Format

Send to

Choose Destination
Mol Cell Proteomics. 2015 Mar;14(3):761-70. doi: 10.1074/mcp.M114.037994. Epub 2015 Jan 20.

Systematic characterization and prediction of post-translational modification cross-talk.

Author information

1
From the ‡Department of Biomedical Informatics, ‖MOE Key Laboratory of Bioinformatics and Bioinformatics Division, TNLIST/Department of Automation, Tsinghua University, Beijing 100084, China; **European Molecular Biology Laboratory, European Bioinformatics Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge, United Kingdom;
2
§Department of Biochemistry and Molecular Biology, and.
3
¶¶Department of Psychiatry and Centre for Genomic Sciences, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong, China.
4
From the ‡Department of Biomedical Informatics.
5
‖MOE Key Laboratory of Bioinformatics and Bioinformatics Division, TNLIST/Department of Automation, Tsinghua University, Beijing 100084, China;
6
From the ‡Department of Biomedical Informatics, ¶Institute of Systems Biomedicine, School of Basic Medical Sciences, Peking University Health Science Center, Beijing 100191, China; litt@hsc.pku.edu.cn.

Abstract

Post-translational modification (PTM)(1) plays an important role in regulating the functions of proteins. PTMs of multiple residues on one protein may work together to determine a functional outcome, which is known as PTM cross-talk. Identification of PTM cross-talks is an emerging theme in proteomics and has elicited great interest, but their properties remain to be systematically characterized. To this end, we collected 193 PTM cross-talk pairs in 77 human proteins from the literature and then tested location preference and co-evolution at the residue and modification levels. We found that cross-talk events preferentially occurred among nearby PTM sites, especially in disordered protein regions, and cross-talk pairs tended to co-evolve. Given the properties of PTM cross-talk pairs, a naïve Bayes classifier integrating different features was built to predict cross-talks for pairwise combination of PTM sites. By using a 10-fold cross-validation, the integrated prediction model showed an area under the receiver operating characteristic (ROC) curve of 0.833, superior to using any individual feature alone. The prediction performance was also demonstrated to be robust to the biases in the collected PTM cross-talk pairs. The integrated approach has the potential for large-scale prioritization of PTM cross-talk candidates for functional validation and was implemented as a web server available at http://bioinfo.bjmu.edu.cn/ptm-x/.

PMID:
25605461
PMCID:
PMC4349993
DOI:
10.1074/mcp.M114.037994
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for HighWire Icon for PubMed Central
Loading ...
Support Center