Computational Elucidation of Protein-Protein Interactions in the Minimal Proteome of JCVI-syn3A

Abstract

Proteins play a vital role in the regulation of biological processes, facilitating the transfer of intermediates and coordinating the sequential steps of biochemical pathways. Protein-protein interactions (PPIs) are crucial molecular events in which two or more proteins bind together, enabling the formation of protein complexes that govern various cellular activities, including signal transduction, gene expression, and enzymatic reactions. Evolutionary correlations arise due to the close proximity of amino acid side chains within these interactions, where amino acids on one side of an interaction surface may restrict which amino acids fit on the other side or encourage mutations that modify the surface. In this study, our aim is to investigate the correlation between protein sequence and structure in mycoplasma mycoides JCVI-syn3A, a minimal cell consisting of 493 genes. We utilize the EVCouplings framework, a coevolution-based approach with probabilistic scores for residue interactions, to predict protein-protein interactions and the specific surfaces that govern them. Our study demonstrates that coevolution-based computational methods can predict protein-protein interactions and their interaction surfaces. After analyzing multiple sequence alignment (MSA) data across 110 protein families, we identify a total of 33 inter-protein interactions. Our analysis of the protein-protein interactions in JCVI-syn3A provides valuable insights into the genetic architecture of Mycoplasma, one of the simplest cellular life forms known, and enhances our understanding of how the earliest cellular life forms might have functioned.