Predicting Protein–Protein Interfaces that Bind Intrinsically Disordered Protein Regions. Issue 17 (9th August 2019)

• Record Type: Journal Article
• Authors: Wong, Eric T.C.; Gsponer, Jörg
• Is Part Of: Journal of molecular biology. Volume 431:Issue 17(2019)
• Page Start: 3157
• Page End: 3178
• Publication Date: 2019-08-09
• Subjects: IDR intrinsically disordered region -- MoRF molecular recognition feature -- CRF conditional random field -- ROC receiver operating characteristic -- AUC area under curve -- TPR true positive rate -- FPR false-positive rate -- MCC Matthews correlation coefficient -- SASA solvent-accessible surface area -- rASA relative accessible surface area; protein–protein interactions -- protein interface prediction -- intrinsically disordered proteins -- molecular recognition features -- protein interface prediction benchmarking; Molecular biology -- Periodicals; Biology -- Periodicals; Biochemistry -- Periodicals; Bacteriology -- Periodicals; Molecular Biology -- Periodicals; Biochemistry -- Periodicals; Biologie moléculaire -- Périodiques; Biologie -- Périodiques; Biochimie -- Périodiques; Moleculaire biologie; Biochemistry; Biology; Molecular biology; Periodicals; 572.805
• DOI: 10.1016/j.jmb.2019.06.010
• Languages: English
• ISSNs: 0022-2836

Abstract

Abstract: A long-standing goal in biology is the complete annotation of function and structure on all protein–protein interactions, a large fraction of which is mediated by intrinsically disordered protein regions (IDRs). However, knowledge derived from experimental structures of such protein complexes is disproportionately small due, in part, to challenges in studying interactions of IDRs. Here, we introduce IDRBind, a computational method that by combining gradient boosted trees and conditional random field models predicts binding sites of IDRs with performance approaching state-of-the-art globular interface predictions, making it suitable for proteome-wide applications. Although designed and trained with a focus on molecular recognition features, which are long interaction-mediating-elements in IDRs, IDRBind also predicts the binding sites of short peptides more accurately than existing specialized predictors. Consistent with IDRBind's specificity, a comparison of protein interface categories uncovered uniform trends in multiple physicochemical properties, positioning molecular recognition feature interfaces between peptide and globular interfaces. Graphical Abstract: Unlabelled Image Highlights: New predictor (IDRBind) for protein sites that bind MoRFs Reliable prediction of protein surfaces that bind specifically disordered regions IDRBind surpasses state-of-the-art predictors for peptide interfaces. Peptide, MoRF, and globular interfaces form continuum in property space.