EUP: Enhanced cross-species prediction of ubiquitination sites via a conditional variational autoencoder network based on ESM2

Ubiquitination is critical in biomedical research. Predicting ubiquitination sites based on deep learning model have advanced the study of ubiquitination. However, traditional supervised model limits in the scenarios where labels are scarcity across species. To address this issue, we introduce EUP,...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:PLoS computational biology Jg. 21; H. 7; S. e1013268
Hauptverfasser: Liu, Junhao, Luo, Zeyu, Wang, Rui, Li, Xin, Sun, Yawen, Chen, Zongqing, Zhang, Yu-Juan
Format: Journal Article
Sprache:Englisch
Veröffentlicht: United States Public Library of Science 16.07.2025
Public Library of Science (PLoS)
Schlagworte:
Animals
Autoencoder
Biology and Life Sciences
Computational Biology - methods
Deep Learning
Lysine
Lysine - metabolism
Machine learning
Physical Sciences
Research and Analysis Methods
Social Sciences
Software
Species Specificity
Ubiquitin
Ubiquitin-proteasome system
Ubiquitination
China
ISSN:1553-7358, 1553-734X, 1553-7358
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Ubiquitination is critical in biomedical research. Predicting ubiquitination sites based on deep learning model have advanced the study of ubiquitination. However, traditional supervised model limits in the scenarios where labels are scarcity across species. To address this issue, we introduce EUP, an online webserver for ubiquitination prediction and model interpretation for multi-species. EUP is constructed by extracting lysine site-dependent features from pretrained language model ESM2. Then, utilizing conditional variational inference to reduce the ESM2 features to a lower-dimensional latent representation. By constructing downstream models built on this latent feature representation, EUP exhibited superior performance in predicting ubiquitination sites across species, while maintaining low inference latency. Furthermore, key features for predicting ubiquitination sites were identified across animals, plants, and microbes. The identification of shared key features that capture evolutionarily conserved traits enhances the interpretability of the EUP model for ubiquitination prediction. EUP is free and available at ( https://eup.aibtit.com/ ).
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Junhao Liu, Zeyu Luo and Rui Wang contributed equally to this research.
The authors declare that no competing interests exist.
ISSN:1553-7358
1553-734X
1553-7358
DOI:10.1371/journal.pcbi.1013268