Predicting protein-protein interactions (PPIs) is crucial for advancing drug discovery. Despite the proposal of numerous advanced computational methods, these approaches often suffer from poor usability for biologists and lack generalization. In this study, we designed a deep learning model based on a coattention mechanism that was capable of both PPI and site prediction and used this model as the foundation for PPI-CoAttNet, a user-friendly, multifunctional web server for PPI prediction. This platform provides comprehensive services for online PPI model training, PPI and site prediction, and prediction of interactions with proteins associated with highly prevalent cancers. In our test set for PPI prediction, PPI-CoAttNet achieved an AUC of 0.9841 and an F1 score of 0.9440, outperforming most state-of-the-art models. Additionally, these results are generated in real time, delivering outcomes within minutes. We also evaluated PPI-CoAttNet for downstream tasks, including novel E3 ligase scoring, demonstrating outstanding accuracy. We believe that this tool will empower researchers, especially those without computational expertise, to leverage AI for accelerating drug development.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1021/acs.jcim.4c01365 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
The ARID1A gene, frequently mutated in cancer, encodes the AT-rich interactive domain-containing protein 1A, a key component of the chromatin remodeling SWI/SNF complex. The ARID1A protein features a conserved DNA-binding domain (ARID domain) of approximately 100 residues crucial for its function. Despite the frequency of mutations, the impact on ARID1A's stability and contribution to cancer progression remains unclear.
View Article and Find Full Text PDFWhile novel deep learning and statistics-based techniques predict accurate structural models for proteins and non-coding RNA, describing their macromolecular conformations in solution is still challenging. Small-angle X-ray scattering (SAXS) in solution is an efficient technique to validate structural predictions by comparing the experimental SAXS profile with those calculated from predicted structures. There are two main challenges in comparing SAXS profiles to RNA structures: the structures often lack cations necessary for stability and charge neutralization, and a single structure inadequately represents the conformational plasticity of RNA.
View Article and Find Full Text PDFPPI-CoAttNet: A Web Server for Protein-Protein Interaction Tasks Using a Coattention Model.
J Chem Inf Model
January 2025
College of Pharmaceutical Sciences, Zhejiang University, Hangzhou 310058, China.
Predicting protein-protein interactions (PPIs) is crucial for advancing drug discovery. Despite the proposal of numerous advanced computational methods, these approaches often suffer from poor usability for biologists and lack generalization. In this study, we designed a deep learning model based on a coattention mechanism that was capable of both PPI and site prediction and used this model as the foundation for PPI-CoAttNet, a user-friendly, multifunctional web server for PPI prediction.
View Article and Find Full Text PDFEnsemble learning-based predictor for driver synonymous mutation with sequence representation.
PLoS Comput Biol
January 2025
School of Biomedical Engineering, Anhui Medical University, Hefei, China.
Synonymous mutations, once considered neutral, are now understood to have significant implications for a variety of diseases, particularly cancer. It is indispensable to identify these driver synonymous mutations in human cancers, yet current methods are constrained by data limitations. In this study, we initially investigate the impact of sequence-based features, including DNA shape, physicochemical properties and one-hot encoding of nucleotides, and deep learning-derived features from pre-trained chemical molecule language models based on BERT.
View Article and Find Full Text PDFCrypt4GH-JS: Securely storing sensitive data online with client-side encryption.
Bioinformatics
January 2025
Justus Liebig University Giessen, Hesse, 35392, Germany.
Article Synopsis
- Crypt4GH-JS is a JavaScript library that enables on-the-fly encryption for web applications without significantly affecting data transfer speeds.
- It supports secure data storage in cloud environments, addressing the growing need for flexible data security solutions.
- The library can be installed via Node Package Manager or accessed on GitHub, and users can also test it through an online demonstration site.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!