Genetic Variant Classifications

A machine learning project for classifying genetic variants based on ClinVar data.

📋 Description

This project analyzes conflicting genetic variants from the ClinVar database and classifies them as Benign or Pathogenic/Likely Pathogenic using machine learning algorithms.

🎯 Key Features

• Loading and preprocessing ClinVar data
• Classification of genetic variants (Benign vs Pathogenic)
• Feature importance analysis for predictions
• Results visualization

📊 Data

The project uses the clinvar_conflicting.csv dataset, which contains information about genetic variants with conflicting interpretations of clinical significance.

Main Features:

• Genomic coordinates: CHROM, POS, REF, ALT
• Allele frequencies: AF_ESP, AF_EXAC, AF_TGP
• Clinical information: CLNSIGINCL, CLNDN, CLNHGVS
• Functional predictions: SIFT, PolyPhen, CADD_PHRED, BLOSUM62
• Annotations: Consequence, IMPACT, SYMBOL, BIOTYPE

🔬 Methodology

1. Data Preprocessing:

   • Simplifying classification labels from CLNSIGINCL
   • Filtering ambiguous variants
   • Binary classification: 0 (Benign) and 1 (Pathogenic/Likely Pathogenic)

2. Machine Learning:

   • Random Forest Classifier implementation
   • Feature importance analysis for model interpretation

📈 Results

The chart shows the importance of various features for genetic variant classification.

🛠️ Technologies

• Python 3.12
• pandas - data processing
• scikit-learn - machine learning
• matplotlib - visualization
• numpy - numerical computations

🚀 Installation and Usage

1. Clone the repository:

git clone https://github.com/3x6dll9ff/Genetic_Variant__Classifications.git
cd Genetic_Variant__Classifications

2. Create a virtual environment:

python3.12 -m venv .venv
source .venv/bin/activate

3. Install dependencies:

pip install pandas scikit-learn matplotlib numpy seaborn jupyter

4. Launch Jupyter Notebook:

jupyter notebook "Genetic_Variant_ Classifications.ipynb"

📝 Project Structure

.
├── Genetic_Variant_ Classifications.ipynb  # Main analysis notebook
├── clinvar_conflicting.csv                 # ClinVar dataset
├── image.png                               # Feature importance visualization
├── .gitignore                              # Git ignore file
└── README.md                               # This file

📚 Data Sources

• ClinVar Database - database of genetic variants and their clinical significance

👨‍💻 Author

Danila Kardashevskii

📄 License

MIT License