Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Oct;12(10):966-8.
doi: 10.1038/nmeth.3505. Epub 2015 Aug 10.

SpeedSeq: Ultra-Fast Personal Genome Analysis and Interpretation

Free PMC article

SpeedSeq: Ultra-Fast Personal Genome Analysis and Interpretation

Colby Chiang et al. Nat Methods. .
Free PMC article


SpeedSeq is an open-source genome analysis platform that accomplishes alignment, variant detection and functional annotation of a 50× human genome in 13 h on a low-cost server and alleviates a bioinformatics bottleneck that typically demands weeks of computation with extensive hands-on expert involvement. SpeedSeq offers performance competitive with or superior to current methods for detecting germline and somatic single-nucleotide variants, structural variants, insertions and deletions, and it includes novel functionality for streamlined interpretation.


Figure 1
Figure 1. SpeedSeq workflow
(a) SpeedSeq converts raw reads into prioritized variants in 13 hours for a 50× human dataset. (b) Germline SNV (N=2,803,144) and (c) indel (N=364,031) receiver operating characteristic (ROC) curves over the Genome in a Bottle (GIAB) truth set for SpeedSeq, GATK Unified Genotyper (GATK-UG) and Haplotype Caller (GATK-HC). (d) Somatic SNV detection ROC curves for a simulated 50× tumor-normal pair using SpeedSeq and three other tools (N=875,206). Open circles (b-d) denote the data points reported in the main text. (e) SpeedSeq’s SV detection performance by quality score of all SVs (black), those with split-read and paired-end support (blue), and those with read-depth support from CNVnator (red), as validated by either PacBio/Moleculo long-reads or the 1000 Genomes Project. (f) Schematic of haplotype-based SV validation showing undetected (open circles), consistently segregating (black circles), and inconsistently segregating (red circles) SVs through the CEPH 1463 pedigree.
Figure 2
Figure 2. Case study in a tumor-normal pair
A SpeedSeq workflow demonstrating the seven succinct commands required to process a tumor-normal pair (TCGA-E2-A14P) from raw FASTQ reads to clinically actionable somatic mutations with predicted damaging consequences. In this tumor, SpeedSeq detected a previously reported somatic gene fusion product between exon 1 of TBL1XR1 and exon 2 of PIK3CA.

Similar articles

See all similar articles

Cited by 105 articles

See all "Cited by" articles


    1. Li H. 2013 1303.3997.
    1. DePristo MA, et al. Nature Genetics. 2011;43:491–498. - PMC - PubMed
    1. Li H, et al. Bioinformatics. 2009;25:2078–2079. - PMC - PubMed
    1. Dewey FE, et al. JAMA. 2014;311:1035–1045. - PMC - PubMed
    1. Faust GG, Hall IM. Bioinformatics. 2014;30:2503–2505. - PMC - PubMed

Online methods references

    1. Tarasov A, Vilella AJ, Cuppen E, Nijman IJ, Prins P. Bioinformatics. 2015 doi:10.1093/bioinformatics/btv098. - PMC - PubMed
    1. Tange O. The USENIX Magazine. 2011;36:42–47.
    1. Cleary JG, et al. J. Comput. Biol. 2014;21:405–419. - PubMed

Publication types