Statistical physics of interacting proteins: impact of dataset size and quality assessed in synthetic sequences

TL;DR

This study uses synthetic data to analyze how dataset size and quality affect the performance of inverse statistical physics methods, like DCA, in predicting protein-protein interactions and residue contacts.

Contribution

It formalizes the relationship between dataset size, quality, and inference accuracy using Ising models and synthetic sequences, providing insights into optimal data utilization.

Findings

DCA performs well with large datasets.

Iterative pairing can predict without training data.

Performance scales quadratically with dataset quality and size.

Abstract

Identifying protein-protein interactions is crucial for a systems-level understanding of the cell. Recently, algorithms based on inverse statistical physics, e.g. Direct Coupling Analysis (DCA), have allowed to use evolutionarily related sequences to address two conceptually related inference tasks: finding pairs of interacting proteins, and identifying pairs of residues which form contacts between interacting proteins. Here we address two underlying questions: How are the performances of both inference tasks related? How does performance depend on dataset size and the quality? To this end, we formalize both tasks using Ising models defined over stochastic block models, with individual blocks representing single proteins, and inter-block couplings protein-protein interactions; controlled synthetic sequence data are generated by Monte-Carlo simulations. We show that DCA is able to address both inference tasks accurately when sufficiently large training sets are available, and that an iterative pairing algorithm (IPA) allows to make predictions even without a training set. Noise in the training data deteriorates performance. In both tasks we find a quadratic scaling relating dataset quality and size that is consistent with noise adding in square-root fashion and signal adding linearly when increasing the dataset. This implies that it is generally good to incorporate more data even if its quality is imperfect, thereby shedding light on the empirically observed performance of DCA applied to natural protein sequences.