Nvidia Parabricks

Parabricks offers end users various collections of tools organized sequentially to analyze the raw data according to the user's requirements, called pipelines. Nevertheless, users can decide to run the tools provided by Parabricks as a standalone, still exploiting GPU acceleration to overcome possible computational bottlenecks. Only some of the provided tools in the suite are GPU-based.{{Cite web |title=Welcome to NVIDIA Parabricks v4.3.1 |url=https://docs.nvidia.com/clara/parabricks/latest/index.html |access-date=2024-07-10 |website=NVIDIA Docs |language=en}}

File:Parabricks-pipeline-overview.png

Overall, all the pipelines share a standard structure. Most of the pipelines are built to analyze FASTQ data resulting from various sequencing technologies (e.g., short- or long-read). Input genomic sequences are firstly aligned and then undergo a quality control process. These two processes provide a BAM or a CRAM file as an intermediate result. Based on this data, the variant calling task that follows employs high-accuracy tools that are already widely used. As output, these pipelines provide the identified mutations in a VCF (or a gVCF).

The germline pipeline offered by Parabricks follows the best practices{{Cite web |date=2015-03-19 |title=Best Practices for Variant Calling with the GATK |url=https://www.broadinstitute.org/partnerships/education/broade/best-practices-variant-calling-gatk-1 |access-date=2024-07-09 |website=@broadinstitute |language=en}} proposed by the Broad Institute in their Genome Analysis ToolKit (GATK).{{Cite web |date=2010-06-08 |title=Genome Analysis Toolkit (GATK) |url=https://www.broadinstitute.org/scientific-community/software/genome-analysis-toolkit-gatk |access-date=2024-07-09 |website=@broadinstitute |language=en}} The germline pipeline operates on the FASTQ files provided as input by the user to call the variants that, belonging to the germ line, can be inherited.

This pipeline analyzes data computing the read alignment with BWA-MEM{{Citation | vauthors = Li H |title=Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM |date=2013-05-26 |arxiv=1303.3997 }}{{Cite web |title=Burrows-Wheeler Aligner |url=https://bio-bwa.sourceforge.net/ |access-date=2024-07-09 |website=bio-bwa.sourceforge.net}} and calling variants using GATK HaplotypeCaller,{{Citation | vauthors = Poplin R, Ruano-Rubio V, DePristo MA, Fennell TJ, Carneiro MO, Van der Auwera GA, Gauthier LD, Levy-Moonshine A |title=Scaling accurate genetic variant discovery to tens of thousands of samples |date=2018-07-24 |url=https://www.biorxiv.org/content/10.1101/201178v3 |access-date=2024-07-09 |language=en |doi=10.1101/201178 }} one of the most relevant tools in the domain for germline variant calling.

Besides the pipeline that resorts to HaplotypeCaller to call variants, Parabricks also offers an alternative pipeline that still calls germline variants but is based on DeepVariant.{{cite journal | vauthors = Poplin R, Chang PC, Alexander D, Schwartz S, Colthurst T, Ku A, Newburger D, Dijamco J, Nguyen N, Afshar PT, Gross SS, Dorfman L, McLean CY, DePristo MA | title = A universal SNP and small-indel variant caller using deep neural networks | journal = Nature Biotechnology | volume = 36 | issue = 10 | pages = 983–987 | date = November 2018 | pmid = 30247488 | doi = 10.1038/nbt.4235 }}{{Citation |title=google/deepvariant |date=2024-07-04 |url=https://github.com/google/deepvariant |access-date=2024-07-09 |publisher=Google}} DeepVariant is a variant caller, developed and maintained by Google, capable of identifying mutations using a deep learning-based approach. The core of DeepVariant is a convolutional neural network (CNN) that identifies variants by transforming this task into an image classification operation. In Parabricks, the inference process is accelerated in hardware. For this pipeline, only T4, V100, and A100 GPUs are supported.

Analyses performed according to this pipeline are compliant with the use of BWA-MEM for the alignment by Google's CNN for variant calling.

Still compliant with GATK best practices, the human_par pipeline allows users to identify mutations in the entire human genome, including sex chromosomes X and Y, and, thus, it is compliant with their ploidy. For male samples, firstly, the pipeline runs HaplotypeCaller on all the regions that do not belong to the X and Y chromosomes and on the pseudoautosomal region with ploidy equal to 1. Then, HaplotypeCaller analyses the X and Y regions without the pseudoautosomal region with ploidy 2. Regarding female samples, instead, the pipeline runs HaplotypeCaller on the entire genome, with ploidy 2.

The sex of the sample can be determined in two main ways:

1. Manually set with the --sample-sex option;
2. Specify the X vs. Y ratio with range options --range-male and --range-female and let the tool automatically infer the sex of the samples based on the X and Y reads count.

The pipeline requires the user to specify at least one of these three options.

As for the germline case, since this pipeline targets the germline variants, the pipeline resorts to BWA-MEM for the alignment, followed by HaplotypeCaller for variant calling.

Parabricks' somatic pipeline is designed to call somatic variants, i.e., those mutations affecting non-reproductive (somatic) cells. This pipeline can analyze both tumor and non-tumor genomes, offering either tumor-only or tumor/normal analyses for comprehensive examinations.

As in the germline pipeline, the alignment task is carried out using BWA-MEM followed by GATK Mutect{{cite journal | vauthors = Cibulskis K, Lawrence MS, Carter SL, Sivachenko A, Jaffe D, Sougnez C, Gabriel S, Meyerson M, Lander ES, Getz G | title = Sensitive detection of somatic point mutations in impure and heterogeneous cancer samples | journal = Nature Biotechnology | volume = 31 | issue = 3 | pages = 213–219 | date = March 2013 | pmid = 23396013 | pmc = 3833702 | doi = 10.1038/nbt.2514 }} to identify the possible mutations. Mutect is used instead of HaplotypeCaller due to its focus on somatic mutations, as opposed to germline mutations targeted by HaplotypeCaller.

This pipeline is optimized for short variant discovery (i.e., Single-nucleotide polymorphisms (SNPs) and indels) in RNAseq data. It follows the Broad Institute's best practices for these types of analyses.

It relies on the STAR aligner,{{cite journal | vauthors = Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR | title = STAR: ultrafast universal RNA-seq aligner | journal = Bioinformatics | volume = 29 | issue = 1 | pages = 15–21 | date = January 2013 | pmid = 23104886 | pmc = 3530905 | doi = 10.1093/bioinformatics/bts635 }} a read aligner specialized for RNA sequences for aligning the reads, and HaplotypeCaller for calling variants.

Parabricks provides a collection of tools to perform genomics analyses, classified into six main categories related to their task. These tools combined constitutes Parabricks' pipelines, and can be also used as-is.

For FASTQ and BAM files processing, the proposed tools are:

• {{Mono|applybsqr}}
• {{Mono|bam2fq}}
• {{Mono|bamsort}}
• {{Mono|bqsr}}
• {{Mono|fq2bam}}
• {{Mono|fq2bamfast}}
• {{Mono|fq2bam_meth}}
• {{Mono|markdup}}
• {{Mono|minimap2}} (beta)

For calling variants, the proposed tools are:

• {{Mono|deepsomatic}}
• {{Mono|deepvariant}}
• {{Mono|deepvariant_germline}}
• {{Mono|germline}} (GATK Germline Pipeline)
• {{Mono|haplotypecaller}}
• {{Mono|mutectcaller}}
• {{Mono|pacbio_germline}} (beta)
• {{Mono|postpon}}
• {{Mono|prepon}}
• {{Mono|somatic}} (Somatic Variant Caller)

For RNA processing, the proposed tools are:

• {{Mono|rna_fq2bam}}
• {{Mono|starfusion}}

For results quality control, the proposed tools are:

• {{Mono|bammetrics}}
• {{Mono|collectmultiplemetrics}}

For processing variants, the proposed tools are:

• {{Mono|dbsnp}}

For processing gVCF files, the proposed tools are:

• {{Mono|genotypegvcf}}
• {{Mono|indexgvcf}}

Not all the listed tools are accelerated on GPU.

Users can download and run Parabricks pipelines on their local servers, allowing for private, on-site data processing and analysis. They also can deploy Parabricks pipelines on cloud platforms, with improved scalability for larger datasets. Supported cloud providers include AWS, GCP, OCI, and Azure.

In the latest release (v4.3.1-1), Parabricks includes support for the Nvidia Grace Hopper super chip. The Nvidia GH200 Grace Hopper Superchip is a heterogeneous platform designed for high-performance computing and artificial intelligence, combining an Nvidia Grace and a Hopper on a single chip.{{Cite book | vauthors = Simakov NA, Jones MD, Furlani TR, Siegmann E, Harrison RJ |chapter=First Impressions of the NVIDIA Grace CPU Superchip and NVIDIA Grace Hopper Superchip for Scientific Workloads |date=2024-01-11 |title=Proceedings of the International Conference on High Performance Computing in Asia-Pacific Region Workshops |chapter-url=https://doi.org/10.1145/3636480.3637097 |series=HPCAsia '24 Workshops |location=New York, NY, USA |publisher=Association for Computing Machinery |pages=36–44 |doi=10.1145/3636480.3637097 |isbn=979-8-4007-1652-2}} This platform enhances application performance using both GPUs and CPUs, offering a programming model aimed at improving performance, portability, and productivity.{{Cite web |title=Grace Hopper Superchip |url=https://docs.nvidia.com/clara/parabricks/latest/GraceHopper.html |access-date=2024-07-10 |website=NVIDIA Docs |language=en}}

Due to the computational power required by genomics workloads, Parabricks has found application in several research studies with different applicative domains, especially in cancer research.{{Cite journal | vauthors = Crowgey EL, Vats P, Franke K, Burnett G, Sethia A, Harkins T, Druley TE | date = July 2021 |title=Abstract 165: Enhanced processing of genomic sequencing data for pediatric cancers: GPUs and machine learning techniques for variant detection |url=https://aacrjournals.org/cancerres/article/81/13_Supplement/165/667493/Abstract-165-Enhanced-processing-of-genomic |journal=Cancer Research |language=en |volume=81 |issue=13_Supplement |pages=165 |doi=10.1158/1538-7445.AM2021-165 |issn=0008-5472|url-access=subscription }}{{cite journal | vauthors = Ng JK, Vats P, Fritz-Waters E, Sarkar S, Sams EI, Padhi EM, Payne ZL, Leonard S, West MA, Prince C, Trani L, Jansen M, Vacek G, Samadi M, Harkins TT, Pohl C, Turner TN | title = de novo variant calling identifies cancer mutation signatures in the 1000 Genomes Project | journal = Human Mutation | volume = 43 | issue = 12 | pages = 1979–1993 | date = December 2022 | pmid = 36054329 | pmc = 9771978 | doi = 10.1002/humu.24455 }}{{cite journal | vauthors = Lee TH, Jang BS, Chang JH, Kim E, Park JH, Chie EK | title = Genomic landscape of locally advanced rectal adenocarcinoma: Comparison between before and after neoadjuvant chemoradiation and effects of genetic biomarkers on clinical outcomes and tumor response | journal = Cancer Medicine | volume = 12 | issue = 14 | pages = 15664–15675 | date = July 2023 | pmid = 37260182 | pmc = 10417181 | doi = 10.1002/cam4.6169 }}

Scientists from Washington University used the Parabricks DeepVariant pipeline for identifying variants (e.g., SNPs and small indels) in long-read Hi-Fi whole-genome sequencing (WGS) data generated with PacBio's Revio SMRT Cell technology.{{cite journal | vauthors = Manuel JG, Heins HB, Crocker S, Neidich JA, Sadzewicz L, Tallon L, Turner TN | title = High Coverage Highly Accurate Long-Read Sequencing of a Mouse Neuronal Cell Line Using the PacBio Revio Sequencer | journal = bioRxiv | date = June 2023 | pmid = 37333171 | pmc = 10274723 | doi = 10.1101/2023.06.06.543940 }}

In addition to the pipelines, individual components of Parabricks have been used as standalone tools in academic settings. For example, the accelerated DeepVariant has been employed in a novel process to reduce the processing time further for WGS Nanopore data.{{cite journal | vauthors = Goenka SD, Gorzynski JE, Shafin K, Fisk DG, Pesout T, Jensen TD, Monlong J, Chang PC, Baid G, Bernstein JA, Christle JW, Dalton KP, Garalde DR, Grove ME, Guillory J, Kolesnikov A, Nattestad M, Ruzhnikov MR, Samadi M, Sethia A, Spiteri E, Wright CJ, Xiong K, Zhu T, Jain M, Sedlazeck FJ, Carroll A, Paten B, Ashley EA | title = Accelerated identification of disease-causing variants with ultra-rapid nanopore genome sequencing | journal = Nature Biotechnology | volume = 40 | issue = 7 | pages = 1035–1041 | date = July 2022 | pmid = 35347328 | pmc = 9287171 | doi = 10.1038/s41587-022-01221-5 }}

In 2022, Nvidia announced a collaboration with the Broad Institute to provide researchers with the benefits of accelerated computing. This partnership includes the entire suite of Nvidia's biomedical hardware-accelerated software suite called Clara, that includes Parabricks and MONAI.{{Cite web |title=The Broad Institute and NVIDIA Bring NVIDIA Clara to Terra Cloud Platform Serving 25,000 Researchers Advancing Biomedical Discovery |url=http://nvidianews.nvidia.com/news/the-broad-institute-and-nvidia-bring-nvidia-clara-to-terra-cloud-platform-serving-25-000-researchers-advancing-biomedical-discovery |access-date=2024-07-09 |website=NVIDIA Newsroom |language=en-us}} Similarly, the Regeneron Genetics Center uses Parabricks to expedite the secondary analysis of the exomes they sequence in their high-throughput sequencing center, leverage the DeepVariant Germline pipeline inside their workflows.{{Cite web |title=UK Biobank Advances Genomics Research with NVIDIA Clara Parabricks |url=https://resources.nvidia.com/en-us-hc-genomics/uk-biobank |access-date=2024-07-09 |website=NVIDIA |language=en}}

• List of bioinformatics software
• List of sequence alignment software

{{Reflist}}

{{refbegin}}

• {{cite journal | vauthors = Franke KR, Crowgey EL | title = Accelerating next generation sequencing data analysis: an evaluation of optimized best practices for Genome Analysis Toolkit algorithms | journal = Genomics & Informatics | volume = 18 | issue = 1 | pages = e10 | date = March 2020 | pmid = 32224843 | doi = 10.5808/GI.2020.18.1.e10 | publisher = BMC | pmc = 7120354 }}
• {{cite journal | vauthors = O'Connell KA, Yosufzai ZB, Campbell RA, Lobb CJ, Engelken HT, Gorrell LM, Carlson TB, Catana JJ, Mikdadi D, Bonazzi VR, Klenk JA | title = Accelerating genomic workflows using NVIDIA Parabricks | journal = BMC Bioinformatics | volume = 24 | issue = 1 | pages = 221 | date = May 2023 | pmid = 37259021 | pmc = 10230726 | doi = 10.1186/s12859-023-05292-2 | author-mask5 = et al. | doi-access = free }}

{{refend}}

• [https://www.nvidia.com/en-us/clara/ NVIDIA Clara]
• [https://www.nvidia.com/en-us/clara/genomics/ NVIDIA Clara for Genomics]

Category:Nvidia software

Category:Bioinformatics software

Category:Medical software