Darwin-WGA: A Co-processor Provides Increased Sensitivity in Whole Genome Alignments with High Speedup

Turakhia, Yatish; Goenka, Sneha D.; Bejerano, Gill; Dally, William J.

doi:10.1109/hpca.2019.00050

Cited by 26 publications

(17 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, poor sensitivity of whole genome aligners can make entire gene exons appear to be missing (deleted) in a query species ( 64 , 65 ), particularly for genes with a high mutation rate, such as positively selected genes. Sensitive approaches to whole genome alignments have been shown to recover missing exons ( 66 ) and may help further reduce the already low FPR for hcoErosion predictions. We provide details on hcoErosion candidates which contradicted UniProt evidence ( Supplementary Table S4 ) in order to facilitate future work investigating and minimizing the few remaining false positive results of our pipeline.…”

Section: Discussionmentioning

confidence: 99%

A fully-automated method discovers loss of mouse-lethal and human-monogenic disease genes in 58 mammals

Turakhia

Chen

Marcovitz

et al. 2020

Nucleic Acids Research

Self Cite

View full text Add to dashboard Cite

Abstract Gene losses provide an insightful route for studying the morphological and physiological adaptations of species, but their discovery is challenging. Existing genome annotation tools focus on annotating intact genes and do not attempt to distinguish nonfunctional genes from genes missing annotation due to sequencing and assembly artifacts. Previous attempts to annotate gene losses have required significant manual curation, which hampers their scalability for the ever-increasing deluge of newly sequenced genomes. Using extreme sequence erosion (amino acid deletions and substitutions) and sister species support as an unambiguous signature of loss, we developed an automated approach for detecting high-confidence gene loss events across a species tree. Our approach relies solely on gene annotation in a single reference genome, raw assemblies for the remaining species to analyze, and the associated phylogenetic tree for all organisms involved. Using human as reference, we discovered over 400 unique human ortholog erosion events across 58 mammals. This includes dozens of clade-specific losses of genes that result in early mouse lethality or are associated with severe human congenital diseases. Our discoveries yield intriguing potential for translational medical genetics and evolutionary biology, and our approach is readily applicable to large-scale genome sequencing efforts across the tree of life.

show abstract

Section: Discussionmentioning

confidence: 99%

A fully-automated method discovers loss of mouse-lethal and human-monogenic disease genes in 58 mammals

Turakhia

Chen

Marcovitz

et al. 2020

Nucleic Acids Research

Self Cite

View full text Add to dashboard Cite

show abstract

“…a In other domains, such as genomics, GPUs provide lower efficiency than FPGAs but offer faster development time. For genomics, we coded the banded Smith-Waterman algorithm 50 in CUDA for the GPU in one daygiving 25× improvement in efficiency over the CPU. Bringing this algorithm up on an FPGA took two months of RTL design and performance tuning-to achieve four times the efficiency of the GPU.…”

Section: Domains Of Applications Can Be Accelerated With Asics Fpgasmentioning

confidence: 99%

“…Upon receipt of each activation, each PE walks the nonzero row entries for that column in its subset of rows, conventional hardware. With specialized hardware, gapped extension becomes feasible 50 giving much better sensitivity when comparing the genomes of distantly-related species-where the alignments have frequent gaps.…”

Section: Domains Of Applications Can Be Accelerated With Asics Fpgasmentioning

confidence: 99%

Domain-specific hardware accelerators

2020

Self Cite

View full text Add to dashboard Cite

show abstract

“…Hardware specialized acceleration is an effective way to overcome the big genomic data problem. However, most prior works focus on only accelerating genome alignment and assembly [34], particular short read alignment [8,14,21,35,37,43]. However, long read alignment and assembly are not the most-time consuming steps in a nanopore sequencing pipeline.…”

Section: Related Workmentioning

confidence: 99%

“…Genome sequencing [8,21,34,35,37] is a cornerstone for enabling personalized medicine, global food security, and virus surveillance. The emerging nanopore genome sequencing technology [15] is revolutionizing the genome research, industry and market due to its ability to generate ultra-long DNA fragments, aka long reads, as well as provide portability.…”

Section: Introductionmentioning

confidence: 99%

Helix

Lou

Janga

Jiang

2020

Proceedings of the ACM International Conference on Parallel Architectures and Compilation Techniques

View full text Add to dashboard Cite

Nanopore genome sequencing is the key to enabling personalized medicine, global food security, and virus surveillance. The state-ofthe-art base-callers adopt deep neural networks (DNNs) to translate electrical signals generated by nanopore sequencers to digital DNA symbols. A DNN-based base-caller consumes 44.5% of total execution time of a nanopore sequencing pipeline. However, it is difficult to quantize a base-caller and build a power-efficient processing-in-memory (PIM) to run the quantized base-caller. Although conventional network quantization techniques reduce the computing overhead of a base-caller by replacing floating-point multiply-accumulations by cheaper fixed-point operations, it significantly increases the number of systematic errors that cannot be corrected by read votes. The power density of prior nonvolatile memory (NVM)-based PIMs has already exceeded memory thermal tolerance even with active heat sinks, because their power efficiency is severely limited by analog-to-digital converters (ADC). Finally, Connectionist Temporal Classification (CTC) decoding and read voting cost 53.7% of total execution time in a quantized base-caller, and thus became its new bottleneck. In this paper, we propose a novel algorithm/architecture codesigned PIM, Helix, to power-efficiently and accurately accelerate nanopore base-calling. From algorithm perspective, we present systematic error aware training to minimize the number of systematic errors in a quantized base-caller. From architecture perspective, we propose a low-power SOT-MRAM-based ADC array to process analog-to-digital conversion operations and improve power efficiency of prior DNN PIMs. Moreover, we revised a traditional NVM-based dot-product engine to accelerate CTC decoding operations, and create a SOT-MRAM binary comparator array to process read voting. Compared to state-of-the-art PIMs, Helix improves base-calling throughput by 6×, throughput per Watt by 11.9× and per 2 by 7.5× without degrading base-calling accuracy. CCS CONCEPTS • Hardware → Spintronics and magnetic technologies; • Applied computing → Computational genomics.

show abstract

Darwin-WGA: A Co-processor Provides Increased Sensitivity in Whole Genome Alignments with High Speedup

Cited by 26 publications

References 56 publications

A fully-automated method discovers loss of mouse-lethal and human-monogenic disease genes in 58 mammals

A fully-automated method discovers loss of mouse-lethal and human-monogenic disease genes in 58 mammals

Domain-specific hardware accelerators

Helix

Contact Info

Product

Resources

About