REPUTE: An OpenCL based Read Mapping Tool for Embedded Genomics

Maheshwari, Sidharth; Shafik, Rishad; Wilson, I. J.; Yakovlev, Alex; Acharyya, Amit

doi:10.23919/date48585.2020.9116238

Cited by 1 publication

(4 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Filtration aims to identify suitable δ +1 kmers in a read to minimize the total candidate locations leading to reduced mapping time as there are fewer DP-based verification cycles. We improve over DP-based filtration method used in [5]; to use proposed memory-aware data structures with fixed bit-vector operations to obtain the desired value of TM as shown in Fig. 1(b).…”

Section: B Filtration For Memory-aware Data Structuresmentioning

confidence: 99%

“…The continuous growth in data has strained the available computational resources, thereby, consumig large amounts of energy [4]. Indeed, the demand of scaling up the computational capabilities energy efficiently is hindering the progress of this crucial and emerging new application [5].…”

Section: Introductionmentioning

confidence: 99%

“…It, however, uses a heuristic filtration methodology. REPUTE [5] proposes a DP based filtration methodology using OpenCL to improve performance and demonstrates an embedded implementation, for chr21, on HiKey970 platform with energy savings of 27× compared to a workstation. Both REPUTE and CORAL, however, use data structures with large memory footprints making it unfit to be used with longer chromosomes (e.g.…”

Section: Introductionmentioning

confidence: 99%

“…We execute read mappers on two systems 1) Intel i7-8750H CPU, 16GB RAM + Nvidia GTX 1050 Ti, 4GB RAM; 2) Odroid N2 with quad-core ARM Cortex-A73 + dual core Cortex-A53, 4GB RAM. Among state-of-the-art read mappers, only Hobbes3 and RazerS3, although oriented towards CPU, ran successfully on HiKey970 (6GB RAM) for comparison in REPUTE [5]. However, with just 4GB RAM on Odroid N2, only Hobbes3 and PLEDGER could be benchmarked.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

PLEDGER: Embedded Whole Genome Read Mapping using Algorithm-HW Co-design and Memory-aware Implementation

Maheshwari

Shafik

Wilson

et al. 2021

2021 Design, Automation &Amp; Test in Europe Conference &Amp; Exhibition (DATE)

Self Cite

View full text Add to dashboard Cite

With over 6000 known genetic disorders, genomics is a key driver to transform the current generation of healthcare from reactive to personalized, predictive, preventive and participatory (P4) form. High throughput sequencing technologies produce large volumes of genomic data, making genome reassembly and analysis computationally expensive in terms of performance and energy. In this paper, we propose an algorithmhardware co-design driven acceleration approach for enabling translational genomics. Core to our approach is a Pyopencl based tooL for gEnomic workloaDs tarGeting Embedded platforms (PLEDGER). PLEDGER is a scalable, portable and energyefficient solution to genomics targeting low-cost embedded platforms. It is a read mapping tool to reassemble genome, which is a crucial prerequisite to genomics. Using bit-vectors and variable level optimisations, we propose a low-memory footprint, dynamic programming based filtration and verification kernel capable of accelerated parallel heterogeneous executions. We demonstrate, for the first time, mapping of real reads to whole human genome on a memory-restricted embedded platform using novel memory-aware preprocessed data structures. We compare the performance and accuracy of PLEDGER with state-of-theart RazerS3, Hobbes3, CORAL and REPUTE on two systems: 1) Intel i7-8750H CPU + Nvidia GTX 1050 Ti; 2) Odroid N2 with 6 cores: 4×Cortex-A73 + 2×Cortex-A53 and Mali GPU. PLEDGER demonstrates persistent energy and accuracy advantages compared to state-of-the-art read mappers producing up to 11× speedups and 5.9× energy savings compared to stateof-the-art hardware resources.

show abstract

Section: B Filtration For Memory-aware Data Structuresmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations