用于细胞类型不可知的调节预测的多模态变压器。

IF 11.1 Q1 CELL BIOLOGY
Cell genomics Pub Date : 2025-02-12 Epub Date: 2025-01-29 DOI:10.1016/j.xgen.2025.100762
Nauman Javed, Thomas Weingarten, Arijit Sehanobish, Adam Roberts, Avinava Dubey, Krzysztof Choromanski, Bradley E Bernstein
{"title":"用于细胞类型不可知的调节预测的多模态变压器。","authors":"Nauman Javed, Thomas Weingarten, Arijit Sehanobish, Adam Roberts, Avinava Dubey, Krzysztof Choromanski, Bradley E Bernstein","doi":"10.1016/j.xgen.2025.100762","DOIUrl":null,"url":null,"abstract":"<p><p>Sequence-based deep learning models have emerged as powerful tools for deciphering the cis-regulatory grammar of the human genome but cannot generalize to unobserved cellular contexts. Here, we present EpiBERT, a multi-modal transformer that learns generalizable representations of genomic sequence and cell type-specific chromatin accessibility through a masked accessibility-based pre-training objective. Following pre-training, EpiBERT can be fine-tuned for gene expression prediction, achieving accuracy comparable to the sequence-only Enformer model, while also being able to generalize to unobserved cell states. The learned representations are interpretable and useful for predicting chromatin accessibility quantitative trait loci (caQTLs), regulatory motifs, and enhancer-gene links. Our work represents a step toward improving the generalization of sequence-based deep neural networks in regulatory genomics.</p>","PeriodicalId":72539,"journal":{"name":"Cell genomics","volume":" ","pages":"100762"},"PeriodicalIF":11.1000,"publicationDate":"2025-02-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11872434/pdf/","citationCount":"0","resultStr":"{\"title\":\"A multi-modal transformer for cell type-agnostic regulatory predictions.\",\"authors\":\"Nauman Javed, Thomas Weingarten, Arijit Sehanobish, Adam Roberts, Avinava Dubey, Krzysztof Choromanski, Bradley E Bernstein\",\"doi\":\"10.1016/j.xgen.2025.100762\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Sequence-based deep learning models have emerged as powerful tools for deciphering the cis-regulatory grammar of the human genome but cannot generalize to unobserved cellular contexts. Here, we present EpiBERT, a multi-modal transformer that learns generalizable representations of genomic sequence and cell type-specific chromatin accessibility through a masked accessibility-based pre-training objective. Following pre-training, EpiBERT can be fine-tuned for gene expression prediction, achieving accuracy comparable to the sequence-only Enformer model, while also being able to generalize to unobserved cell states. The learned representations are interpretable and useful for predicting chromatin accessibility quantitative trait loci (caQTLs), regulatory motifs, and enhancer-gene links. Our work represents a step toward improving the generalization of sequence-based deep neural networks in regulatory genomics.</p>\",\"PeriodicalId\":72539,\"journal\":{\"name\":\"Cell genomics\",\"volume\":\" \",\"pages\":\"100762\"},\"PeriodicalIF\":11.1000,\"publicationDate\":\"2025-02-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11872434/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Cell genomics\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1016/j.xgen.2025.100762\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/1/29 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q1\",\"JCRName\":\"CELL BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Cell genomics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1016/j.xgen.2025.100762","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/1/29 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"CELL BIOLOGY","Score":null,"Total":0}
引用次数: 0

摘要

基于序列的深度学习模型已经成为破译人类基因组顺式调控语法的强大工具,但不能推广到未观察到的细胞环境。在这里,我们提出了EpiBERT,一个多模式转换器,通过一个基于可达性的预训练目标来学习基因组序列和细胞类型特异性染色质可达性的可泛化表示。在预训练之后,EpiBERT可以对基因表达预测进行微调,达到与仅序列Enformer模型相当的准确性,同时也能够推广到未观察到的细胞状态。学习表征是可解释的,并且有助于预测染色质可达性定量性状位点(caqtl)、调控基序和增强基因链接。我们的工作代表着朝着改善基于序列的深度神经网络在调控基因组学中的推广迈出了一步。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
A multi-modal transformer for cell type-agnostic regulatory predictions.

Sequence-based deep learning models have emerged as powerful tools for deciphering the cis-regulatory grammar of the human genome but cannot generalize to unobserved cellular contexts. Here, we present EpiBERT, a multi-modal transformer that learns generalizable representations of genomic sequence and cell type-specific chromatin accessibility through a masked accessibility-based pre-training objective. Following pre-training, EpiBERT can be fine-tuned for gene expression prediction, achieving accuracy comparable to the sequence-only Enformer model, while also being able to generalize to unobserved cell states. The learned representations are interpretable and useful for predicting chromatin accessibility quantitative trait loci (caQTLs), regulatory motifs, and enhancer-gene links. Our work represents a step toward improving the generalization of sequence-based deep neural networks in regulatory genomics.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
7.10
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信