Predicting the Sequence-Dependent Backbone Dynamics of Intrinsically Disordered Proteins.

How the sequences of intrinsically disordered proteins (IDPs) code for functions is still an enigma. Dynamics, in particular residue-specific dynamics, holds crucial clues. Enormous efforts have been spent to characterize residue-specific dynamics of IDPs, mainly through NMR spin relaxation experiments. Here we present a sequence-based method, SeqDYN, for predicting residue-specific backbone dynamics of IDPs. SeqDYN employs a mathematical model with 21 parameters and is trained on 45 IDPs. It provides not only rapid, accurate prediction but also insightful physical interpretation of sequence-dependent IDP dynamics.