Skip to content

README.md

Latest commit

 

History

History
115 lines (83 loc) · 3.37 KB

README.md

File metadata and controls

115 lines (83 loc) · 3.37 KB

Variant Effect Predictor

A tool for predicting the effects of genomic variants on protein-coding genes, focusing on synonymous and nonsynonymous changes.

Overview

This tool analyzes Variant Call Format (VCF) files to determine the effects of SNPs (Single Nucleotide Polymorphisms) on protein-coding genes. It uses a reference genome (FASTA) and gene annotations (GenBank) to:

  1. Identify if variants fall within coding regions
  2. Determine if coding variants cause amino acid changes
  3. Classify variants as:
    • Synonymous (no amino acid change)
    • Nonsynonymous (amino acid change)
    • Stop gain (introducing a premature stop codon)
    • Stop loss (removing a stop codon)
    • Intergenic (not in a coding region)

Requirements

  • Python 3.6+
  • BioPython
  • pandas (for batch processing)
  • matplotlib and seaborn (for visualization)

Install dependencies:

pip install biopython pandas matplotlib seaborn
# use conda instead

Usage

Processing a Single VCF File

python vep.py INPUT.vcf REFERENCE.fasta ANNOTATION.gbk -o OUTPUT.tsv

Arguments:

  • INPUT.vcf: Input variant file in VCF format
  • REFERENCE.fasta: Reference genome in FASTA format
  • ANNOTATION.gbk: Gene annotation file in GenBank format
  • -o, --output: Output file name (default: variant_effects.tsv)

Processing Multiple VCF Files

python batch.py --vcf_dir /path/to/vcfs  --pattern "*.vcf.gz" --fasta REFERENCE.fasta --genbank ANNOTATION.gbk --output_dir results

Arguments:

  • --vcf_dir: Directory containing VCF files
  • --fasta: Reference genome in FASTA format
  • --genbank: Gene annotation file in GenBank format
  • --output_dir: Output directory (default: variant_results)
  • --pattern: Pattern to match VCF files (default: *.vcf)

Output Format

The tool generates a tab-delimited file with the following columns:

  • CHROM: Chromosome/contig name
  • POS: Variant position
  • REF: Reference base
  • ALT: Alternate base
  • EFFECT: Effect type (synonymous, nonsynonymous, stop_gain, stop_loss, intergenic)
  • GENE: Gene name
  • PRODUCT: Gene product description
  • REF_CODON: Reference codon
  • ALT_CODON: Alternate codon
  • REF_AA: Reference amino acid
  • ALT_AA: Alternate amino acid
  • CODON_POS: Position in the codon (1-based)
  • AA_POS: Amino acid position (1-based)

Batch Processing Output

The batch processor generates:

  • Individual TSV files for each processed VCF
  • A combined results file (combined_effects.tsv)
  • A summary text file with statistics (variant_summary.txt)
  • Visualization plots in the "plots" directory:
    • Distribution of variant effects
    • Variant effects by sample
    • Top affected genes

Examples

Using the provided test data:

# Process single file
python vep.py test1.vcf data/KX894508/KX894508.1_155920_2012_Israel_19_Dec_2012.fna data/KX894508/genes.gbk

# Process multiple files
python batch.py --vcf_dir GIRAFFE_1 --fasta data/KX894508/KX894508.1_155920_2012_Israel_19_Dec_2012.fna --genbank data/KX894508/genes.gbk

Testing with KX894508

The provided test case uses:

  • The KX894508 genome (Lumpy skin disease virus isolate)
  • Gene annotations from genes.gbk
  • Variants from test1.vcf

Limitations

  • This tool currently only handles simple SNPs (no indels or complex variants)
  • Focuses on protein-coding regions (doesn't analyze UTRs, introns, etc.)
  • Assumes the GBK and FASTA files are properly aligned