Author(s)

Zheyao Gao

Corresponding Author

Zheyao Gao

Abstract

Phosphorylated proteomics is a discipline that studies all phosphorylated proteins involved in the life process and plays an important role in proteomics. With the advancement of technology, there have been many experimental methods for protein phosphorylation and bioinformatics-based computing strategies. The application of data-driven machine learning methods in phosphate-catalyzed research has matured and has become the mainstream method in this field. This paper mainly summarizes the main methods and principles of machine learning, including Bayesian Decision Theory (BDT), Random Forests (RFs), AdaBoost, Support Vector Machines (SVMs)and etc. Finally, the application of these methods in the field of phosphorylation is reviewed in this article.

Keywords

Machine learning phosphorylation, bayesian decision theory, feature selection, phosphoproteomics.