RAEPI:RAEPI:基于受限注意力机制预测促进剂与增强剂之间的相互作用

IF 3.9 2区 生物学 Q1 MATHEMATICAL & COMPUTATIONAL BIOLOGY
Wanjing Zhang, Mingyang Zhang, Min Zhu
{"title":"RAEPI:RAEPI:基于受限注意力机制预测促进剂与增强剂之间的相互作用","authors":"Wanjing Zhang, Mingyang Zhang, Min Zhu","doi":"10.1007/s12539-024-00669-0","DOIUrl":null,"url":null,"abstract":"<p><p>Enhancer-promoter interactions (EPIs) are crucial in gene transcription regulation and cell differentiation. Traditional biological experiments are costly and time-consuming, motivating the development of computational prediction methods. However, existing EPI prediction methods inadequately capture the intricate direct interactions between enhancer and promoter sequences, which limits their prediction performance to some extent. In this work, we propose an innovative attention-based approach RAEPI, which uses convolutional neural networks to extract initial features of enhancers and promoters, combined with a specially designed Restricted Attention mechanism with Query-Key-Value constrained to simulate the interactions between them for further feature extraction. To improve cross-cell line prediction, we employ a transfer learning strategy for pre-training. Furthermore, we extracted sequence motifs to evaluate the RAEPI's effectiveness from a visualization perspective. Experimental results show that RAEPI achieves competitive prediction performance to existing methods on the benchmark dataset.</p>","PeriodicalId":13670,"journal":{"name":"Interdisciplinary Sciences: Computational Life Sciences","volume":" ","pages":""},"PeriodicalIF":3.9000,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"RAEPI: Predicting Enhancer-Promoter Interactions Based on Restricted Attention Mechanism.\",\"authors\":\"Wanjing Zhang, Mingyang Zhang, Min Zhu\",\"doi\":\"10.1007/s12539-024-00669-0\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Enhancer-promoter interactions (EPIs) are crucial in gene transcription regulation and cell differentiation. Traditional biological experiments are costly and time-consuming, motivating the development of computational prediction methods. However, existing EPI prediction methods inadequately capture the intricate direct interactions between enhancer and promoter sequences, which limits their prediction performance to some extent. In this work, we propose an innovative attention-based approach RAEPI, which uses convolutional neural networks to extract initial features of enhancers and promoters, combined with a specially designed Restricted Attention mechanism with Query-Key-Value constrained to simulate the interactions between them for further feature extraction. To improve cross-cell line prediction, we employ a transfer learning strategy for pre-training. Furthermore, we extracted sequence motifs to evaluate the RAEPI's effectiveness from a visualization perspective. Experimental results show that RAEPI achieves competitive prediction performance to existing methods on the benchmark dataset.</p>\",\"PeriodicalId\":13670,\"journal\":{\"name\":\"Interdisciplinary Sciences: Computational Life Sciences\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":3.9000,\"publicationDate\":\"2024-11-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Interdisciplinary Sciences: Computational Life Sciences\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.1007/s12539-024-00669-0\",\"RegionNum\":2,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"MATHEMATICAL & COMPUTATIONAL BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Interdisciplinary Sciences: Computational Life Sciences","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1007/s12539-024-00669-0","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATHEMATICAL & COMPUTATIONAL BIOLOGY","Score":null,"Total":0}
引用次数: 0

摘要

增强子-启动子相互作用(EPIs)在基因转录调控和细胞分化中至关重要。传统的生物学实验既费钱又费时,因此人们开始开发计算预测方法。然而,现有的 EPI 预测方法不能充分捕捉增强子和启动子序列之间错综复杂的直接相互作用,这在一定程度上限制了它们的预测性能。在这项工作中,我们提出了一种基于注意力的创新方法 RAEPI,该方法利用卷积神经网络提取增强子和启动子的初始特征,并结合专门设计的限制注意力机制(Restricted Attention mechanism)和查询键值(Query-Key-Value)约束来模拟它们之间的相互作用,从而进一步提取特征。为了改进跨细胞系预测,我们采用了迁移学习策略进行预训练。此外,我们还提取了序列主题,从可视化角度评估 RAEPI 的有效性。实验结果表明,RAEPI 在基准数据集上取得了与现有方法相当的预测性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
RAEPI: Predicting Enhancer-Promoter Interactions Based on Restricted Attention Mechanism.

Enhancer-promoter interactions (EPIs) are crucial in gene transcription regulation and cell differentiation. Traditional biological experiments are costly and time-consuming, motivating the development of computational prediction methods. However, existing EPI prediction methods inadequately capture the intricate direct interactions between enhancer and promoter sequences, which limits their prediction performance to some extent. In this work, we propose an innovative attention-based approach RAEPI, which uses convolutional neural networks to extract initial features of enhancers and promoters, combined with a specially designed Restricted Attention mechanism with Query-Key-Value constrained to simulate the interactions between them for further feature extraction. To improve cross-cell line prediction, we employ a transfer learning strategy for pre-training. Furthermore, we extracted sequence motifs to evaluate the RAEPI's effectiveness from a visualization perspective. Experimental results show that RAEPI achieves competitive prediction performance to existing methods on the benchmark dataset.

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
Interdisciplinary Sciences: Computational Life Sciences
Interdisciplinary Sciences: Computational Life Sciences MATHEMATICAL & COMPUTATIONAL BIOLOGY-
CiteScore
8.60
自引率
4.20%
发文量
55
期刊介绍: Interdisciplinary Sciences--Computational Life Sciences aims to cover the most recent and outstanding developments in interdisciplinary areas of sciences, especially focusing on computational life sciences, an area that is enjoying rapid development at the forefront of scientific research and technology. The journal publishes original papers of significant general interest covering recent research and developments. Articles will be published rapidly by taking full advantage of internet technology for online submission and peer-reviewing of manuscripts, and then by publishing OnlineFirstTM through SpringerLink even before the issue is built or sent to the printer. The editorial board consists of many leading scientists with international reputation, among others, Luc Montagnier (UNESCO, France), Dennis Salahub (University of Calgary, Canada), Weitao Yang (Duke University, USA). Prof. Dongqing Wei at the Shanghai Jiatong University is appointed as the editor-in-chief; he made important contributions in bioinformatics and computational physics and is best known for his ground-breaking works on the theory of ferroelectric liquids. With the help from a team of associate editors and the editorial board, an international journal with sound reputation shall be created.
×
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