Accelerated Seeding for Genome Sequence Alignment with Enumerated Radix Trees

Subramaniyan, Arun; Wadden, Jack; Goliya, Kush; Ozog, Nathan; Wu, Xinzhou; Narayanasamy, Satish; Blaauw, David; Das, Reetuparna

doi:10.1109/isca52012.2021.00038

Cited by 19 publications

(8 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SMEMs found in the SMEM search that are longer than the minimum seed length threshold are selected as seeds. In the following, we denote the SMEM search algorithm of BWA-MEM2 (Vasimuddin et al, 2019) and ERT (Subramaniyan et al, 2021) as SMEM-BWA and SMEM-ERT, respectively. SMEM search algorithm and extension.…”

Section: The Seeding Algorithm Of Bwa-mem2 and Ertmentioning

confidence: 99%

“…In the process, the alignment scores of the extended seeds are calculated using the Banded-Smith-Waterman (BSW) algorithm (Vasimuddin et al, 2019). Many studies (Ahmed et al, 2015;Ho et al, 2019;Houtgast et al, 2018;Subramaniyan et al, 2021;Tárraga et al, 2014) have shown that the seeding phase is a major performance bottleneck in the popular alignment software BWA-MEM2 (Vasimuddin et al, 2019) and Bowtie 2 (Langmead and Salzberg, 2012). In particular, finding an exact match of short readsor substrings of short reads-within the reference DNA sequence, is the main problem that fundamentally constrains the performance of the seeding phase (Langmead and Salzberg, 2012;Li, 2013;Li et al, 2009).…”

Section: Introductionmentioning

confidence: 99%

“…Recently, several new index structures have been studied to solve the exact match search problem in an efficient manner. The state-of-the-art index structures fall into two major categories: traditional index-based (Subramaniyan et al, 2021;Vasimuddin et al, 2019) and machine-learning-based (Ho et al, 2019(Ho et al, , 2021Kirsche et al, 2021). Examples of the traditional index structures are FM-index (Langmead and Salzberg, 2012;Li, 2013;Li and Durbin, 2009;Li et al, 2009) and enumerated radix tree (ERT) (Subramaniyan et al, 2021) index.…”

Section: Introductionmentioning

confidence: 99%

“…The state-of-the-art index structures fall into two major categories: traditional index-based (Subramaniyan et al, 2021;Vasimuddin et al, 2019) and machine-learning-based (Ho et al, 2019(Ho et al, , 2021Kirsche et al, 2021). Examples of the traditional index structures are FM-index (Langmead and Salzberg, 2012;Li, 2013;Li and Durbin, 2009;Li et al, 2009) and enumerated radix tree (ERT) (Subramaniyan et al, 2021) index. FM-index is a compressed version of the suffix array (Ferragina and Manzini, 2001), which progressively extends a substring from a single character and finds its exact match in the reference DNA sequence.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

BWA-MEME: BWA-MEM emulated with a machine learning approach

2022

View full text Add to dashboard Cite

Motivation The growing use of next-generation sequencing and enlarged sequencing throughput require efficient short-read alignment, where seeding is one of the major performance bottlenecks. The key challenge in the seeding phase is searching for exact matches of substrings of short reads in the reference DNA sequence. Existing algorithms, however, present limitations in performance due to their frequent memory accesses. Results This paper presents BWA-MEME, the first full-fledged short read alignment software that leverages learned indices for solving the exact match search problem for efficient seeding. BWA-MEME is a practical and efficient seeding algorithm based on a suffix array search algorithm that solves the challenges in utilizing learned indices for SMEM search which is extensively used in the seeding phase. Our evaluation shows that BWA-MEME achieves up to 3.45x speedup in seeding throughput over BWA-MEM2 by reducing the number of instructions by 4.60x, memory accesses by 8.77x, and LLC misses by 2.21x, while ensuring the identical SAM output to BWA-MEM2. Availability The source code and test scripts are available for academic use at https://github.com/kaist-ina/BWA-MEME/. Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

Section: The Seeding Algorithm Of Bwa-mem2 and Ertmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

BWA-MEME: BWA-MEM emulated with a machine learning approach

2022

View full text Add to dashboard Cite

show abstract

“…BWA-MEM2 ( Vasimuddin et al 2019 ) uses a smaller sampling interval for storing BWT data structure and suffix array that introduces a four times memory overhead, but it is still limited by the data prefetch bound. BWA-MEM2 next adopts enumerated radix tree (ERT) ( Subramaniyan et al 2021 ) to entirely replace the FM-index, and performs the super maximal exact match (SMEM) search algorithm ( Li 2012 ) in a zigzag fashion to reduce the unsuccessful SMEM extensions. By encoding the suffixes of the reference sequence in a radix tree, ERT significantly improves the space locality of seeding at a cost of 64 GB memory for human genome index.…”

Section: Introductionmentioning

confidence: 99%

A compressive seeding algorithm in conjunction with reordering-based compression

Ji,

Zhou,

Ruan

et al. 2024

Bioinformatics

View full text Add to dashboard Cite

Motivation Seeding is a rate-limiting stage in sequence alignment for next-generation sequencing (NGS) reads. The existing optimization algorithms typically utilize hardware and machine-learning techniques to accelerate seeding. However, an efficient solution provided by professional NGS compressors has been largely overlooked by far. In addition to achieving remarkable compression ratios by reordering reads, these compressors provide valuable insights for downstream alignment that reveal the repetitive computations accounting for more than 50% of BWA-MEM seeding procedure at common sequencing coverage. Nevertheless, the exploited redundancy information is not fully realized or utilized. Results In this study, we present a compressive seeding algorithm, named CompSeed, to fill the gap. CompSeed, in collaboration with the existing reordering-based compression tools, finishes the BWA-MEM seeding process in about half the time by caching all intermediate seeding results in compact trie structures to directly answer repetitive inquiries that frequently cause random memory accesses. Furthermore, CompSeed demonstrates better performance as sequencing coverage increases, as it focuses solely on the small informative portion of sequencing reads after compression. The innovative strategy highlights the promising potential of integrating sequence compression and alignment to tackle the ever-growing volume of sequencing data. Availability CompSeed is available at https://github.com/i-xiaohu/CompSeed Supplementary information Supplementary data are available at Bioinformatics online.

show abstract

BWA-MEME: BWA-MEM emulated with a machine learning approach

Jung

Han

2021

Preprint

View full text Add to dashboard Cite

The growing use of next-generation sequencing and enlarged sequencing throughput require efficient short-read alignment, where seeding is one of the major performance bottlenecks. The key challenge in the seeding phase is searching for exact matches of substrings of short reads in the reference DNA sequence. Existing algorithms, however, present limitations in performance due to their frequent memory accesses. This paper presents BWA-MEME, the first full-fledged short read alignment software that leverages learned indices for solving the exact match search problem for efficient seeding. BWA-MEME is a practical and efficient seeding algorithm based on a suffix array search algorithm that solves the challenges in utilizing learned indices for SMEM search which is extensively used in the seeding phase. Our evaluation shows that BWA-MEME achieves up to 3.45x speedup in seeding throughput over BWA-MEM2 by reducing the number of instructions by 4.60x, memory accesses by 8.77x, and LLC misses by 2.21x, while ensuring the identical SAM output to BWA-MEM2.

show abstract

Accelerated Seeding for Genome Sequence Alignment with Enumerated Radix Trees

Cited by 19 publications

References 47 publications

BWA-MEME: BWA-MEM emulated with a machine learning approach

BWA-MEME: BWA-MEM emulated with a machine learning approach

A compressive seeding algorithm in conjunction with reordering-based compression

BWA-MEME: BWA-MEM emulated with a machine learning approach

Contact Info

Product

Resources

About