Integrative annotation and knowledge discovery of kinase post-translational modifications and cancer-associated mutations through federated protein ontologies and resources

• Published in: Scientific reports Vol. 8; no. 1; pp. 6518 - 16
• Main Authors: Huang, Liang-Chin, Ross, Karen E., Baffi, Timothy R., Drabkin, Harold, Kochut, Krzysztof J., Ruan, Zheng, D’Eustachio, Peter, McSkimming, Daniel, Arighi, Cecilia, Chen, Chuming, Natale, Darren A., Smith, Cynthia, Gaudet, Pascale, Newton, Alexandra C., Wu, Cathy, Kannan, Natarajan
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 25-04-2018; Nature Publishing Group
• Subjects: 42/109; 42/70; 631/114/129/2044; 631/114/2401; 631/337/458; 631/45/275; 631/67/69; 82/80; Animals; Bioinformatics; Cancer; Cell Line; Chlorocebus aethiops; CHO Cells; Computational Biology - methods; COS Cells; Cricetulus; Epidermal growth factor receptors; Gene Ontology; Genetic Variation - genetics; Genomes; Humanities and Social Sciences; Humans; Hydrophobicity; Informatics; Kinases; Mice; multidisciplinary; Mutation; Mutation - genetics; Neoplasms - genetics; Ontology; Phosphorylation; Phosphorylation - genetics; Post-translation; Protein kinase; Protein kinase C; Protein Kinases - genetics; Protein Processing, Post-Translational - genetics; Proteins; Proteins - genetics; Science; Science (multidisciplinary); Translation; Translocation
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-018-24457-1

Many bioinformatics resources with unique perspectives on the protein landscape are currently available. However, generating new knowledge from these resources requires interoperable workflows that support cross-resource queries. In this study, we employ federated queries linking information from the Protein Kinase Ontology, iPTMnet, Protein Ontology, neXtProt, and the Mouse Genome Informatics to identify key knowledge gaps in the functional coverage of the human kinome and prioritize understudied kinases, cancer variants and post-translational modifications (PTMs) for functional studies. We identify 32 functional domains enriched in cancer variants and PTMs and generate mechanistic hypotheses on overlapping variant and PTM sites by aggregating information at the residue, protein, pathway and species level from these resources. We experimentally test the hypothesis that S768 phosphorylation in the C-helix of EGFR is inhibitory by showing that oncogenic variants altering S768 phosphorylation increase basal EGFR activity. In contrast, oncogenic variants altering conserved phosphorylation sites in the ‘hydrophobic motif’ of PKCβII (S660F and S660C) are loss-of-function in that they reduce kinase activity and enhance membrane translocation. Our studies provide a framework for integrative, consistent, and reproducible annotation of the cancer kinomes.