Reconstructing the free-energy landscape of Met-enkephalin using dihedral Principal Component Analysis and Well-tempered Metadynamics

TL;DR

This paper introduces a method combining Well-Tempered Metadynamics with dihedral Principal Component Analysis to efficiently reconstruct the free-energy landscape of Met-enkephalin, overcoming the challenge of selecting relevant collective variables.

Contribution

The study presents a novel approach that couples WTmetaD with dihedral PCA-derived CVs, providing a robust and generic way to improve free-energy surface reconstruction for proteins.

Findings

dPCA-derived CVs effectively capture backbone conformations

The method produces a rugged free-energy landscape for Met-enkephalin

A criterion is proposed to select the optimal number of CVs

Abstract

Well-Tempered Metadynamics (WTmetaD) is an efficient method to enhance the reconstruction of the free-energy surface of proteins. WTmetaD guarantees a faster convergence in the long time limit in comparison with the standard metadynamics. It still suffers however from the same limitation, i.e. the non trivial choice of pertinent collective variables (CVs). To circumvent this problem, we couple WTmetaD with a set of CVs generated from a dihedral Principal Component Analysis (dPCA) on the Ramachadran dihedral angles describing the backbone structure of the protein. The dPCA provides a generic method to extract relevant CVs built from internal coordinates. We illustrate the robustness of this method in the case of the small and very diffusive Metenkephalin pentapeptide, and highlight a criterion to limit the number of CVs necessary to biased the metadynamics simulation. The free-energy landscape (FEL) of Met-enkephalin built on CVs generated from dPCA is found rugged compared with the FEL built on CVs extracted from PCA of the Cartesian coordinates of the atoms.