Skip to content

Installation

Quick Install

pip install py-gbcms
gbcms --version

System Requirements

PyPI wheels require glibc 2.34+ (Ubuntu 22.04+, RHEL 9+, Debian 12+). For older systems, see Legacy Linux.

docker pull ghcr.io/msk-access/py-gbcms:2.3.0
docker run --rm ghcr.io/msk-access/py-gbcms:2.3.0 gbcms --help

git clone https://github.com/msk-access/py-gbcms.git
cd py-gbcms
pip install -e ".[dev]"

Requirements

Component Requirement
Python 3.10+
OS Linux (glibc 2.34+), macOS, Windows (WSL2)
Memory 4GB+ (8GB for large BAMs)

For Nextflow Workflow

  • Nextflow 21.10.3+
  • Docker or Singularity

Legacy Linux (RHEL 8)

For RHEL 8, CentOS 8, or other systems with glibc < 2.34:

# Create conda environment with dependencies
conda create -n gbcms python=3.11 openssl rust clang cmake -c conda-forge
conda activate gbcms

# Install from source
git clone https://github.com/msk-access/py-gbcms.git
cd py-gbcms
pip install .

docker pull ghcr.io/msk-access/py-gbcms:2.3.0
docker run --rm -v $(pwd):/data ghcr.io/msk-access/py-gbcms:2.3.0 gbcms --help

singularity pull docker://ghcr.io/msk-access/py-gbcms:2.3.0
singularity exec py-gbcms_2.3.0.sif gbcms --help

Why not pip install?

The PyPI wheels require glibc 2.34+. On RHEL 8 (glibc 2.28), pip would fall back to source compilation which requires Rust and OpenSSL 3.0 dev libraries. Using conda provides these dependencies automatically.

Verification

# Check installation
gbcms --version
# Expected: 2.3.0

# Test help
gbcms --help

Docker Usage

docker run --rm \
  -v $(pwd):/data \
  ghcr.io/msk-access/py-gbcms:2.3.0 \
  gbcms run \
    --variants /data/variants.vcf \
    --bam /data/sample.bam \
    --fasta /data/reference.fa \
    --output-dir /data/results/

Docker Volume

Use -v to mount your data directory.

Troubleshooting

Module Not Found

pip uninstall py-gbcms && pip install py-gbcms

BAM Index Missing

samtools index sample.bam

Docker Permission Denied

sudo usermod -aG docker $USER && newgrp docker

glibc Version Error

If you see GLIBC_2.34 not found, use the Legacy Linux instructions.

Upgrade

pip install --upgrade py-gbcms

Next Steps