Out-of-equilibrium selection pressure enhances inference from protein sequence data

TL;DR

This paper demonstrates that out-of-equilibrium selection pressures and fluctuating selection strengths can improve the accuracy of coevolution-based inference methods for protein structure and interactions.

Contribution

It introduces a minimal model showing how non-equilibrium dynamics enhance inference from protein sequence data, extending to realistic synthetic data and interaction partner inference.

Findings

Fluctuating selection improves structural contact inference.

Out-of-equilibrium noise enhances inference accuracy.

Results extend to synthetic data and interaction prediction.

Abstract

Homologous proteins have similar three-dimensional structures and biological functions that shape their sequences. The resulting coevolution-driven correlations underlie methods from Potts models to AlphaFold, which infer protein structure and function from sequences. Using a minimal model, we show that fluctuating selection strength and the onset of new selection pressures improve coevolution-based inference of structural contacts. Our conclusions extend to realistic synthetic data and to the inference of interaction partners. Out-of-equilibrium noise arising from ubiquitous variations in natural selection thus enhances, rather than hinders, the success of inference from protein sequences.