GPSD: a hybrid learning framework for the prediction of phosphatase-specific dephosphorylation sites
Open Access
- 22 November 2024
- journal article
- research article
- Published by Oxford University Press (OUP) in Briefings in Bioinformatics
- Vol. 26 (1)
Abstract
Protein phosphorylation is dynamically and reversibly regulated by protein kinases and protein phosphatases, and plays an essential role in orchestrating a wide range of biological processes. Although a number of tools have been developed for predicting kinase-specific phosphorylation sites (p-sites), computational prediction of phosphatase-specific dephosphorylation sites remains to be a great challenge. In this study, we manually curated 4393 experimentally identified site-specific phosphatase–substrate relationships for 3463 dephosphorylation sites occurring on phosphoserine, phosphothreonine, and/or phosphotyrosine residues, from the literature and public databases. Then, we developed a hybrid learning framework, the group-based prediction system for the prediction of phosphatase-specific dephosphorylation sites (GPSD). For model training, we integrated 10 types of sequence features and utilized three types of machine learning methods, including penalized logistic regression, deep neural networks, and transformer neural networks. First, a pretrained model was constructed using 561 416 nonredundant p-sites and then fine-tuned to generate computational models for predicting general dephosphorylation sites. In addition, 103 individual phosphatase-specific predictors were constructed via transfer learning and meta-learning. For site prediction, one or multiple protein sequences in FASTA format could be inputted, and the prediction results will be shown together with additional annotations, such as protein–protein interactions, structural information, and disorder propensity. The online service of GPSD is freely available at https://gpsd.biocuckoo.cn/. We believe that GPSD can serve as a valuable tool for further analysis of dephosphorylation.Keywords
Funding Information
- National Key Research and Development Program of China (2022YFC2704304, 2021YFF0702000)
- Natural Science Foundation of China (32341020, 32341021, 31930021)
- Hubei Innovation Group Project (2021CFA005)
- Hubei Province Postdoctoral Outstanding Talent Tracking Support Program
- Interdisciplinary Research Program of HUST (2023JCYJ010, 2024JCYJ013)
- Research Core Facilities for Life Science
This publication has 58 references indexed in Scilit:
- IUPred: web server for the prediction of intrinsically unstructured regions of proteins based on estimated energy contentBioinformatics, 2005
- Structural basis of protein phosphatase 1 regulationNature, 2004
- Phosphorylation of p53 at serine 37 is important for transcriptional activity and regulation in response to DNA damageOncogene, 2004
- The Protein Kinase Complement of the Human GenomeScience, 2002
- The SPOT technique as a tool for studying protein tyrosine phosphatase substrate specificitiesProtein Science, 2002
- Phosphorylation and dephosphorylation of histidine residues in proteinsEuropean Journal of Biochemistry, 2002
- The YRD Motif Is a Major Determinant of Substrate and Inhibitor Specificity in T-cell Protein-tyrosine PhosphataseJournal of Biological Chemistry, 2001
- Protein kinases and phosphatases: The Yin and Yang of protein phosphorylation and signalingCell, 1995
- Dephosphorylation of tau protein and Alzheimer paired helical filaments by calcmeurin and phosphatase‐2AFEBS Letters, 1993
- CONVERSION OF PHOSPHORYLASE b TO PHOSPHORYLASE a IN MUSCLE EXTRACTSJournal of Biological Chemistry, 1955