ABSTRACT
As autonomous units, IDPs display considerable conformational heterogeneity under standard physiological conditions (aqueous solutions, pH 7.0-7.4, 150-mM monovalent salt, low millimolar concentrations of di-and multivalent ions, and temperatures in the 25-37°C range). is implies that distinct and disparate conformations of equivalent free energies are readily sampled via spontaneous uctuations. In order to achieve a coherent understanding of how conformational heterogeneity leads to protein function, we need accurate descriptions of the conformations accessible to an IDP under given solution conditions. Computer simulations play an important role in describing the conformational ensembles of IDPs. ey are used either de novo to o er quantitative insights regarding archetypal IDP sequences or in synergy with data from spectroscopic experiments, which serve as restraints, in order to provide interpretations of experimental data. In either mode, simulation results are reliable only if accurate descriptions of conformational ensembles are achievable. Additionally, it is necessary to be able to perform large numbers of independent simulations for a range of sequences if we are to obtain quantitative insights regarding the relationships between information encoded in amino acid sequences and the ensembles they sample under given solution conditions (Mao, Lyle, and Pappu 2013). Due to their conformational heterogeneity, IDPs create unique challenges for computer simulations: We desire an optimal combination of eciency, accuracy, and throughput, and this cannot be achieved using standard molecular dynamics approaches or methods that are designed for protein structure prediction.
