Trajectory-based training enables protein simulations with accurate folding and Boltzmann ensembles in cpu-hours

• Published in: PLoS computational biology Vol. 14; no. 12; p. e1006578
• Main Authors: Jumper, John M, Faruk, Nabil F, Freed, Karl F, Sosnick, Tobin R
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 01.12.2018; Public Library of Science (PLoS)
• Subjects: Bayes Theorem; Bayesian analysis; Biochemistry; Bioinformatics; Biology and Life Sciences; Chains; Chemistry; Computational Biology - methods; Computational efficiency; Computer applications; Computer Simulation; Computing time; Divergence; Energy; Folding; Free energy; Hydrogen; Learning algorithms; Machine Learning; Model accuracy; Molecular biology; Molecular Dynamics Simulation - statistics & numerical data; Organic chemistry; Parameter estimation; Parameterization; Physical Sciences; Physics; Protein Conformation; Protein Folding; Proteins; Proteins - chemistry; Research and Analysis Methods; Software; Solvents; Thermodynamics; Tradeoffs; Training; United States > US; Illinois; Chicago Illinois
• ISSN: 1553-7358; 1553-734X; 1553-7358
• DOI: 10.1371/journal.pcbi.1006578

An ongoing challenge in protein chemistry is to identify the underlying interaction energies that capture protein dynamics. The traditional trade-off in biomolecular simulation between accuracy and computational efficiency is predicated on the assumption that detailed force fields are typically well-parameterized, obtaining a significant fraction of possible accuracy. We re-examine this trade-off in the more realistic regime in which parameterization is a greater source of error than the level of detail in the force field. To address parameterization of coarse-grained force fields, we use the contrastive divergence technique from machine learning to train from simulations of 450 proteins. In our procedure, the computational efficiency of the model enables high accuracy through the precise tuning of the Boltzmann ensemble. This method is applied to our recently developed Upside model, where the free energy for side chains is rapidly calculated at every time-step, allowing for a smooth energy landscape without steric rattling of the side chains. After this contrastive divergence training, the model is able to de novo fold proteins up to 100 residues on a single core in days. This improved Upside model provides a starting point both for investigation of folding dynamics and as an inexpensive Bayesian prior for protein physics that can be integrated with additional experimental or bioinformatic data.