Predicting tyrosine phosphorylation from site-modification network

Phosphorylation plays a great role in regulating variety of cellular processes. Identification of the tyrosine phosphorylation is helpful for understanding the post-translational modifications regulation processes. Although a number of computational approaches have been developed, most of them only consider local sequence information, and currently few studies are concerned about the site with in situ PTM information, which refers to different types of PTMs occurring on the same modification site. In this study, we introduce a novel method (SMNBI) based on a site-modification network, which is constructed by integrating the site with in situ PTM information to predict tyrosine phosphorylation. In order to verify the effectiveness of our SMNBI method, we compare it with Protein-based collaborative filtering (ProteinCF). The results clearly show the superior performance of the SMNBI. Furthermore, to verify the importance of the site-modification network, we compare it with SVM and Bayesian decision theory based on local sequence information. These results demonstrate the site-modification network is useful for predicting tyrosine phosphorylation.