Accelerated Whole Genome Sequencing Analysis

Experience high-throughput, high-accuracy DNA and RNA sequencing with NVIDIA Clara Parabricks.

Power faster, more accurate genomic analysis with NVIDIA Clara™ Parabricks®. A GPU-accelerated computational genomics application framework, it delivers accurate and accelerated analysis of next-generation sequencing (NGS) data for researchers, clinical teams, medical centers and sequencing centers for cancer sequencing projects, population studies, RNA-Seq, and many more applications.

View Document: Parabricks Product Sheet (PDF 304 kB)
High Throughput

High Throughput

Process up to 60 whole genomes per day on a single server.

Better Accuracy

Better Accuracy

No need to sacrifice accuracy to reduce turnaround time. A suite of germline and somatic callers is available.

Flexible Pipelines

Flexible Pipelines

Create customized, accelerated pipelines by configuring over 50 tools for any specific pipeline or application need.

30–60X Faster

30–60X Faster

Reduce computing costs up to 50 percent by running 30–60X faster than CPU-only solutions.

Accelerated Applications for Genomics

From FASTQ to Binary Alignment Map (BAM) to Variant Call Format (VCF), Clara Parabricks accelerates genomic analysis for somatic, germline, and structural variant workflows. It also improves sequencer speeds and consensus read base calling accuracy with a suite of over 50 tools.

Cancer Sequencing

Accelerate cancer sequencing workflows up to 40X with five somatic callers, support for gene panels, and enhanced accuracy when using molecular barcodes or unique molecular indices.

Whole Genome Sequencing

Accelerate whole genome analysis up to 60X with a breadth of tools for streamlined end-to-end workflows, including single nucleotide polymorphisms (SNPs), insertions and deletions (indels), and structural variant calling.

Exome Sequencing

Analyze exomes in under five minutes with high-accuracy variant calling and a full suite of germline and somatic callers.

RNA Sequencing

Access a suite of tools to accelerate RNA sequencing, from splice-aware alignment to fusion detection to gene expression quantification.

FASTQ to Binary Alignment Map (BAM) to VCF

Unmatched Performance and Accuracy

Accelerate Germline Variant Analysis

From FastQ to VCF, Clara Parabricks accelerates runtimes across a series of hardware configurations with NVIDIA A100 Tensor Core GPUs. Genomic researchers can experience acceleration across every step of their analysis workflows, from alignment to sorting to variant calling. When more GPUs are used, a near-linear scaling in compute times is observed compared to CPU-only systems, allowing over 80X acceleration.

Analyze a Whole Human Genome at 30X Coverage in Less Than 30 Minutes

NVIDIA runtimes were achieved on an NVIDIA DGX™ A100 320GB system. All analyses were run on HG002 Genome In A Bottle (GIAB NIST) sequenced to 30X coverage, with variant calls made with GATKv4.1.0 HaplotypeCaller.

Maximize Accuracy in Germline Calling

With Clara Parabricks, multiple germline callers can be run simultaneously to improve accuracy, minimizing false positive and/or false negative calls. As shown in this upset plot, when running both DeepVariant and HaplotypeCaller, DeepVariant was able to report  approximately 5,000 variants missed by HaplotypeCaller.

Maximize Accuracy in Germline Calling

Most commonly used reference for germline detection HG002 Genome In A Bottle (GIAB NIST) sequenced to 30X coverage. Data was run g4dn.12x large GPUs on AWS.

Accelerate Somatic Mutation Analysis

Clara Parabricks speeds up runtimes for four somatic callers, accelerating analysis by up to 66X compared to CPU. With Mutect2 and Muse, whole human genome analysis can be reduced from over 30 hours to under one hour. A total of five somatic callers are part of the Parabricks software suite.

Accelerate Cancer Sequencing up to 66X

Performance runtimes were achieved for the CPU baseline using 32vcpu with a 128GB node, while the Parabricks instances used 8 NVIDIA V100 Tensor Core GPUs with a 488GB node. The data used was 50X SEQC-II, while the metrics for LoFreq used a 30X dataset.

Improve Somatic Calling Visibility and Accuracy

Clara Parabricks lets you utilize the strengths of multiple callers simultaneously to improve accuracy and extract more insights from your data. A suite of five somatic callers—MuSE, Mutect2, LoFreq, Strelka2, and Somatic Sniper—can be used for comprehensive mutation detection. Run them individually or in combination to uncover more variants at faster speeds with reliable results.

Improve Somatic Calling Visibility and Accuracy

High-confidence regions of Sequencing Quality Control Phase (SEQC-II) data showing the somatic variants for five different callers. Data was run g4dn.12x large GPUs on AWS.

Next-Generation Sequencing on-Premises or in the Cloud

Running NVIDIA Clara Parabricks requires a minimum of two GPUs. On-premise licenses are annual subscriptions on a per-GPU basis, which include full access to all pipelines in the NVIDIA Clara Parabricks software suite.

NVIDIA DGX A100

On-Premises Genomic Analysis Solutions

Running on one NVIDIA DGX™ A100, Clara Parabricks analyzes up to 25,000 whole genomes per year (30X coverage) with Genomic Analysis Toolkit (GATK) from Broad Institute. Throughput scales linearly beyond a single DGX A100.

Learn more about DGX A100

Genomic Analysis in the Cloud

In the cloud, time is money. Choose from a range of GPU instances capable of analyzing genomes in a matter of minutes. For easy deployment on AWS has Clara Parabricks available in an Amazon Machine Image (AMI) for immediate use.

Explore the AWS Marketplace
AWS Marketplace : NVIDIA Clara Parabricks Pipelines

How Clara Parabricks Is Making a Real-World Impact

DeepVariant with Clara Parabricks

Regeneron Genetics Center

See how Regeneron is using DeepVariant with Clara Parabricks to generate scalable, high-quality data that can be reproduced across the genomics community.

Watch webinar
NVIDIA DGX A100 and Clara Parabricks

National Biobank of Thailand (NBT)

NBT reduced its whole genome sequencing data processing by four months and its individual user processing time from over 30 hours to just 1-2 hours with NVIDIA DGX A100 and Clara Parabricks.

Read article
NVIDIA V100 Tensor Core GPUs

Translational Genomics Research Institute (TGen)

With NVIDIA Clara Parabricks and NVIDIA V100 Tensor Core GPUs, TGen sped up genomic sequencing analysis by 6X to help determine personalized treatments for children with rare neurological disorders.

Read success story
NVIDIA Clara Parabricks

Washington University in Saint Louis (WUSTL)

Integrating GPU-based steps with NVIDIA Clara Parabricks enabled Tychele Turner of WUSTL to drastically reduce the runtime for de novo variant calling.

Watch webinar

Resources

  • Enterprise Support
  • Documentation

Parabricks Pipelines Support Services provides you with an easy, reliable way to improve productivity and reduce system downtime for your production analysis systems. NVIDIA Support Services for Parabricks includes:

  • Direct communication with NVIDIA technical and domain experts
  • Help with installation and usage of Parabricks Pipelines
  • Rapid response and timely issue resolution (cases are accepted for 24/7 remote support)
  • Access to all software upgrades, maintenance releases, bug fixes, and security patches
  • Help with temporary workarounds for confirmed bugs

Get Started Today

Already have an active support contract?

Log into the support portal >

Any additional questions?

Contact parabricks-eval-support@nvidia.com

Getting Started

Clara Parabricks can be set up in just 10 minutes with a 90-day free trial. Watch a short video for a step-by-step walkthrough or learn more about the software suite and tools available in the Parabricks technical documentation.

Start a free 90-day trial of NVIDIA Clara Parabricks today or contact our sales team to get a quote.

Start Trial
Get a Quote

Get The Latest genomics News

  1. Section
Sign up for enterprise news, announcements, and more from NVIDIA.

Platforms

Products

Developers