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

Junhao Liu, Zeyu Luo, Rui Wang, Xin Li, Yawen Sun, Zongqing Chen and Yu-Juan Zhang

PLOS Computational Biology, 2025, vol. 21, issue 7, 1-23

Abstract: 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/).Author summary: We present EUP, an AI-powered web-based tool designed to predict ubiquitination sites on protein sequences. Ubiquitination is a critical regulatory process in cells, where small ubiquitin molecules are attached to specific lysine residues to control protein stability and function. Identifying these modification sites is essential for understanding cellular mechanisms, but experimental detection is time-consuming and often unavailable for many non-model organisms. To address this challenge, EUP leverages a pretrained protein language model to extract informative features from sequences and applies variational inference to predict likely ubiquitination sites. The tool supports prediction across animals, plants, and microbes, and highlights both conserved and species-specific patterns. This provides users with interpretable insights into how ubiquitination may vary across evolution. EUP is freely available at (https://eup.aibtit.com) and offers a practical solution for researchers studying protein regulation across diverse biological systems.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1013268 (text/html)
https://journals.plos.org/ploscompbiol/article/fil ... 13268&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1013268

DOI: 10.1371/journal.pcbi.1013268

Access Statistics for this article

More articles in PLOS Computational Biology from Public Library of Science
Bibliographic data for series maintained by ploscompbiol ().