3 – Phosphorylation motif discovery

Phosphorylation motifs are consensus amino acid patterns around the phosphorylation sites. The discovery of these types of motifs can provide more precise rules to guide the recognition of kinase–substrate interactions. To date, people have gathered large numbers of experimentally verified phosphorylation sites, making it possible to conduct the substrate-driven motif extraction with data mining techniques. In this chapter, we first provide the background and the problem description of the phosphorylation motif discovery. Then, we discuss how this motif discovery problem is equivalent to a discriminative pattern mining problem. Consequently, the key steps in the data mining process for extracting phosphorylation motifs are described with sufficient details. Additionally, we give particular emphasis on representative phosphorylation motif mining algorithms and permutation-based phosphorylation motif validation methods. Finally, we conclude the chapter with a discussion on the limitations of current solutions, with a highlight on future research directions.