Zixin Duan, Yafeng Liang, Xin Xiu, Wenjie Ma, Hu Mei
{"title":"ResUbiNet:用于泛素化位点预测的新型深度学习架构","authors":"Zixin Duan, Yafeng Liang, Xin Xiu, Wenjie Ma, Hu Mei","doi":"10.2174/0113892029331751240820111158","DOIUrl":null,"url":null,"abstract":"Introduction: Ubiquitination, a unique post-translational modification, plays a cardinal role in diverse cellular functions such as protein degradation, signal transduction, DNA repair, and regulation of cell cycle. Method: Thus, accurate prediction of potential ubiquitination sites is an urgent requirement for exploring the ubiquitination mechanism as well as the disease pathogenesis associated with ubiquitination processes. Results: This study introduces a novel deep learning architecture, ResUbiNet, which utilized a protein language model (ProtTrans), amino acid properties, and BLOSUM62 matrix for sequence embedding and multiple state-of-the-art architectural components, i.e., transformer, multi-kernel convolution, residual connection, and squeeze-and-excitation for feature extractions. Conclusion: The results of cross-validation and external tests showed that the ResUbiNet model achieved better prediction performances in comparison with the available hCKSAAP_UbSite, RUBI, MDCapsUbi, and MusiteDeep models.","PeriodicalId":10803,"journal":{"name":"Current Genomics","volume":"7 1","pages":""},"PeriodicalIF":1.8000,"publicationDate":"2024-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"ResUbiNet: A Novel Deep Learning Architecture for Ubiquitination Site Prediction\",\"authors\":\"Zixin Duan, Yafeng Liang, Xin Xiu, Wenjie Ma, Hu Mei\",\"doi\":\"10.2174/0113892029331751240820111158\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Introduction: Ubiquitination, a unique post-translational modification, plays a cardinal role in diverse cellular functions such as protein degradation, signal transduction, DNA repair, and regulation of cell cycle. Method: Thus, accurate prediction of potential ubiquitination sites is an urgent requirement for exploring the ubiquitination mechanism as well as the disease pathogenesis associated with ubiquitination processes. Results: This study introduces a novel deep learning architecture, ResUbiNet, which utilized a protein language model (ProtTrans), amino acid properties, and BLOSUM62 matrix for sequence embedding and multiple state-of-the-art architectural components, i.e., transformer, multi-kernel convolution, residual connection, and squeeze-and-excitation for feature extractions. Conclusion: The results of cross-validation and external tests showed that the ResUbiNet model achieved better prediction performances in comparison with the available hCKSAAP_UbSite, RUBI, MDCapsUbi, and MusiteDeep models.\",\"PeriodicalId\":10803,\"journal\":{\"name\":\"Current Genomics\",\"volume\":\"7 1\",\"pages\":\"\"},\"PeriodicalIF\":1.8000,\"publicationDate\":\"2024-08-28\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Current Genomics\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://doi.org/10.2174/0113892029331751240820111158\",\"RegionNum\":4,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q4\",\"JCRName\":\"BIOCHEMISTRY & MOLECULAR BIOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Current Genomics","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.2174/0113892029331751240820111158","RegionNum":4,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q4","JCRName":"BIOCHEMISTRY & MOLECULAR BIOLOGY","Score":null,"Total":0}
ResUbiNet: A Novel Deep Learning Architecture for Ubiquitination Site Prediction
Introduction: Ubiquitination, a unique post-translational modification, plays a cardinal role in diverse cellular functions such as protein degradation, signal transduction, DNA repair, and regulation of cell cycle. Method: Thus, accurate prediction of potential ubiquitination sites is an urgent requirement for exploring the ubiquitination mechanism as well as the disease pathogenesis associated with ubiquitination processes. Results: This study introduces a novel deep learning architecture, ResUbiNet, which utilized a protein language model (ProtTrans), amino acid properties, and BLOSUM62 matrix for sequence embedding and multiple state-of-the-art architectural components, i.e., transformer, multi-kernel convolution, residual connection, and squeeze-and-excitation for feature extractions. Conclusion: The results of cross-validation and external tests showed that the ResUbiNet model achieved better prediction performances in comparison with the available hCKSAAP_UbSite, RUBI, MDCapsUbi, and MusiteDeep models.
期刊介绍:
Current Genomics is a peer-reviewed journal that provides essential reading about the latest and most important developments in genome science and related fields of research. Systems biology, systems modeling, machine learning, network inference, bioinformatics, computational biology, epigenetics, single cell genomics, extracellular vesicles, quantitative biology, and synthetic biology for the study of evolution, development, maintenance, aging and that of human health, human diseases, clinical genomics and precision medicine are topics of particular interest. The journal covers plant genomics. The journal will not consider articles dealing with breeding and livestock.
Current Genomics publishes three types of articles including:
i) Research papers from internationally-recognized experts reporting on new and original data generated at the genome scale level. Position papers dealing with new or challenging methodological approaches, whether experimental or mathematical, are greatly welcome in this section.
ii) Authoritative and comprehensive full-length or mini reviews from widely recognized experts, covering the latest developments in genome science and related fields of research such as systems biology, statistics and machine learning, quantitative biology, and precision medicine. Proposals for mini-hot topics (2-3 review papers) and full hot topics (6-8 review papers) guest edited by internationally-recognized experts are welcome in this section. Hot topic proposals should not contain original data and they should contain articles originating from at least 2 different countries.
iii) Opinion papers from internationally recognized experts addressing contemporary questions and issues in the field of genome science and systems biology and basic and clinical research practices.