2013/08/20 by Andrew J. Bordner, Bordner, Andrew J., Hans D. Mittelmann +1
Biochemistry, Genetics and Molecular Biology · #Biomolecules (q-bio.BM) #FOS: Biological sciences #Microbial Metabolic Engineering and Bioproduction #Populations and Evolution (q-bio.PE) #Protein Structure and Dynamics #RNA and protein synthesis mechanisms
paper · pdf · doi:10.48550/arxiv.1308.4342
openalex publication_date 2013/08/20 · openalex created_date 2022/08/14 · openalex updated_date 2026/07/28
Despite the importance of a thermodynamically stable structure with a\nconserved fold for protein function, almost all evolutionary models neglect\nsite-site correlations that arise from physical interactions between\nneighboring amino acid sites. This is mainly due to the difficulty in\nformulating a computationally tractable model since rate matrices can no longer\nbe used. Here we introduce a general framework, based on factor graphs, for\nconstructing probabilistic models of protein evolution with site\ninterdependence. Conveniently, efficient approximate inference algorithms, like\nBelief Propagation, can be used to calculate likelihoods for these models. We\nfit an amino acid substitution model of this type that accounts for both\nsolvent accessibility and site-site correlations. Comparisons of the new model\nwith rate matrix models and a model accounting only for solvent accessibility\ndemonstrate that it better fits the sequence data. We also examine evolution\nwithin a family of homohexameric enzymes and find that site-site correlations\nbetween most contacting subunits contribute to a higher likelihood. In\naddition, we show that the new substitution model has a similar mathematical\nform to the one introduced in (Rodrigue et al. 2005), although with different\nparameter interpretations and values. We also perform a statistical analysis of\nthe effects of amino acids at neighboring sites on substitution probabilities\nand find a significant perturbation of most probabilities, further supporting\nthe significant role of site-site interactions in protein evolution and\nmotivating the development of new evolutionary models like the one described\nhere. Finally, we discuss possible extensions and applications of the new\nsubstitution models.\n