Module Gene Overlap Analysis

A web application for analyzing gene overlaps between different modules across datasets. This tool enables researchers to identify statistically significant overlaps between gene modules from their own data and reference datasets.

Website:https://overlapmodule.onrender.com

Features

• Upload your own gene module data in CSV or Excel format
• Compare against built-in reference datasets
• Perform statistical testing using Fisher's Exact Test or Hypergeometric Test
• Visualize overlaps with interactive heatmaps
• Review significant overlaps with adjustable significance thresholds
• Download lists of overlapping genes for further analysis
• Apply Benjamini-Hochberg correction for multiple testing

Installation

1. Clone this repository:

   git clone https://github.com/HuihongLi/OverlapModule.git
   cd OverlapModule
   

2. Create a virtual environment (recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   

3. Install required packages:

   pip install -r requirements.txt
   

4. Create the data directory:

   mkdir -p data
   

5. Add your reference datasets (optional):

   • Place an Excel file at data/data.xlsx with reference datasets in separate sheets
   • Each dataset should have at least the columns 'Gene' and 'Module'

Usage

1. Start the application:

   python server.py
   

2. Open a web browser and navigate to:

   http://127.0.0.1:8050/
   

3. Upload your dataset:

   • Your file must contain at least the columns 'Gene' and 'Module'
   • Both Excel (.xlsx) and CSV (.csv) formats are supported
   • Use the provided example file as a reference for formatting

4. Configure analysis parameters:

   • Select a reference dataset to compare against
   • Adjust the total gene universe size (default: 20,000)
   • Set the significance threshold (default: 0.05)
   • Choose minimum overlap count (default: 5)
   • Select the statistical test method

5. Run the analysis and view results:

   • Interactive heatmap shows overlap significance (-log10 of adjusted p-value)
   • Table of significant overlaps with key statistics
   • Click on any overlap to view and download detailed gene lists

Input File Format

Your input file should contain at least the following columns:

• Gene: Gene identifiers (symbols or IDs)
• Module: Module/cluster identifiers for each gene

Example:

Gene,Module
BRCA1,module1
TP53,module1
MYC,module2
PTEN,module2

Statistical Methods

The application offers two statistical approaches:

1. Fisher's Exact Test: Tests the independence of row and column variables in a contingency table.

2. Hypergeometric Test: Calculates the probability of finding k or more overlapping genes between two sets.

Both methods are adjusted for multiple testing using the Benjamini-Hochberg procedure to control the false discovery rate.

Directory Structure

OverlapModule/
├── server.py               # Main application code
├── data/                # Directory for datasets
│   ├── data.xlsx        # Built-in reference datasets
│   └── example.csv      # Example dataset for users
├── requirements.txt     # Python dependencies
└── README.md            # This documentation file

Requirements

• Python 3.7+
• See requirements.txt for Python package dependencies

License

MIT License

Citation

If you use this tool in your research, please cite:

Li, Huihong, et al. "Shared Transcriptomic Signatures Reveal Synaptic Pruning as a Link Between Alzheimer’s Disease and Epilepsy." bioRxiv (2024): 2024-10.