A Method for Determining Structure Ensemble of Large Disordered Protein: Application to a Mechanosensing Protein

• DOI: 10.1021/jacs.8b04792
• Publication Date: 2018-08-20T11:30:37Z
• Authors: Wei Liu; Xiao Liu; Guanhua Zhu; Lanyuan Lu; Daiwen Yang
• Keywords: Intrinsically Disordered Proteins; Feature (linguistics); Ensemble Learning

Structure characterization of intrinsically disordered proteins (IDPs) remains a key obstacle in understanding their functional mechanisms. Due to the highly dynamic feature IDPs, structure ensembles instead static unique structures are often derived from experimental data. Several state-of-the-art computational methods have been developed select an optimal ensemble pregenerated pool, but they suffer low efficiency for large IDPs. Here we present matching pursuit genetic algorithm (MPGA) determination, which takes advantages both (MP) reduce search space and (GA) restriction on constraint types. The MPGA method is validated using reference with predefined structures. In comparison conventional GA, much less time utility demonstrated by application determination mechanosensing protein domain 306 amino acids. determined reveals that N-terminal region 1–240 more compact than C-terminal 240–306. structural explains why only small portion YXXP tyrosine residues can be phosphorylated easily kinases absence extension force phosphorylation force-dependent.