Lost in Projection? Gaussian Filtering Recovers Hidden Conformational States

TL;DR

This paper introduces a Gaussian filtering method to correct projection artifacts in molecular dynamics data, enabling better identification of conformational states and revealing hidden metastable states.

Contribution

The study demonstrates that Gaussian low-pass filtering of high-dimensional MD data can recover hidden conformational states and improve state definition in free energy landscapes.

Findings

Gaussian filtering restores hidden states in MD simulations.

Number of microstates increases significantly after filtering.

Filtered states are more structurally stable and long-lived.

Abstract

To interpret molecular dynamics (MD) simulations, it is common practice to reduce the dimensionality of the molecular coordinates to a low-dimensional collective variable $x$. Projecting the high-dimensional MD data onto $x$ yields a free energy landscape $\Delta G(x)$, which highlights low-energy regions corresponding to conformational states. The accurate definition of these states, however, is often impeded by projection artifacts, resulting in artificially shortened state lifetimes or even the complete disappearance of states from the analysis. As demonstrated for a two-dimensional toy model, Gaussian low-pass filtering of the high-dimensional MD coordinates can restore the underlying free energy landscape, allowing to recover previously hidden states. When applied to an all-atom folding trajectory of HP35, the number of microstates increases by an order of magnitude, which leads to metastable states that are long-lived and much better defined structurally, even compared to dynamically cored state trajectories.