Computational prediction of phosphorylation sites (p-sites) with their cognate protein kinases (PKs) is greatly helpful for further experimental design. Previously, our group developed and maintained a series of PK-specific p-site predictors. In 2004, we developed a novel algorithm of group-based phosphorylation site predicting and scoring (GPS) 1.0, based on a hypothesis of short similar peptides exhibiting similar biological functions. We considerably refined the algorithm and constructed an online service of GPS 1.1, which could predict p-sites for 71 PK clusters. Later, we presented GPS 2.0 and 2.1 (renamed as Group-based Prediction System), in which two methods of matrix mutation (MaM) and motif length selection (MLS) were designed to improve the prediction accuracy, whereas the scoring strategy of GPS 1.0 & 1.1 was not changed. Using 3,417 known PK-specific p-sites as the training data set, GPS 2.1 contained 213 individual predictors, and could hierarchically predict specific p-sites for 408 human PKs. We also developed GPS 2.2, 3.0 and 4.0 algorithms, which were used for the prediction of other types of post-translational modification (PTM) sites but not phosphorylation. Since its release, GPS series have been widely used and prompted many biological discoveries.

Recently, we released GPS 5.0, by developing two novel methods of position weight determination (PWD) and scoring matrix optimization (SMO) to improve the performance for predicting kinase-specific p-sites. Besides serine/threonine or tyrosine kinases, the prediction of dual-specificity kinase-specific p-sites was also supported. In the classical module of GPS 5.0, 617 individual predictors were constructed for predicting p-sites of 479 human PKs. To extend the application of GPS 5.0, a species-specific module was implemented to predict kinase-specific p-sites for 44,795 PKs in 161 eukaryotes. In addition, structural features such as secondary structures, surface accessibilities and disorder regions were annotated for the predicted p-sites.

The GPS 5.0 is freely available for academic research at: http://gps.biocuckoo.cn.

This website is linked in ExPASy Proteomics Tools page.

For publication of results please cite the following article:

GPS 5.0: An update on the prediction of kinase-specific phosphorylation sites in proteins
Chenwei Wang, Haodong Xu, Shaofeng Lin, Wankun Deng, Jiaqi Zhou, Ying Zhang, Ying Shi, Di Peng, Yu Xue*. Genomics, Proteomics & Bioinformatics. 2020. S1672-0229(20)30027-9.

[Abstract] [Full Text] [Supplemental Data]

GPS 2.1: enhanced prediction of kinase-specific phosphorylation sites with an algorithm of motif length selection. Yu Xue, Zexian Liu, Jun Cao, Qian Ma, Xinjiao Gao, Qingqi Wang, Changjiang Jin, Yanhong Zhou, Longping Wen, and Jian Ren. Protein Engineering, Design and Selection (2011);24 (3): 255-260

[Abstract] [Full Text] [Supplemental Data]

GPS 2.0, a Tool to Predict Kinase-specific Phosphorylation Sites in Hierarchy. Yu Xue#, Jian Ren#, Xinjiao Gao, Changjiang Jin, Longping Wen*, and Xuebiao Yao*. Mol Cell Proteomics. 2008; 7: 1598-1608

[Abstract] [Full Text] [Supplemental Data]

GPS: a comprehensive www server for phosphorylation sites prediction. Yu Xue, Fengfeng Zhou, Minjie Zhu, Kashif Ahmed, Guoliang Chen and Xuebiao Yao*. Nucleic Acids Res. 2005; 33 (suppl 2): W184-W187

[Abstract] [Full Text]