Abstract
Phosphorylation is the most important type of protein post-translational modification. Accordingly, reliable identification of kinase-mediated phosphorylation has important implications for functional annotation of phosphorylated substrates and characterization of cellular signalling pathways. The local sequence context surrounding potential phosphorylation sites is considered to harbour the most relevant information for phosphorylation site prediction models. However, currently there is a lack of condensed vector representation for this important contextual information, despite the presence of varying residue-level features that can be constructed from sequence homology profiles, structural information, and physicochemical properties. To address this issue, we present PhosContext2vec which is a distributed representation of residue-level sequence contexts for potential phosphorylation sites and demonstrate its application in both general and kinase-specific phosphorylation site predictions. Benchmarking experiments indicate that PhosContext2vec could achieve promising predictive performance compared with several other existing methods for phosphorylation site prediction. We envisage that PhosContext2vec, as a new sequence context representation, can be used in combination with other informative residue-level features to improve the classification performance in a number of related bioinformatics tasks that require appropriate residue-level feature vector representation and extraction. The web server of PhosContext2vec is publicly available at http://phoscontext2vec.erc.monash.edu/.
Highlights
Phosphorylation is the most common type of protein post-translational modification (PTM), which plays an important role in almost all cellular processes in eukaryotes[1,2]
We evaluated the prediction performance of constructed models using four measurements, including sensitivity (Eq 1), specificity (Eq 2), the Matthews coefficients of correlation (MCC) (Eq 3), and the area under the ROC curve (AUC)[9]
We conducted comprehensive benchmarking experiments to compare the performance of PhosContext2vec and seven existing predictors for phosphorylation site prediction
Summary
For kinase-specific phosphorylation site prediction, we combined phosphorylation site data obtained from Phospho.ELM18 and UniProt[20] and constructed training and independent test datasets according to the following steps. For each of the three phosphorylation site types S, T, and Y, and each of the five kinase families AGC/PKC, AGC/PKA, CMGC/CK2, CMGC/CDK, and TK/SRC, an independent SVM model was constructed based on the proposed contextual feature vector in conjunction with six other residue-level feature groups. We generated the distributed representation of residue-level sequence contexts within specific contextual window sizes This residue-level sequence context was referred to as PSP (m, n) in previous studies, namely the Phosphorylation Site Peptide, which was composed of m upstream residues and n downstream residues of the target site[21].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
