{ "cells": [ { "cell_type": "code", "execution_count": null, "id": "3c3f683e", "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 = \"Gastroesophageal_reflux_disease_(GERD)\"\n", "cohort = \"GSE28302\"\n", "\n", "# Input paths\n", "in_trait_dir = \"../../input/GEO/Gastroesophageal_reflux_disease_(GERD)\"\n", "in_cohort_dir = \"../../input/GEO/Gastroesophageal_reflux_disease_(GERD)/GSE28302\"\n", "\n", "# Output paths\n", "out_data_file = \"../../output/preprocess/Gastroesophageal_reflux_disease_(GERD)/GSE28302.csv\"\n", "out_gene_data_file = \"../../output/preprocess/Gastroesophageal_reflux_disease_(GERD)/gene_data/GSE28302.csv\"\n", "out_clinical_data_file = \"../../output/preprocess/Gastroesophageal_reflux_disease_(GERD)/clinical_data/GSE28302.csv\"\n", "json_path = \"../../output/preprocess/Gastroesophageal_reflux_disease_(GERD)/cohort_info.json\"\n" ] }, { "cell_type": "markdown", "id": "bd106a12", "metadata": {}, "source": [ "### Step 1: Initial Data Loading" ] }, { "cell_type": "code", "execution_count": null, "id": "93ccc943", "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": "d7985c69", "metadata": {}, "source": [ "### Step 2: Dataset Analysis and Clinical Feature Extraction" ] }, { "cell_type": "code", "execution_count": null, "id": "b933bcd7", "metadata": {}, "outputs": [], "source": [ "I'll provide correct code for this step.\n", "\n", "```python\n", "import pandas as pd\n", "import os\n", "import json\n", "from typing import Optional, Callable, Dict, Any\n", "import numpy as np\n", "\n", "# 1. Gene Expression Data Availability\n", "# Based on the background information, this dataset contains genome-wide expression profiling\n", "# using Illumina whole-genome Beadarray on RNA from esophageal biopsy tissues\n", "is_gene_available = True\n", "\n", "# 2. Data Availability and Conversion\n", "\n", "# 2.1 Trait data - Barrett's esophagus related to GERD\n", "trait_row = 0 # \"tissue type\" row\n", "\n", "# Function to convert Barrett's esophagus data to binary values\n", "def convert_trait(value):\n", " if value is None or pd.isna(value):\n", " return None\n", " if \":\" in str(value):\n", " value = str(value).split(\":\", 1)[1].strip()\n", " \n", " if \"barrett\" in value.lower() or \"be\" in value.lower():\n", " return 1 # Barrett's esophagus\n", " elif \"normal\" in value.lower() or \"squamous\" in value.lower():\n", " return 0 # Normal esophageal tissue (control)\n", " elif \"adenocarcinoma\" in value.lower() or \"tumor\" in value.lower():\n", " return None # Exclude cancer samples as we're focusing on GERD/Barrett's\n", " return None\n", "\n", "# 2.2 Age data\n", "age_row = 4 # \"subject age (years)\" row\n", "\n", "def convert_age(value):\n", " if value is None or pd.isna(value):\n", " return None\n", " if \":\" in str(value):\n", " value = str(value).split(\":\", 1)[1].strip()\n", " \n", " try:\n", " return float(value)\n", " except (ValueError, TypeError):\n", " return None\n", "\n", "# 2.3 Gender data\n", "gender_row = 3 # \"subject gender\" row\n", "\n", "def convert_gender(value):\n", " if value is None or pd.isna(value):\n", " return None\n", " if \":\" in str(value):\n", " value = str(value).split(\":\", 1)[1].strip().lower()\n", " \n", " if \"female\" in value:\n", " return 0\n", " elif \"male\" in value:\n", " return 1\n", " return None\n", "\n", "# 3. Save metadata - initial filtering\n", "is_trait_available = trait_row is not None\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 (if trait data is available)\n", "if trait_row is not None:\n", " # Create sample characteristics dictionary\n", " sample_char_dict = {\n", " 0: ['tissue type: normal esophageal squamous', \"tissue type: Barrett's esophagus (without dysplasia)\", 'tissue type: esophageal adenocarcinoma tumor'],\n", " 1: ['individual id: 53072', 'individual id: 53073', 'individual id: 54011', 'individual id: 52036', 'individual id: 53016', 'individual id: 53053', 'individual id: 53029', 'individual id: 53164', 'individual id: 52011', 'individual id: 53015', 'individual id: 54036', 'individual id: 54080', 'individual id: 52040', 'individual id: 54013', 'individual id: 53154', 'individual id: 52039', 'individual id: 54005', 'individual id: 54045', 'individual id: 54077', 'individual id: 53005', 'individual id: 53032', 'individual id: 53052', 'individual id: 54025', 'individual id: 53092', 'individual id: 53100', 'individual id: 53038', 'individual id: 53059', 'individual id: 53118', 'individual id: 53097', 'individual id: 53114'],\n", " 2: ['histology review type (see paper for details): slide', 'histology review type (see paper for details): path info'],\n", " 3: ['subject gender: female', 'subject gender: male'],\n", " 4: ['subject age (years): 73', 'subject age (years): 55', 'subject age (years): 66', 'subject age (years): 21', 'subject age (years): 48', 'subject age (years): 41', 'subject age (years): 31', 'subject age (years): 80', 'subject age (years): 45', 'subject age (years): 75', 'subject age (years): 60', 'subject age (years): 72', 'subject age (years): 56', 'subject age (years): 47', 'subject age (years): 78', 'subject age (years): 65', 'subject age (years): 68', 'subject age (years): 43', 'subject age (years): 67', 'subject age (years): 69', 'subject age (years): 57', 'subject age (years): 77', 'subject age (years): 61', 'subject age (years): 79', 'subject age (years): 70', 'subject age (years): 62', 'subject age (years): 71', 'subject age (years): 63', 'subject age (years): 52', 'subject age (years): 74'],\n", " 5: ['sample barcode: 1477791129_A', 'sample barcode: 1477791124_A', 'sample barcode: 1477791144_A', 'sample barcode: 1477791133_D', 'sample barcode: 1477791127_E', 'sample barcode: 1477791086_D', 'sample barcode: 1477791133_E', 'sample barcode: 1477791143_E', 'sample barcode: 1477791139_F', 'sample barcode: 1477791133_A', 'sample barcode: 1477791128_F', 'sample barcode: 1477791109_A', 'sample barcode: 1477791135_B', 'sample barcode: 1477791115_B', 'sample barcode: 1477791114_C', 'sample barcode: 1477791125_A', 'sample barcode: 1477791113_B', 'sample barcode: 1477791112_F', 'sample barcode: 1477791110_F', 'sample barcode: 1477791107_A', 'sample barcode: 1477791143_C', 'sample barcode: 1477791124_D', 'sample barcode: 1477791127_D', 'sample barcode: 1477791139_B', 'sample barcode: 1477791144_D', 'sample barcode: 1477791086_C', 'sample barcode: 1477791134_B', 'sample barcode: 1477791110_E', 'sample barcode: 1477791139_E', 'sample barcode: 1477791129_B'],\n", " 6: [np.nan, 'matching cn sample id: GSM265790', 'matching cn sample id: GSM266075', 'matching cn sample id: GSM265786', 'matching cn sample id: GSM265500', 'matching cn sample id: GSM265789', 'matching cn sample id: GSM266703', 'matching cn sample id: GSM266074', 'matching cn sample id: GSM266706', 'matching cn sample id: GSM265808', 'sample barcode: 1477791107_E', 'matching cn sample id: GSM265787', 'matching cn sample id: GSM266708', 'matching cn sample id: GSM266660', 'matching cn sample id: GSM265809', 'matching cn sample id: GSM266705', 'matching cn sample id: GSM266707', 'matching cn sample id: GSM266119', 'matching cn sample id: GSM265501', 'matching cn sample id: GSM265791', 'matching cn sample id: GSM266715', 'matching cn sample id: GSM266659', 'matching cn sample\n" ] }, { "cell_type": "markdown", "id": "5351b418", "metadata": {}, "source": [ "### Step 3: Dataset Analysis and Clinical Feature Extraction" ] }, { "cell_type": "code", "execution_count": null, "id": "697c70af", "metadata": {}, "outputs": [], "source": [ "```python\n", "# Import necessary libraries\n", "import pandas as pd\n", "import os\n", "import numpy as np\n", "import json\n", "from typing import Optional, Callable, Dict, Any, List, Union\n", "\n", "# -------- 1. First, let's check what files are available in the input directory --------\n", "print(f\"Checking files in: {in_cohort_dir}\")\n", "available_files = os.listdir(in_cohort_dir)\n", "print(\"Available files:\", available_files)\n", "\n", "# Look for appropriate files that might contain sample characteristics\n", "potential_clinical_files = [f for f in available_files if 'clinical' in f.lower() or 'sample' in f.lower() or '.soft' in f.lower()]\n", "print(\"Potential clinical files:\", potential_clinical_files)\n", "\n", "# If we find a SOFT file, let's use that as it contains complete information\n", "soft_files = [f for f in available_files if f.endswith('.soft')]\n", "if soft_files:\n", " soft_file = os.path.join(in_cohort_dir, soft_files[0])\n", " print(f\"Using SOFT file: {soft_file}\")\n", " \n", " # Reading the SOFT file to extract sample characteristics\n", " with open(soft_file, 'r') as f:\n", " lines = f.readlines()\n", " \n", " # Extract sample characteristics\n", " sample_data = {}\n", " current_sample = None\n", " \n", " for line in lines:\n", " line = line.strip()\n", " if line.startswith(\"^SAMPLE\"):\n", " parts = line.split(\" = \")\n", " if len(parts) > 1:\n", " current_sample = parts[1]\n", " sample_data[current_sample] = {}\n", " elif line.startswith(\"!Sample_\") and current_sample is not None:\n", " parts = line.split(\" = \")\n", " if len(parts) > 1:\n", " key = parts[0].replace(\"!Sample_\", \"\")\n", " value = parts[1]\n", " if key not in sample_data[current_sample]:\n", " sample_data[current_sample][key] = value\n", " else:\n", " if not isinstance(sample_data[current_sample][key], list):\n", " sample_data[current_sample][key] = [sample_data[current_sample][key]]\n", " sample_data[current_sample][key].append(value)\n", " \n", " # Convert to DataFrame\n", " all_keys = set()\n", " for sample_dict in sample_data.values():\n", " all_keys.update(sample_dict.keys())\n", " \n", " clinical_data = pd.DataFrame(index=list(sample_data.keys()), columns=list(all_keys))\n", " for sample, sample_dict in sample_data.items():\n", " for key, value in sample_dict.items():\n", " clinical_data.loc[sample, key] = value\n", " \n", " # Transpose to have characteristics as rows\n", " clinical_data = clinical_data.transpose()\n", "\n", "else:\n", " # If no SOFT file, try to find a matrix file\n", " matrix_files = [f for f in available_files if 'matrix' in f.lower() or 'series_matrix' in f.lower()]\n", " if matrix_files:\n", " matrix_file = os.path.join(in_cohort_dir, matrix_files[0])\n", " print(f\"Using matrix file: {matrix_file}\")\n", " \n", " # Read the file to extract sample characteristics\n", " with open(matrix_file, 'r') as f:\n", " lines = f.readlines()\n", " \n", " # Extract sample characteristics\n", " sample_data = {}\n", " sample_ids = []\n", " \n", " for line in lines:\n", " line = line.strip()\n", " if line.startswith(\"!Sample_\"):\n", " parts = line.split(\"\\t\")\n", " key = parts[0].replace(\"!Sample_\", \"\")\n", " values = parts[1:]\n", " \n", " if key == \"geo_accession\":\n", " sample_ids = values\n", " for sample_id in sample_ids:\n", " sample_data[sample_id] = {}\n", " \n", " if sample_ids:\n", " for i, sample_id in enumerate(sample_ids):\n", " if i < len(values):\n", " sample_data[sample_id][key] = values[i]\n", " \n", " # Convert to DataFrame\n", " all_keys = set()\n", " for sample_dict in sample_data.values():\n", " all_keys.update(sample_dict.keys())\n", " \n", " clinical_data = pd.DataFrame(index=list(sample_data.keys()), columns=list(all_keys))\n", " for sample, sample_dict in sample_data.items():\n", " for key, value in sample_dict.items():\n", " clinical_data.loc[sample, key] = value\n", " \n", " # Transpose to have characteristics as rows\n", " clinical_data = clinical_data.transpose()\n", " \n", " else:\n", " # If no suitable files found, create a dummy DataFrame and mark data as unavailable\n", " print(\"No suitable files found for clinical data.\")\n", " clinical_data = pd.DataFrame()\n", " is_gene_available = False\n", " is_trait_available = False\n", " \n", " # Save metadata indicating data is not available\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", " # Exit early\n", " print(f\"Data not available for {cohort}. Metadata saved.\")\n", " exit()\n", "\n", "# Display the clinical data\n", "print(\"Preview of clinical data:\")\n", "print(clinical_data.head())\n", "\n", "# -------- 2. Check the unique values in each row to identify relevant information --------\n", "unique_values = {}\n", "for i in range(len(clinical_data.index)):\n", " row_name = clinical_data.index[i]\n", " values = clinical_data.iloc[i].unique()\n", " unique_values[i] = {\n", " \"name\": row_name,\n", " \"values\": values,\n", " \"count\": len(values)\n", " }\n", " print(f\"Row {i} - {row_name}: {values}\")\n", "\n", "# -------- 3. Determine availability and conversion functions based on the data --------\n", "\n", "# 3.1 Check if gene expression data is available\n", "# Look for platform information that suggests gene expression\n", "is_gene_available = True\n", "platform_rows = [i for i, info in unique_values.items() if \"platform\" in str(info[\"name\"]).lower()]\n", "if platform_rows:\n", " platform_values = [str(v).lower() for v in unique_values[platform_rows[0]][\"values\"]]\n", " # If platform indicates miRNA or methylation, mark gene data as unavailable\n", " if any((\"mirna\" in v or \"methylation\" in v) for v in platform_values):\n", " is_gene_available = False\n", "\n", "# 3.2 Identify the row indices for trait, age, and gender\n", "\n", "# For GERD (Gastroesophageal reflux disease)\n", "trait_row = None\n", "for i, info in unique_values.items():\n", " row_name = str(info[\"name\"]).lower()\n", " values = [str(v).lower() for v in info[\"values\"]]\n", " \n", " # Look for rows that might contain GERD information\n", " if (\"gerd\" in row_name or \"reflux\" in row_name or \"disease\" in row_name or \n", " \"diagnosis\" in row_name or \"condition\" in row_name or \"group\" in row_name):\n", " if any((\"gerd\" in v or \"reflux\" in v or \"control\" in v or \"normal\" in v or \"disease\" in v) for v in values):\n", " trait_row = i\n", " break\n", "\n", "# If we couldn't find a direct trait row, check if we can infer from sample descriptions\n", "if trait_row is None:\n", " for i, info in unique_values.items():\n", " row_name = str(info[\"name\"]).lower()\n", " values = [str(v).lower() for v in info[\"values\"]]\n", " \n", " if (\"title\" in row_name or \"description\" in row_name or \"characteristic\" in row_name):\n", " if any((\"gerd\" in v or \"reflux\" in v or \"control\" in v or \"normal\" in v) for v in values):\n", " trait_row = i\n", " break\n", "\n", "# For age\n", "age_row = None\n", "for i, info in unique_values.items():\n", " row_name = str(\n" ] }, { "cell_type": "markdown", "id": "6556c5ab", "metadata": {}, "source": [ "### Step 4: Gene Data Extraction" ] }, { "cell_type": "code", "execution_count": null, "id": "c1e02fd3", "metadata": {}, "outputs": [], "source": [ "# 1. Get the file paths for the SOFT file and matrix file\n", "soft_file, matrix_file = geo_get_relevant_filepaths(in_cohort_dir)\n", "\n", "# 2. First, let's examine the structure of the matrix file to understand its format\n", "import gzip\n", "\n", "# Peek at the first few lines of the file to understand its structure\n", "with gzip.open(matrix_file, 'rt') as file:\n", " # Read first 100 lines to find the header structure\n", " for i, line in enumerate(file):\n", " if '!series_matrix_table_begin' in line:\n", " print(f\"Found data marker at line {i}\")\n", " # Read the next line which should be the header\n", " header_line = next(file)\n", " print(f\"Header line: {header_line.strip()}\")\n", " # And the first data line\n", " first_data_line = next(file)\n", " print(f\"First data line: {first_data_line.strip()}\")\n", " break\n", " if i > 100: # Limit search to first 100 lines\n", " print(\"Matrix table marker not found in first 100 lines\")\n", " break\n", "\n", "# 3. Now try to get the genetic data with better error handling\n", "try:\n", " gene_data = get_genetic_data(matrix_file)\n", " print(gene_data.index[:20])\n", "except KeyError as e:\n", " print(f\"KeyError: {e}\")\n", " \n", " # Alternative approach: manually extract the data\n", " print(\"\\nTrying alternative approach to read the gene data:\")\n", " with gzip.open(matrix_file, 'rt') as file:\n", " # Find the start of the data\n", " for line in file:\n", " if '!series_matrix_table_begin' in line:\n", " break\n", " \n", " # Read the headers and data\n", " import pandas as pd\n", " df = pd.read_csv(file, sep='\\t', index_col=0)\n", " print(f\"Column names: {df.columns[:5]}\")\n", " print(f\"First 20 row IDs: {df.index[:20]}\")\n", " gene_data = df\n" ] }, { "cell_type": "markdown", "id": "7a3810b0", "metadata": {}, "source": [ "### Step 5: Gene Identifier Review" ] }, { "cell_type": "code", "execution_count": null, "id": "8035056c", "metadata": {}, "outputs": [], "source": [ "# Let's analyze the gene identifiers in the gene expression data\n", "\n", "# The identifiers appear to start with 'GI_' followed by a number and a suffix like '-S', '-A', or '-I'\n", "# These are not standard human gene symbols (like BRCA1, TP53, etc.)\n", "# These look like GenBank or other database IDs that need to be mapped to standard gene symbols\n", "\n", "# Looking at examples like:\n", "# GI_10047089-S\n", "# GI_10047091-S \n", "# These appear to be GenInfo Identifiers (GI numbers) which were used by NCBI\n", "\n", "# Conclusion based on biomedical knowledge:\n", "requires_gene_mapping = True\n" ] }, { "cell_type": "markdown", "id": "9506b260", "metadata": {}, "source": [ "### Step 6: Gene Annotation" ] }, { "cell_type": "code", "execution_count": null, "id": "d6d76448", "metadata": {}, "outputs": [], "source": [ "# 1. Let's first examine the structure of the SOFT file before trying to parse it\n", "import gzip\n", "\n", "# Look at the first few lines of the SOFT file to understand its structure\n", "print(\"Examining SOFT file structure:\")\n", "try:\n", " with gzip.open(soft_file, 'rt') as file:\n", " # Read first 20 lines to understand the file structure\n", " for i, line in enumerate(file):\n", " if i < 20:\n", " print(f\"Line {i}: {line.strip()}\")\n", " else:\n", " break\n", "except Exception as e:\n", " print(f\"Error reading SOFT file: {e}\")\n", "\n", "# 2. Now let's try a more robust approach to extract the gene annotation\n", "# Instead of using the library function which failed, we'll implement a custom approach\n", "try:\n", " # First, look for the platform section which contains gene annotation\n", " platform_data = []\n", " with gzip.open(soft_file, 'rt') as file:\n", " in_platform_section = False\n", " for line in file:\n", " if line.startswith('^PLATFORM'):\n", " in_platform_section = True\n", " continue\n", " if in_platform_section and line.startswith('!platform_table_begin'):\n", " # Next line should be the header\n", " header = next(file).strip()\n", " platform_data.append(header)\n", " # Read until the end of the platform table\n", " for table_line in file:\n", " if table_line.startswith('!platform_table_end'):\n", " break\n", " platform_data.append(table_line.strip())\n", " break\n", " \n", " # If we found platform data, convert it to a DataFrame\n", " if platform_data:\n", " import pandas as pd\n", " import io\n", " platform_text = '\\n'.join(platform_data)\n", " gene_annotation = pd.read_csv(io.StringIO(platform_text), delimiter='\\t', \n", " low_memory=False, on_bad_lines='skip')\n", " print(\"\\nGene annotation preview:\")\n", " print(preview_df(gene_annotation))\n", " else:\n", " print(\"Could not find platform table in SOFT file\")\n", " \n", " # Try an alternative approach - extract mapping from other sections\n", " with gzip.open(soft_file, 'rt') as file:\n", " for line in file:\n", " if 'ANNOTATION information' in line or 'annotation information' in line:\n", " print(f\"Found annotation information: {line.strip()}\")\n", " if line.startswith('!Platform_title') or line.startswith('!platform_title'):\n", " print(f\"Platform title: {line.strip()}\")\n", " \n", "except Exception as e:\n", " print(f\"Error processing gene annotation: {e}\")\n" ] }, { "cell_type": "markdown", "id": "c5392219", "metadata": {}, "source": [ "### Step 7: Gene Identifier Mapping" ] }, { "cell_type": "code", "execution_count": null, "id": "9c46f692", "metadata": {}, "outputs": [], "source": [ "# 1. Get gene annotation data from the SOFT file (need to re-extract it)\n", "gene_annotation = get_gene_annotation(soft_file)\n", "print(f\"Gene annotation shape: {gene_annotation.shape}\")\n", "print(\"Gene annotation columns:\")\n", "print(gene_annotation.columns.tolist())\n", "print(\"Sample of gene annotation data:\")\n", "print(gene_annotation.head())\n", "\n", "# 2. Create a gene mapping dataframe using the ID and GB_ACC columns\n", "# ID column contains the same identifiers as in the gene expression data\n", "# GB_ACC contains RefSeq accessions which we'll use for gene mapping\n", "mapping_df = pd.DataFrame({\n", " 'ID': gene_annotation['ID'],\n", " 'Gene': gene_annotation['GB_ACC']\n", "})\n", "mapping_df = mapping_df.dropna(subset=['Gene']) # Remove rows with missing gene information\n", "print(f\"Mapping dataframe shape: {mapping_df.shape}\")\n", "print(\"First few rows of mapping dataframe:\")\n", "print(mapping_df.head())\n", "\n", "# 3. Define a custom function to extract gene identifiers from RefSeq accessions\n", "# This is needed because the standard extract_human_gene_symbols function \n", "# wouldn't work well with RefSeq IDs\n", "def extract_gene_from_refseq(refseq_id):\n", " \"\"\"Extract a gene identifier from RefSeq accession without filtering\"\"\"\n", " if not isinstance(refseq_id, str):\n", " return []\n", " \n", " # For RefSeq accessions, return the accession without version number\n", " if refseq_id.startswith('NM_') or refseq_id.startswith('NR_') or refseq_id.startswith('XM_'):\n", " # Remove version number if present (e.g., NM_001234.2 -> NM_001234)\n", " base_id = refseq_id.split('.')[0]\n", " return [base_id]\n", " \n", " return []\n", "\n", "# 4. Apply custom mapping to convert probe-level data to gene-level data\n", "# First, modify the mapping DataFrame to use our custom extraction function\n", "mapping_df['Gene'] = mapping_df['Gene'].apply(extract_gene_from_refseq)\n", "mapping_df['num_genes'] = mapping_df['Gene'].apply(len)\n", "mapping_df = mapping_df.explode('Gene')\n", "mapping_df = mapping_df.dropna(subset=['Gene'])\n", "\n", "if mapping_df.empty:\n", " print(\"WARNING: No valid mappings found between probes and genes\")\n", " # Use probe IDs directly as fallback\n", " gene_data_mapped = gene_data.copy()\n", " gene_data_mapped.index = gene_data_mapped.index.map(lambda x: x.split('-')[0] if '-' in x else x)\n", "else:\n", " # Set ID as index to allow joining with expression data\n", " mapping_df.set_index('ID', inplace=True)\n", " \n", " # Get expression columns (exclude ID column if it exists)\n", " expr_cols = [col for col in gene_data.columns if col != 'ID']\n", " \n", " # Join mapping with expression data\n", " merged_df = mapping_df.join(gene_data)\n", " \n", " # Distribute expression values based on number of genes per probe\n", " merged_df[expr_cols] = merged_df[expr_cols].div(merged_df['num_genes'].replace(0, 1), axis=0)\n", " \n", " # Sum expression values for each gene\n", " gene_data_mapped = merged_df.groupby('Gene')[expr_cols].sum()\n", "\n", "# Use the mapped data\n", "gene_data = gene_data_mapped\n", "\n", "print(f\"Gene data shape after mapping: {gene_data.shape}\")\n", "print(\"First few mapped genes:\")\n", "print(gene_data.index[:10].tolist())\n", "\n", "# Save a sample of the gene expression data to observe its structure\n", "print(\"Sample of gene expression data:\")\n", "print(gene_data.iloc[:5, :5] if not gene_data.empty else \"No gene data available\")\n" ] }, { "cell_type": "markdown", "id": "83944c46", "metadata": {}, "source": [ "### Step 8: Data Normalization and Linking" ] }, { "cell_type": "code", "execution_count": null, "id": "065687c9", "metadata": {}, "outputs": [], "source": [ "# 1. Load the gene expression data saved in step 6\n", "gene_data = pd.read_csv(out_gene_data_file, index_col=0)\n", "print(f\"Loaded gene data shape: {gene_data.shape}\")\n", "\n", "# Normalize gene symbols using NCBI Gene database\n", "normalized_gene_data = normalize_gene_symbols_in_index(gene_data)\n", "print(f\"Gene data shape after normalization: {normalized_gene_data.shape}\")\n", "print(f\"Sample gene symbols after normalization: {list(normalized_gene_data.index[:10])}\")\n", "\n", "# Save the normalized gene data (overwrite the previous file with normalized data)\n", "os.makedirs(os.path.dirname(out_gene_data_file), exist_ok=True)\n", "normalized_gene_data.to_csv(out_gene_data_file)\n", "print(f\"Normalized gene data saved to {out_gene_data_file}\")\n", "\n", "# 2. Load the clinical data created in step 2\n", "clinical_df = pd.read_csv(out_clinical_data_file)\n", "print(f\"Loaded clinical data shape: {clinical_df.shape}\")\n", "\n", "# If clinical_df doesn't have a proper index, fix it\n", "if 'Unnamed: 0' in clinical_df.columns:\n", " clinical_df = clinical_df.set_index('Unnamed: 0')\n", "elif not clinical_df.index.name:\n", " # Just in case the index needs to be set from data\n", " clinical_features = 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 if age_row is not None else None,\n", " gender_row=gender_row,\n", " convert_gender=convert_gender if gender_row is not None else None\n", " )\n", " clinical_df = clinical_features\n", "\n", "# Link clinical and genetic data\n", "linked_data = geo_link_clinical_genetic_data(clinical_df, normalized_gene_data)\n", "print(f\"Linked data shape: {linked_data.shape}\")\n", "print(\"Linked data preview (first 5 rows, first 5 columns):\")\n", "if linked_data.shape[1] >= 5:\n", " print(linked_data.iloc[:5, :5])\n", "else:\n", " print(linked_data.head())\n", "\n", "# 3. Handle missing values\n", "print(\"\\nMissing values before handling:\")\n", "print(f\" Trait ({trait}) missing: {linked_data[trait].isna().sum()} out of {len(linked_data)}\")\n", "if 'Age' in linked_data.columns:\n", " print(f\" Age missing: {linked_data['Age'].isna().sum()} out of {len(linked_data)}\")\n", "if 'Gender' in linked_data.columns:\n", " print(f\" Gender missing: {linked_data['Gender'].isna().sum()} out of {len(linked_data)}\")\n", "\n", "gene_cols = [col for col in linked_data.columns if col not in [trait, 'Age', 'Gender']]\n", "if gene_cols:\n", " print(f\" Genes with >20% missing: {sum(linked_data[gene_cols].isna().mean() > 0.2)}\")\n", " print(f\" Samples with >5% missing genes: {sum(linked_data[gene_cols].isna().mean(axis=1) > 0.05)}\")\n", "\n", "cleaned_data = handle_missing_values(linked_data, trait)\n", "print(f\"Data shape after handling missing values: {cleaned_data.shape}\")\n", "\n", "# 4. Evaluate bias in trait and demographic features\n", "is_trait_biased = False\n", "if len(cleaned_data) > 0:\n", " trait_biased, cleaned_data = judge_and_remove_biased_features(cleaned_data, trait)\n", " is_trait_biased = trait_biased\n", "else:\n", " print(\"No data remains after handling missing values.\")\n", " is_trait_biased = True\n", "\n", "# 5. Final validation and save\n", "is_usable = validate_and_save_cohort_info(\n", " is_final=True, \n", " cohort=cohort, \n", " info_path=json_path, \n", " is_gene_available=len(normalized_gene_data) > 0, \n", " is_trait_available=True, \n", " is_biased=is_trait_biased, \n", " df=cleaned_data,\n", " note=f\"Dataset contains gene expression data for {trait} analysis.\"\n", ")\n", "\n", "# 6. Save if usable\n", "if is_usable and len(cleaned_data) > 0:\n", " os.makedirs(os.path.dirname(out_data_file), exist_ok=True)\n", " cleaned_data.to_csv(out_data_file)\n", " print(f\"Linked data saved to {out_data_file}\")\n", "else:\n", " print(\"Data was determined to be unusable or empty and was not saved\")" ] } ], "metadata": {}, "nbformat": 4, "nbformat_minor": 5 }