py-gbcms

Get Base Counts Multi-Sample — High-performance variant counting from BAM files

What It Does

GBCMS extracts allele counts and variant metrics at specified positions in BAM files:

flowchart LR
    subgraph Input
        VCF[VCF/MAF]
        BAM[BAM Files]
    end
    
    subgraph Output
        Counts[Allele Counts]
        Metrics[VAF, Strand Bias]
    end
    
    VCF --> Engine[gbcms]
    BAM --> Engine
    Engine --> Counts
    Engine --> Metrics

Key Metrics

Metric

Formula

Description

VAF

AD / (RD + AD)

Variant Allele Frequency

Strand Bias

Fisher's exact test

Detect sequencing artifacts

Fragment Counts

Deduplicated pairs

PCR-aware counting

Quick Start

# Install
pip install py-gbcms

# Run
gbcms run --variants variants.vcf --bam sample.bam --fasta ref.fa --output-dir results/

→ Full Installation Guide | → CLI Examples

Choose Your Workflow

flowchart TD
    Start[How many samples?] --> Few{1-10 samples}
    Start --> Many{10+ samples}
    
    Few --> CLI[Use CLI<br/>gbcms run ...]
    Many --> HPC{HPC cluster?}
    
    HPC --> |Yes| Nextflow[Use Nextflow<br/>nextflow run ...]
    HPC --> |No| CLI
    
    style CLI fill:#4caf50,color:white
    style Nextflow fill:#2196f3,color:white

Workflow

Best For

Guide

CLI

1-10 samples, local/single server

Quick Start

Nextflow

10+ samples, HPC/SLURM

Nextflow Guide

Architecture

Python/Rust hybrid for maximum performance:

flowchart TB
    subgraph Python["🐍 Python"]
        CLI[CLI] --> Pipeline[Orchestration]
        Pipeline --> IO[VCF/MAF I/O]
    end
    
    subgraph Rust["🦀 Rust"]
        Counter[BAM Counting]
        Stats[Fisher's Test]
    end
    
    Pipeline --> Counter
    Counter --> Stats
    
    style Python fill:#3776ab,color:#fff
    style Rust fill:#dea584,color:#000