identifiers and which to gene symbols\n", "# From previous analysis, the gene annotation has 'ID' for probe IDs and 'ENTREZ_GENE_ID' for Entrez Gene IDs\n", "probe_col = 'ID'\n", "gene_col = 'ENTREZ_GENE_ID'\n", "\n", "# 2. Extract the two columns from the gene annotation dataframe to create the mapping dataframe\n", "print(\"Creating gene mapping DataFrame...\")\n", "mapping_data = get_gene_mapping(gene_annotation, probe_col, gene_col)\n", "print(f\"Created mapping between {probe_col} and {gene_col}\")\n", "print(f\"Mapping data shape: {mapping_data.shape}\")\n", "print(\"Sample of mapping data:\")\n", "print(mapping_data.head())\n", "\n", "# 3. We need to ensure our mapping works by examining the data formats\n", "# Let's get a fresh copy of the gene expression data\n", "gene_data = get_genetic_data(matrix_file)\n", "print(f\"Gene expression data shape: {gene_data.shape}\")\n", "\n", "# Create a custom mapping approach\n", "# First, get the overlap of probe IDs between expression data and annotation\n", "common_probes = set(gene_data.index) & set(mapping_data['ID'])\n", "print(f\"Number of probes in expression data: {len(gene_data.index)}\")\n", "print(f\"Number of probes in mapping data: {len(mapping_data['ID'])}\")\n", "print(f\"Number of common probes: {len(common_probes)}\")\n", "\n", "# Filter mapping to only include probes that exist in expression data\n", "valid_mapping = mapping_data[mapping_data['ID'].isin(common_probes)]\n", "print(f\"Valid mapping shape after filtering: {valid_mapping.shape}\")\n", "\n", "# Create a direct mapping from probe ID to Entrez Gene ID\n", "probe_to_gene = {}\n", "for idx, row in valid_mapping.iterrows():\n", " probe_id = row['ID'] \n", " gene_id = str(row['Gene']) # Convert to string\n", " \n", " if probe_id not in probe_to_gene:\n", " probe_to_gene[probe_id] = []\n", " probe_to_gene[probe_id].append(gene_id)\n", "\n", "# Create a new gene expression DataFrame\n", "result = pd.DataFrame()\n", "\n", "# For each probe, distribute its expression to its mapped genes\n", "for probe_id, gene_ids in probe_to_gene.items():\n", " if not gene_ids: # Skip if no genes mapped\n", " continue\n", " \n", " # Get probe expression data\n", " probe_expr = gene_data.loc[probe_id]\n", " \n", " # Distribute expression equally among genes\n", " weight = 1.0 / len(gene_ids)\n", " \n", " for gene_id in gene_ids:\n", " # Skip empty gene IDs\n", " if not gene_id or gene_id == 'nan':\n", " continue\n", " \n", " # Add weighted expression to the gene\n", " if gene_id in result.index:\n", " result.loc[gene_id] += probe_expr * weight\n", " else:\n", " result.loc[gene_id] = probe_expr * weight\n", "\n", "print(f\"Converted probe-level data to gene-level expression\")\n", "print(f\"Gene data shape after mapping: {result.shape}\")\n", "\n", "if not result.empty:\n", " print(\"First 10 gene symbols after mapping:\")\n", " print(result.index[:10].tolist())\n", " \n", " # Check for top genes with highest expression to verify data quality\n", " print(\"\\nGenes with highest mean expression:\")\n", " mean_expression = result.mean(axis=1).sort_values(ascending=False)\n", " print(mean_expression.head(10))\n", " \n", " # Update gene_data with our processed result\n", " gene_data = result\n", "else:\n", " print(\"\\nWarning: No genes were mapped. Check the mapping process.\")\n" ] }, { "cell_type": "markdown", "id": "74e09ccc", "metadata": {}, "source": [ "### Step 7: Gene Identifier Mapping" ] }, { "cell_type": "code", "execution_count": null, "id": "2b52a8fb", "metadata": {}, "outputs": [], "source": [ "# 1. Determine which column in gene annotation corresponds to gene identifiers and which to gene symbols\n", "# From previous analysis, the gene annotation has 'ID' for probe IDs and 'ENTREZ_GENE_ID' for Entrez Gene IDs\n", "prob_col = 'ID'\n", "gene_col = 'ENTREZ_GENE_ID'\n", "\n", "# 2. Extract the two columns from the gene annotation dataframe to create the mapping dataframe\n", "print(\"Creating gene mapping DataFrame...\")\n", "mapping_data = get_gene_mapping(gene_annotation, prob_col, gene_col)\n", "print(f\"Created mapping between {prob_col} and {gene_col}\")\n", "print(f\"Mapping data shape: {mapping_data.shape}\")\n", "print(\"Sample of mapping data:\")\n", "print(mapping_data.head())\n", "\n", "# 3. Convert probe-level measurements to gene expression data by applying the gene mapping\n", "print(\"Converting probe-level measurements to gene-level expression data...\")\n", "\n", "# Use the library function to convert probe-level data to gene-level expression data\n", "gene_data = apply_gene_mapping(gene_data, mapping_data)\n", "print(f\"Resulting gene expression data shape: {gene_data.shape}\")\n", "\n", "# Show a sample of the resulting gene data\n", "print(\"Sample of gene expression data:\")\n", "if not gene_data.empty:\n", " print(\"First 10 gene symbols after mapping:\")\n", " print(gene_data.index[:10].tolist())\n", " \n", " # Check for top genes with highest expression to verify data quality\n", " print(\"\\nGenes with highest mean expression:\")\n", " mean_expression = gene_data.mean(axis=1).sort_values(ascending=False)\n", " print(mean_expression.head(10))\n", "else:\n", " print(\"WARNING: No genes were mapped successfully.\")" ] } ], "metadata": {}, "nbformat": 4, "nbformat_minor": 5 }