GPS-Uber: a hybrid-learning framework for prediction of general and E3-specific lysine ubiquitination sites

Wang, Chenwei
Tan, Xiaodan
Tang, Dachao
Gou, Yujie
Han, Cheng
Ning, Wanshan
Lin, Shaofeng
Zhang, Weizhi
Chen, Miaomiao
Peng, Di
Xue, Yu

¹Department of Bioinformatics and Systems Biology, MOE Key Laboratory of Molecular Biophysics, Hubei Bioinformatics and Molecular Imaging Key Laboratory, Center for Artificial Intelligence Biology, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan, Hubei 430074, China

Corresponding author: Yu Xue, Department of Bioinformatics and Systems Biology, MOE Key Laboratory of Molecular Biophysics, Hubei Bioinformatics and Molecular Imaging Key Laboratory, Center for Artificial Intelligence Biology, College of Life Science and Technology, Huazhong University of Science and Technology, Wuhan, Hubei 430074, China. Tel.: Phone: +86-27-87793903; Fax: Fax: +86-27-87793172; E-mail: e-mail; [email protected].

^*Chenwei Wang and Xiaodan Tan authors contributed equally to this work.

Received September 3, 2021; revision received December 11, 2021; Accepted December 14, 2021

Chenwei Wang is a postdoc scientist at Huazhong University of Science and Technology. He mainly focuses on the development of new algorithms to predict functional PTM events from multi-omic data. He developed a number of computational methods including iCMod, cMAK, iFPS and iFIP to predict functional kinases, substrates and interacting partners involved in regulating circadian, autophagy and ageing. He was also a major developer of EPSD for collecting known protein phosphorylation sites in eukaryotes.

Xiaodan Tan is a master student at Huazhong University of Science and Technology. She mainly focuses on the collection and annotation of protein lysine modifications.

Dachao Tang is a PhD student at Huazhong University of Science and Technology. His major research is identification of potentially important proteins involved in regulating autophagy.

Yujie Gou is a PhD student at Huazhong University of Science and Technology. She focuses on the development of new deep learning frameworks to process biomedical imagingdata.

Cheng Han is a PhD student at Huazhong University of Science and Technology. He is working on the computational prediction of PTM sites.

Wanshan Ning is a postdoc scientist at Huazhong University of Science and Technology. His major research interest is focused on using artificial intelligence methods to analyze sequence, multi-omics and imaging data. He built a new hybrid-learning architecture named HybridSucc for predicting general and species-specific succinylation sites. He also developed GPS-Palm to predict palmitoylation sites, HUST-19 to predict COVID-19 clinical outcomes and POC-19 to prioritize protein biomarkers of COVID-19.

Shaofeng Lin is a PhD student at Huazhong University of Science and Technology. His major research interest is focused on the integration of PTM data. He was a major developer of EPSD. He was also a major developer of iUUCD 2.0, a database of ubiquitin and ubiquitin-like conjugations.

Weizhi Zhang is a PhD student at Huazhong University of Science and Technology. He mainly focuses on developing machine learning algorithms based on multi-omic data. He developed a new method named iCAL to predict cancer mutations that change autophagy selectivity.

Miaomiao Chen is a PhD student at Huazhong University of Science and Technology. She is working on the prediction of phosphorylation sites using deep learning algorithms.

Di Peng is a postdoc scientist at Huazhong University of Science and Technology. His major research interests are focused on experimentally discovering new PTM regulators, substrates and sites in the regulation of diverse biological processes, with the combination of computational predictions.

Yu Xue is a professor at Huazhong University of Science and Technology. He has started to work in the field of PTM Bioinformatics since 2004. He is interested in using both computational and experimental approaches to exploit how functional PTM events can be precisely orchestrated to regulate various biological processes, such as autophagy, circadian and cell fate determination. He is also involved in the establishment of a new interdisciplinary field, artificial intelligence biology (AIBIO) in China.

Briefings in Bioinformatics 23(2):p bbab574, March 2022. | DOI: 10.1093/bib/bbab574

Abstract

As an important post-translational modification, lysine ubiquitination participates in numerous biological processes and is involved in human diseases, whereas the site specificity of ubiquitination is mainly decided by ubiquitin-protein ligases (E3s). Although numerous ubiquitination predictors have been developed, computational prediction of E3-specific ubiquitination sites is still a great challenge. Here, we carefully reviewed the existing tools for the prediction of general ubiquitination sites. Also, we developed a tool named GPS-Uber for the prediction of general and E3-specific ubiquitination sites. From the literature, we manually collected 1311 experimentally identified site-specific E3-substrate relations, which were classified into different clusters based on corresponding E3s at different levels. To predict general ubiquitination sites, we integrated 10 types of sequence and structure features, as well as three types of algorithms including penalized logistic regression, deep neural network and convolutional neural network. Compared with other existing tools, the general model in GPS-Uber exhibited a highly competitive accuracy, with an area under curve values of 0.7649. Then, transfer learning was adopted for each E3 cluster to construct E3-specific models, and in total 112 individual E3-specific predictors were implemented. Using GPS-Uber, we conducted a systematic prediction of human cancer-associated ubiquitination events, which could be helpful for further experimental consideration. GPS-Uber will be regularly updated, and its online service is free for academic research at http://gpsuber.biocuckoo.cn/.