Parabricks

Introduction

NVIDIA’s Clara Parabricks brings next generation sequencing to GPUs, accelerating an array of gold-standard tooling such as BWA-MEM, GATK4, Google’s DeepVariant, and many more. Users can achieve a 30-60x acceleration and 99.99% accuracy for variant calling when comparing against CPU-only BWA-GATK4 pipelines, meaning a single server can process up to 60 whole genomes per day. These tools can be easily integrated into current pipelines with drop-in replacement commands to quickly bring speed and data-center scale to a range of applications including germline, somatic and RNA workflows.

Versions

• 4.0.0-1

• 4.2.1-1

Commands

Example job

Adjust slurm options based on job requirements (slurm cheat sheet):

#!/bin/bash
#SBATCH -p partitionName  # batch, gpu, preempt, mpi or your group's own partition
#SBATCH -t 1:00:00  # Runtime limit (D-HH:MM:SS)
#SBATCH -N 1   # Number of nodes
#SBATCH -n 1   # Number of tasks per node
#SBATCH -c 4   # Number of CPU cores per task
#SBATCH --mem=8G       # Memory required per node
#SBATCH --job-name=parabricks  # Job name
#SBATCH --mail-type=FAIL,BEGIN,END     # Send an email when job fails, begins, and finishes
#SBATCH --mail-user=your.email@tufts.edu       # Email address for notifications
#SBATCH --error=%x-%J-%u.err   # Standard error file: <job_name>-<job_id>-<username>.err
#SBATCH --output=%x-%J-%u.out  # Standard output file: <job_name>-<job_id>-<username>.out

module purge   ### Optional, but highly recommended.
module load parabricks/XXXX    ### Latest version is recommended.