Arioc: GPU-accelerated alignment of short bisulfite-treated reads

Richard Wilton, Xin Li, Andrew P Feinberg, Alexander S. Szalay

Research output: Contribution to journalArticle

Abstract

Motivation: The alignment of bisulfite-treated DNA sequences (BS-seq reads) to a large genome involves a significant computational burden beyond that required to align non-bisulfite-treated reads. In the analysis of BS-seq data, this can present an important performance bottleneck that can be mitigated by appropriate algorithmic and software-engineering improvements. One strategy is to modify the read-alignment algorithms by integrating the logic related to BS-seq alignment, with the goal of making the software implementation amenable to optimizations that lead to higher speed and greater sensitivity than might otherwise be attainable. Results: We evaluated this strategy using Arioc, a short-read aligner that uses GPU (generalpurpose graphics processing unit) hardware to accelerate computationally-expensive programming logic. We integrated the BS-seq computational logic into both GPU and CPU code throughout the Arioc implementation. We then carried out a read-by-read comparison of Arioc's reported alignments with the alignments reported by well-known CPU-based BS-seq read aligners. With simulated reads, Arioc's accuracy is equal to or better than the other read aligners we evaluated. With human sequencing reads, Arioc's throughput is at least 10 times faster than existing BS-seq aligners across a wide range of sensitivity settings.

Original languageEnglish (US)
Pages (from-to)2673-2675
Number of pages3
JournalBioinformatics
Volume34
Issue number15
DOIs
StatePublished - Jan 1 2018

Fingerprint

Graphics Processing Unit
Alignment
Software
Program processors
Logic
Genome
Logic programming
DNA sequences
Logic Programming
Software Engineering
DNA Sequence
Sequencing
Accelerate
Software engineering
High Speed
Throughput
Genes
Graphics processing unit
hydrogen sulfite
Hardware

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Computational Mathematics

Cite this

Arioc : GPU-accelerated alignment of short bisulfite-treated reads. / Wilton, Richard; Li, Xin; Feinberg, Andrew P; Szalay, Alexander S.

In: Bioinformatics, Vol. 34, No. 15, 01.01.2018, p. 2673-2675.

Research output: Contribution to journalArticle

Wilton, Richard ; Li, Xin ; Feinberg, Andrew P ; Szalay, Alexander S. / Arioc : GPU-accelerated alignment of short bisulfite-treated reads. In: Bioinformatics. 2018 ; Vol. 34, No. 15. pp. 2673-2675.
@article{efd15519635f4a53a6aaaa9ab302929f,
title = "Arioc: GPU-accelerated alignment of short bisulfite-treated reads",
abstract = "Motivation: The alignment of bisulfite-treated DNA sequences (BS-seq reads) to a large genome involves a significant computational burden beyond that required to align non-bisulfite-treated reads. In the analysis of BS-seq data, this can present an important performance bottleneck that can be mitigated by appropriate algorithmic and software-engineering improvements. One strategy is to modify the read-alignment algorithms by integrating the logic related to BS-seq alignment, with the goal of making the software implementation amenable to optimizations that lead to higher speed and greater sensitivity than might otherwise be attainable. Results: We evaluated this strategy using Arioc, a short-read aligner that uses GPU (generalpurpose graphics processing unit) hardware to accelerate computationally-expensive programming logic. We integrated the BS-seq computational logic into both GPU and CPU code throughout the Arioc implementation. We then carried out a read-by-read comparison of Arioc's reported alignments with the alignments reported by well-known CPU-based BS-seq read aligners. With simulated reads, Arioc's accuracy is equal to or better than the other read aligners we evaluated. With human sequencing reads, Arioc's throughput is at least 10 times faster than existing BS-seq aligners across a wide range of sensitivity settings.",
author = "Richard Wilton and Xin Li and Feinberg, {Andrew P} and Szalay, {Alexander S.}",
year = "2018",
month = "1",
day = "1",
doi = "10.1093/bioinformatics/bty167",
language = "English (US)",
volume = "34",
pages = "2673--2675",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "15",

}

TY - JOUR

T1 - Arioc

T2 - GPU-accelerated alignment of short bisulfite-treated reads

AU - Wilton, Richard

AU - Li, Xin

AU - Feinberg, Andrew P

AU - Szalay, Alexander S.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Motivation: The alignment of bisulfite-treated DNA sequences (BS-seq reads) to a large genome involves a significant computational burden beyond that required to align non-bisulfite-treated reads. In the analysis of BS-seq data, this can present an important performance bottleneck that can be mitigated by appropriate algorithmic and software-engineering improvements. One strategy is to modify the read-alignment algorithms by integrating the logic related to BS-seq alignment, with the goal of making the software implementation amenable to optimizations that lead to higher speed and greater sensitivity than might otherwise be attainable. Results: We evaluated this strategy using Arioc, a short-read aligner that uses GPU (generalpurpose graphics processing unit) hardware to accelerate computationally-expensive programming logic. We integrated the BS-seq computational logic into both GPU and CPU code throughout the Arioc implementation. We then carried out a read-by-read comparison of Arioc's reported alignments with the alignments reported by well-known CPU-based BS-seq read aligners. With simulated reads, Arioc's accuracy is equal to or better than the other read aligners we evaluated. With human sequencing reads, Arioc's throughput is at least 10 times faster than existing BS-seq aligners across a wide range of sensitivity settings.

AB - Motivation: The alignment of bisulfite-treated DNA sequences (BS-seq reads) to a large genome involves a significant computational burden beyond that required to align non-bisulfite-treated reads. In the analysis of BS-seq data, this can present an important performance bottleneck that can be mitigated by appropriate algorithmic and software-engineering improvements. One strategy is to modify the read-alignment algorithms by integrating the logic related to BS-seq alignment, with the goal of making the software implementation amenable to optimizations that lead to higher speed and greater sensitivity than might otherwise be attainable. Results: We evaluated this strategy using Arioc, a short-read aligner that uses GPU (generalpurpose graphics processing unit) hardware to accelerate computationally-expensive programming logic. We integrated the BS-seq computational logic into both GPU and CPU code throughout the Arioc implementation. We then carried out a read-by-read comparison of Arioc's reported alignments with the alignments reported by well-known CPU-based BS-seq read aligners. With simulated reads, Arioc's accuracy is equal to or better than the other read aligners we evaluated. With human sequencing reads, Arioc's throughput is at least 10 times faster than existing BS-seq aligners across a wide range of sensitivity settings.

UR - http://www.scopus.com/inward/record.url?scp=85055193469&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055193469&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/bty167

DO - 10.1093/bioinformatics/bty167

M3 - Article

C2 - 29554207

AN - SCOPUS:85055193469

VL - 34

SP - 2673

EP - 2675

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 15

ER -