{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T07:45:08Z","timestamp":1783410308543,"version":"3.54.6"},"reference-count":41,"publisher":"Oxford University Press (OUP)","issue":"3","license":[{"start":{"date-parts":[[2025,2,6]],"date-time":"2025-02-06T00:00:00Z","timestamp":1738800000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,3,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:sec>\n                  <jats:title>Motivation<\/jats:title>\n                  <jats:p>DNA sequence alignment, an important genomic task, involves assigning short DNA reads to the most probable locations on an extensive reference genome. Conventional methods tackle this challenge in two steps: genome indexing followed by efficient search to locate likely positions for given reads. Building on the success of Large Language Models in encoding text into embeddings, where the distance metric captures semantic similarity, recent efforts have encoded DNA sequences into vectors using Transformers and have shown promising results in tasks involving classification of short DNA sequences. Performance at sequence classification tasks does not, however, guarantee sequence alignment, where it is necessary to conduct a genome-wide search to align every read successfully, a significantly longer-range task by comparison.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Results<\/jats:title>\n                  <jats:p>We bridge this gap by developing a \u201cEmbed-Search-Align\u201d (ESA) framework, where a novel Reference-Free DNA Embedding (RDE) Transformer model generates vector embeddings of reads and fragments of the reference in a shared vector space; read-fragment distance metric is then used as a surrogate for sequence similarity. ESA introduces: (i) Contrastive loss for self-supervised training of DNA sequence representations, facilitating rich reference-free, sequence-level embeddings, and (ii) a DNA vector store to enable search across fragments on a global scale. RDE is 99% accurate when aligning 250-length reads onto a human reference genome of 3 gigabases (single-haploid), rivaling conventional algorithmic sequence alignment methods such as Bowtie and BWA-Mem. RDE far exceeds the performance of six recent DNA-Transformer model baselines such as Nucleotide Transformer, Hyena-DNA, and shows task transfer across chromosomes and species.<\/jats:p>\n               <\/jats:sec>\n               <jats:sec>\n                  <jats:title>Availability and implementation<\/jats:title>\n                  <jats:p>Please see https:\/\/anonymous.4open.science\/r\/dna2vec-7E4E\/readme.md.<\/jats:p>\n               <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btaf041","type":"journal-article","created":{"date-parts":[[2025,2,3]],"date-time":"2025-02-03T12:17:32Z","timestamp":1738585052000},"source":"Crossref","is-referenced-by-count":9,"title":["Embed-Search-Align: DNA sequence alignment using Transformer models"],"prefix":"10.1093","volume":"41","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8495-9416","authenticated-orcid":false,"given":"Pavan","family":"Holur","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering, UCLA , Los Angeles, California, 90024,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"K C","family":"Enevoldsen","sequence":"additional","affiliation":[{"name":"Center for Humanities Computing, Aarhus University , Aarhus, 8000,","place":["Denmark"]},{"name":"Center for Quantitative Genetics and Genomics, Aarhus University , Aarhus, 8000,","place":["Denmark"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shreyas","family":"Rajesh","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, UCLA , Los Angeles, California, 90024,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lajoyce","family":"Mboning","sequence":"additional","affiliation":[{"name":"Department of Chemistry and Biochemistry, UCLA , Los Angeles, California, 90024,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thalia","family":"Georgiou","sequence":"additional","affiliation":[{"name":"Department of Biochemistry, Biophysics, and Structural Biology (MBIDP), UCLA , Los Angeles, California, 90024,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Louis-S","family":"Bouchard","sequence":"additional","affiliation":[{"name":"Department of Chemistry and Biochemistry, UCLA , Los Angeles, California, 90024,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Matteo","family":"Pellegrini","sequence":"additional","affiliation":[{"name":"Department of Molecular, Cell, and Developmental Biology, UCLA , Los Angeles, California, 90024,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vwani","family":"Roychowdhury","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, UCLA , Los Angeles, California, 90024,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2025,2,6]]},"reference":[{"key":"2025031903123372300_btaf041-B1","first-page":"12449","volume-title":"Adv Neural Inf Process Syst","author":"Baevski","year":"2020;"},{"key":"2025031903123372300_btaf041-B2"},{"key":"2025031903123372300_btaf041-B3","first-page":"1877","volume-title":"Adv Neural Inf Process Syst"},{"key":"2025031903123372300_btaf041-B4","author":"Chen"},{"key":"2025031903123372300_btaf041-B5","doi-asserted-by":"crossref","first-page":"404","DOI":"10.1093\/biomet\/26.4.404","article-title":"The use of confidence or fiducial limits illustrated in the case of the binomial","volume":"26","author":"Clopper","year":"1934","journal-title":"Biometrika"},{"key":"2025031903123372300_btaf041-B6","doi-asserted-by":"crossref","first-page":"1422","DOI":"10.1093\/bioinformatics\/btp163","article-title":"Biopython: freely available Python tools for computational molecular biology and bioinformatics","volume":"25","author":"Cock","year":"2009","journal-title":"Bioinformatics"},{"key":"2025031903123372300_btaf041-B7"},{"key":"2025031903123372300_btaf041-B8","volume-title":"Nat Methods"},{"key":"2025031903123372300_btaf041-B9","doi-asserted-by":"crossref","first-page":"3712","DOI":"10.1093\/bioinformatics\/btaa265","article-title":"Vargas: heuristic-free alignment for assessing linear and graph read aligners","volume":"36","author":"Darby","year":"2020","journal-title":"Bioinformatics"},{"key":"2025031903123372300_btaf041-B10","first-page":"4171","author":"Devlin","year":"2019"},{"key":"2025031903123372300_btaf041-B11","author":"Dosovitskiy","year":"2021"},{"key":"2025031903123372300_btaf041-B12","volume-title":"NAR","author":"Fishman"},{"key":"2025031903123372300_btaf041-B13","first-page":"6894","author":"Gao","year":"2021"},{"key":"2025031903123372300_btaf041-B14","first-page":"1735","author":"Hadsell","year":"2006"},{"key":"2025031903123372300_btaf041-B15","author":"Holur","year":"2023"},{"key":"2025031903123372300_btaf041-B16","doi-asserted-by":"crossref","first-page":"593","DOI":"10.1093\/bioinformatics\/btr708","article-title":"ART: a next-generation sequencing read simulator","volume":"28","author":"Huang","year":"2012","journal-title":"Bioinformatics"},{"key":"2025031903123372300_btaf041-B17","doi-asserted-by":"crossref","first-page":"1098","DOI":"10.1109\/JRPROC.1952.273898","article-title":"A method for the construction of minimum-redundancy codes","volume":"40","author":"Huffman","year":"1952","journal-title":"Proc IRE"},{"key":"2025031903123372300_btaf041-B18","doi-asserted-by":"crossref","first-page":"2112","DOI":"10.1093\/bioinformatics\/btab083","article-title":"DNABERT: pre-trained bidirectional encoder representations from transformers model for DNA-language in genome","volume":"37","author":"Ji","year":"2021","journal-title":"Bioinformatics"},{"key":"2025031903123372300_btaf041-B19","doi-asserted-by":"crossref","first-page":"535","DOI":"10.1109\/TBDATA.2019.2921572","article-title":"Billion-scale similarity search with gpus","volume":"7","author":"Johnson","year":"2021","journal-title":"IEEE Trans Big Data"},{"key":"2025031903123372300_btaf041-B20","author":"Kille"},{"key":"2025031903123372300_btaf041-B21","doi-asserted-by":"crossref","first-page":"421","DOI":"10.1093\/bioinformatics\/bty648","article-title":"Scaling read aligners to hundreds of threads on general-purpose processors","volume":"35","author":"Langmead","year":"2019","journal-title":"Bioinformatics"},{"key":"2025031903123372300_btaf041-B22","first-page":"9459","article-title":"Retrieval-augmented generation for knowledge-intensive nlp tasks","volume":"33","author":"Lewis","year":"2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"2025031903123372300_btaf041-B23","doi-asserted-by":"crossref","first-page":"3094","DOI":"10.1093\/bioinformatics\/bty191","article-title":"Minimap2: pairwise alignment for nucleotide sequences","volume":"34","author":"Li","year":"2018","journal-title":"Bioinformatics"},{"key":"2025031903123372300_btaf041-B24","doi-asserted-by":"crossref","first-page":"381","DOI":"10.3389\/fgene.2014.00381","article-title":"Mappability and read length","volume":"5","author":"Li","year":"2014","journal-title":"Front Genet"},{"key":"2025031903123372300_btaf041-B25","doi-asserted-by":"crossref","first-page":"312","DOI":"10.1038\/s41586-023-05896-x","article-title":"A draft human pangenome reference","volume":"617","author":"Liao","year":"2023","journal-title":"Nature"},{"key":"2025031903123372300_btaf041-B26","first-page":"824","volume-title":"IEEE Trans Pattern Anal Mach Intell","author":"Malkov","year":"2018"},{"key":"2025031903123372300_btaf041-B27","doi-asserted-by":"crossref","first-page":"935","DOI":"10.1137\/0222058","article-title":"Suffix arrays: a new method for on-line string searches","volume":"22","author":"Manber","year":"1993","journal-title":"SIAM J Comput"},{"key":"2025031903123372300_btaf041-B28","doi-asserted-by":"crossref","first-page":"i127","DOI":"10.1093\/bioinformatics\/btz354","article-title":"Locality-sensitive hashing for the edit distance","volume":"35","author":"Mar\u00e7ais","year":"2019","journal-title":"Bioinformatics"},{"key":"2025031903123372300_btaf041-B29","author":"McInnes","year":"2018"},{"key":"2025031903123372300_btaf041-B30","first-page":"2014","author":"Muennighoff","year":"2023"},{"key":"2025031903123372300_btaf041-B31","first-page":"43177","volume-title":"Adv Neural Inf Process Syst","year":"2023"},{"key":"2025031903123372300_btaf041-B32","doi-asserted-by":"crossref","first-page":"44","DOI":"10.1126\/science.abj6987","article-title":"The complete sequence of a human genome","volume":"376","author":"Nurk","year":"2022","journal-title":"Science"},{"key":"2025031903123372300_btaf041-B33"},{"key":"2025031903123372300_btaf041-B34"},{"key":"2025031903123372300_btaf041-B35","first-page":"1","article-title":"Re-inventing willis","volume":"502","author":"Simkin","year":"2011","journal-title":"Phys Rep"},{"key":"2025031903123372300_btaf041-B36","first-page":"369","author":"Smith","year":"2019"},{"key":"2025031903123372300_btaf041-B37","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Adv Neural Inf Process Syst"},{"key":"2025031903123372300_btaf041-B38","author":"Zhou"},{"key":"2025031903123372300_btaf041-B39","first-page":"97","author":"Zhu","year":"2007"},{"key":"2025031903123372300_btaf041-B40","doi-asserted-by":"crossref","first-page":"160025","DOI":"10.1038\/sdata.2016.25","article-title":"Extensive sequencing of seven human genomes to characterize benchmark reference materials","volume":"3","author":"Zook","year":"2016","journal-title":"Sci Data"},{"key":"2025031903123372300_btaf041-B41","first-page":"683","volume-title":"Int J High Perform Comput Appl","author":"Zvyagin","year":"2023"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btaf041\/61778456\/btaf041.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/41\/3\/btaf041\/61778456\/btaf041.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/41\/3\/btaf041\/61778456\/btaf041.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,19]],"date-time":"2025-03-19T03:12:56Z","timestamp":1742353976000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btaf041\/8003678"}},"subtitle":[],"editor":[{"given":"Pier Luigi","family":"Martelli","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,2,6]]},"references-count":41,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,3,4]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btaf041","relation":{},"ISSN":["1367-4811"],"issn-type":[{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2025,3]]},"published":{"date-parts":[[2025,2,6]]},"article-number":"btaf041"}}