GPS 2.1: enhanced prediction of kinase-specific phosphorylation sites with an algorithm of motif length selection

作者：Yu Xue, Zexian Liu, Jun Cao, Qian Ma, Xiang Gao, R. Wang, C. Jin, Yanhong Zhou, Longping Wen, Jian Ren · 发表于：Protein Engineering Design and Selection · 年份：2010 · DOI：10.1093/protein/gzq094 · 被引用次数：264 · 研究领域：Machine Learning in Bioinformatics、Glycosylation and Glycoproteins Research、Genomics and Phylogenetic Studies

As the most important post-translational modification of proteins, phosphorylation plays essential roles in all aspects of biological processes. Besides experimental approaches, computational prediction of phosphorylated proteins with their kinase-specific phosphorylation sites has also emerged as a popular strategy, for its low-cost, fast-speed and convenience. In this work, we developed a kinase-specific phosphorylation sites predictor of GPS 2.1 (Group-based Prediction System), with a novel but simple approach of motif length selection (MLS). By this approach, the robustness of the prediction system was greatly improved. All algorithms in GPS old versions were also reserved and integrated in GPS 2.1. The online service and local packages of GPS 2.1 were implemented in JAVA 1.5 (J2SE 5.0) and freely available for academic researches at: http://gps.biocuckoo.org.