Variant Calling Pipeline

A Nextflow pipeline for variant calling between two complete genomes using MUMmer or Minimap2.

Features

• SNP and small indel detection
• Optional structural variant (SV) detection
• Support for both MUMmer and Minimap2 aligners
• Standardized VCF output
• Automatic sorting and normalization of variants

Prerequisites

• Nextflow
• Docker
• Python 3.9+

Installation

# Clone the repository
git clone https://github.com/yourusername/variant-caller.git
cd variant-caller

# Build Docker image
docker build -t macbio/variant-caller:0.1.0 -f docker/Dockerfile .

Usage

Basic SNP calling:

nextflow run main.nf \
  --reference ref.fa \
  --query query.fa

Include structural variant detection:

nextflow run main.nf \
  --reference ref.fa \
  --query query.fa \
  --detect_sv

Use Minimap2 instead of MUMmer:

nextflow run main.nf \
  --reference ref.fa \
  --query query.fa \
  --aligner minimap2

Parameters

Required Arguments

• --reference: Path to reference genome (FASTA format)
• --query: Path to query genome (FASTA format)

Optional Arguments

• --aligner: Choice of alignment tool (mummer/minimap2) [default: mummer]
• --outdir: Output directory [default: results]
• --detect_sv: Enable structural variant detection [default: false]

SV Detection Options

These parameters are only used when --detect_sv is enabled:

• --min_sv_size: Minimum size for structural variants [default: 50]
• --max_sv_size: Maximum size for structural variants [default: 100000]

Output

The pipeline generates:

• Alignment files (delta/PAF format)
• Variant calls in VCF format
• Normalized and sorted final VCF
• Optional structural variant calls

Docker Containers

The pipeline uses the following Docker containers:

• macbio/mummer-arm64: MUMmer aligner
• macbio/minimap2-arm64: Minimap2 aligner
• macbio/variant-caller: Custom variant processing tools

Known Issues

• All sequence files need to be within the './genomes' subdirectory
• Compressed fasta files (BGZIP or other) are problematic

License

MIT