A multi-scale coevolutionary approach to predict interactions between protein domains

Abstract

Interacting proteins and protein domains coevolve on multiple scales, from their correlated presence across species, to correlations in amino-acid usage. Genomic databases provide rapidly growing data for variability in genomic protein content and in protein sequences, calling for computational predictions of unknown interactions. We first introduce the concept of direct phyletic couplings, based on global statistical models of phylogenetic profiles. They strongly increase the accuracy of predicting pairs of related protein domains beyond simpler correlation-based approaches like phylogenetic profiling (80% vs. 30-50% positives out of the 1000 highest-scoring pairs). Combined with the direct coupling analysis of inter-protein residue-residue coevolution, we provide multi-scale evidence for direct but unknown interaction between protein families. An in-depth discussion shows these to be biologically sensible and directly experimentally testable. Negative phyletic couplings highlight alternative solutions for the same functionality, including documented cases of convergent evolution. Thereby our work proves the strong potential of global statistical modeling approaches to genome-wide coevolutionary analysis, far beyond the established use for individual protein complexes and domain-domain interactions. Author summaryInteractions between proteins and their domains are at the basis of almost all biological processes. To complement labor intensive and error-prone experimental approaches to the genome-scale characterization of such interactions, we propose a computational approach based upon rapidly growing protein-sequence databases. To maintain interaction in the course of evolution, proteins and their domains are required to coevolve: evolutionary changes in the interaction partners appear correlated across several scales, from correlated presence-absence patterns of proteins across species, up to correlations in the amino-acid usage. Our approach combines these different scales within a common mathematical-statistical inference framework, which is inspired by the so-called direct coupling analysis. It is able to predict currently unknown, but biologically sensible interaction, and to identify cases of convergent evolution leading to alternative solutions for a common biological task. Thereby our work illustrates the potential of global statistical inference for the genome-scale coevolutionary analysis of interacting proteins and protein domains.