{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T13:15:32Z","timestamp":1783689332500,"version":"3.55.0"},"reference-count":11,"publisher":"Oxford University Press (OUP)","issue":"Supplement_1","license":[{"start":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T00:00:00Z","timestamp":1783382400000},"content-version":"vor","delay-in-days":6,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001804","name":"Canada Research Chair","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100001804","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Machine Learning for Genomics and Healthcare","award":["CRC-2021\u201300547"],"award-info":[{"award-number":["CRC-2021\u201300547"]}]},{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council","doi-asserted-by":"crossref","award":["RGPIN-2016\u201305174"],"award-info":[{"award-number":["RGPIN-2016\u201305174"]}],"id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100000024","name":"CIHR","doi-asserted-by":"publisher","award":["PJT-540722"],"award-info":[{"award-number":["PJT-540722"]}],"id":[{"id":"10.13039\/501100000024","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Fonds de recherche du Qu\u00e9bec\u2013Nature et technologies","award":["336340"],"award-info":[{"award-number":["336340"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,7,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Genome sequence-to-function (S2F) models are widely used to interpret base-resolution functional genomics assays. Most S2F models are trained and evaluated against observed counts and profile-shapes using statistical objectives and fidelity metrics. These choices are well motivated, but they are DNA-independent. At the same time, experimental measurements arise from DNA-dependent assays with distinct characteristics. This mismatch motivates a complementary DNA-aware evaluation of S2F-predicted and experimental functional genomic tracks.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We study DNA-dependency of experimental and S2F-predicted tracks using track-conditional genome language models (cgLMs). cgLMs predict masked nucleotides from a conditioning track under controlled DNA visibility. Across ATAC-seq and TF ChIP-seq peaks from GM12878 and K562, cgLM-probing reveals a consistent masked DNA-decodability gap between many experimental and S2F-predicted tracks. In particular, single-task (e.g. BPNet) and multi-task (e.g. AlphaGenome) S2F-predicted tracks enabled cgLMs to recover masked nucleotides with significantly higher accuracy and confidence than matched experimental tracks. Analyses of nonpeak and dinucleotide-shuffled sequences show that this gap is not confined to peaks and is not captured by standard DNA-agnostic profile-shape fidelity metrics alone. ChromBPNet Tn5-denoised predictions were an exception and behaved closer to the experimental regime, suggesting that staged training may reduce the gap. We then convert this diagnostic into a critic-derived objective, DNA-dependency matching (DDM), using a frozen multi-headed cgLM critic. We introduce Critic-Guided Profile-Shape Editing (CGPSE), a preliminary post hoc debiasing framework for frozen S2F models. In GM12878 ATAC-seq, CGPSE partially reduces the masked DNA-decodability gap for AlphaGenome and BPNet predictions, while exposing a tradeoff with profile-shape fidelity.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and Implementation<\/jats:title>\n                    <jats:p>https:\/\/github.com\/li-lab-mcgill\/dna-aware-s2f-eval.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag266","type":"journal-article","created":{"date-parts":[[2026,7,7]],"date-time":"2026-07-07T16:31:12Z","timestamp":1783441872000},"source":"Crossref","is-referenced-by-count":0,"title":["DNA-aware evaluation and debiasing of sequence-to-function models"],"prefix":"10.1093","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7626-7647","authenticated-orcid":false,"given":"Doruk","family":"Cakmakci","sequence":"first","affiliation":[{"name":"School of Computer Science, McGill University , Montreal, QC H3A 0E9,","place":["Canada"]},{"name":"Mila\u2014Quebec AI Institute , Montr\u00e9al, QC H2S 3H1,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3844-4865","authenticated-orcid":false,"given":"Yue","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science, McGill University , Montreal, QC H3A 0E9,","place":["Canada"]},{"name":"Mila\u2014Quebec AI Institute , Montr\u00e9al, QC H2S 3H1,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,7,7]]},"reference":[{"key":"2026071008251461100_btag266-B1","doi-asserted-by":"crossref","first-page":"1196","DOI":"10.1038\/s41592-021-01252-x","article-title":"Effective gene expression prediction from sequence by integrating long-range interactions","volume":"18","author":"Avsec","year":"2021","journal-title":"Nat Methods"},{"key":"2026071008251461100_btag266-B2","doi-asserted-by":"crossref","first-page":"354","DOI":"10.1038\/s41588-021-00782-6","article-title":"Base-resolution models of transcription-factor binding reveal soft motif syntax","volume":"53","author":"Avsec","year":"2021","journal-title":"Nat Genet"},{"key":"2026071008251461100_btag266-B3","first-page":"1206","author":"Avsec","year":"2026"},{"key":"2026071008251461100_btag266-B4","doi-asserted-by":"crossref","first-page":"57","DOI":"10.1038\/nature11247","article-title":"An integrated encyclopedia of DNA elements in the human genome","volume":"489","author":"ENCODE Project Consortium","year":"2012","journal-title":"Nature"},{"key":"2026071008251461100_btag266-B5","doi-asserted-by":"crossref","first-page":"949","DOI":"10.1038\/s41588-024-02053-6","article-title":"Predicting RNA-SEQ coverage from DNA sequence as a unifying model of gene regulation","volume":"57","author":"Linder","year":"2025","journal-title":"Nat Genet"},{"key":"2026071008251461100_btag266-B6","doi-asserted-by":"publisher","author":"Liu","year":"2025","DOI":"10.1101\/2025.04.30.651381"},{"key":"2026071008251461100_btag266-B7","doi-asserted-by":"publisher","author":"Nair","year":"2025","DOI":"10.1101\/2025.11.06.687036"},{"key":"2026071008251461100_btag266-B8","doi-asserted-by":"publisher","first-page":"2024","DOI":"10.1101\/2024.12.25.630221","author":"Pampari","year":"2025"},{"key":"2026071008251461100_btag266-B9","doi-asserted-by":"crossref","first-page":"3942","DOI":"10.1609\/aaai.v32i1.11671","article-title":"Film: visual reasoning with a general conditioning layer","volume":"32","author":"Perez","year":"2018","journal-title":"In Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"2026071008251461100_btag266-B10","doi-asserted-by":"publisher","author":"Schreiber","year":"2025","DOI":"10.1101\/2025.08.08.669296"},{"key":"2026071008251461100_btag266-B11","doi-asserted-by":"publisher","author":"Schreiber","year":"2025","DOI":"10.1101\/2025.04.22.650035"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/Supplement_1\/btag266\/68709275\/btag266.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/Supplement_1\/btag266\/68709275\/btag266.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,10]],"date-time":"2026-07-10T12:25:26Z","timestamp":1783686326000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag266\/8726307"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,1]]},"references-count":11,"journal-issue":{"issue":"Supplement_1","published-print":{"date-parts":[[2026,7,1]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag266","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,7]]},"published":{"date-parts":[[2026,7,1]]},"article-number":"btag266"}}