文献类型：期刊文献

英文题名：DGCPPISP: a PPI site prediction model based on dynamic graph convolutional network and two-stage transfer learning

作者：Feng, Zijian[1,2];Huang, Weihong[1,2];Li, Haohao[2];Zhu, Hancan[3];Kang, Yanlei[1];Li, Zhong[1,2]

机构：[1]Huzhou Univ, Sch Informat Engn, Zhejiang Prov Key Lab Smart Management & Applicat, Huzhou 313000, Zhejiang, Peoples R China;[2]Zhejiang Sci Tech Univ, Coll Sci, Hangzhou 310018, Zhejiang, Peoples R China;[3]Shaoxing Univ, Sch Math Phys & Informat, Shaoxing 312000, Zhejiang, Peoples R China

年份：2024

卷号：25

期号：1

外文期刊名：BMC BIOINFORMATICS

收录：SCI-EXPANDED(收录号：WOS:001281494200002)、、EI(收录号：20243216810141)、Scopus(收录号：2-s2.0-85200232503)、WOS

基金：Not applicable.DAS:The source code and data information is publicly available at https://github.com/Mrfengdashen/DGCPPISP.

语种：英文

外文关键词：PPI site prediction; Graph convolutional network; Transfer learning

外文摘要：BackgroundProteins play a pivotal role in the diverse array of biological processes, making the precise prediction of protein-protein interaction (PPI) sites critical to numerous disciplines including biology, medicine and pharmacy. While deep learning methods have progressively been implemented for the prediction of PPI sites within proteins, the task of enhancing their predictive performance remains an arduous challenge.ResultsIn this paper, we propose a novel PPI site prediction model (DGCPPISP) based on a dynamic graph convolutional neural network and a two-stage transfer learning strategy. Initially, we implement the transfer learning from dual perspectives, namely feature input and model training that serve to supply efficacious prior knowledge for our model. Subsequently, we construct a network designed for the second stage of training, which is built on the foundation of dynamic graph convolution.ConclusionsTo evaluate its effectiveness, the performance of the DGCPPISP model is scrutinized using two benchmark datasets. The ensuing results demonstrate that DGCPPISP outshines competing methods in terms of performance. Specifically, DGCPPISP surpasses the second-best method, EGRET, by margins of 5.9%, 10.1%, and 13.3% for F1-measure, AUPRC, and MCC metrics respectively on Dset_186_72_PDB164. Similarly, on Dset_331, it eclipses the performance of the runner-up method, HN-PPISP, by 14.5%, 19.8%, and 29.9% respectively.