Collective Variable for Metadynamics Derived from AlphaFold Output

TL;DR

This paper introduces a novel collective variable derived from AlphaFold outputs to enhance protein folding simulations using metadynamics, demonstrating its potential in structure refinement and mutation outcome prediction.

Contribution

It presents a new AlphaFold-based collective variable for metadynamics, enabling improved protein folding simulations and broader applications.

Findings

Successful folding simulation of Trp-cage mini-protein

AlphaFold-based CV predicts folding equilibria

Potential for mutation outcome prediction

Abstract

AlphaFold is a neural-network-based tool for the prediction of 3D structures of protein. In CASP14, a blind structure prediction challenge, it performed significantly better than other competitors, which makes it the best available structure prediction tool. One of the outputs of AlphaFold is the probability profile of residue-residue distances. This makes it possible to score any conformation of the studied protein to express its compliance with the AlphaFold model. Here we show how this score can be used to drive protein folding simulation by metadynamics and parallel tempering metadynamics. By parallel tempering metadynamics, we simulated folding of a mini-protein Trp-cage beta hairpin and predicted their folding equilibria. We see the potential of AlphaFold-based collective variable in applications beyond structure prediction, such as in structure refinement or prediction of the outcome of a mutation.