CONSTRUCT: an algorithmic tool for identifying functional or structurally important regions in protein tertiary structure

Abstract

In the realm of protein-coding genes, evolutionary rates show considerable variability. Essential or highly expressed proteins evolve more slowly, and within a protein, different amino acid sites evolve at different rates. Accurately modeling this variation is critical for identifying structurally or functionally important amino acid sites. Standard methods such as Rate4Site assume independent substitution rates across sites, and the most conserved ones are widely distributed in protein tertiary structure. This is biologically unrealistic because functional sites often cluster together. Here, we developed CONSTRUCT, an improved strategy for identifying functional and structurally important regions in protein tertiary structure. Given a set of orthologous sequences, CONSTRUCT first calculates site-specific substitution rates using the Rate4site model, which are then weighted by the rates of neighboring amino acid sites within a range of window sizes. The optimal window size is determined by the strongest spatial correlation, if present. This method can be analyzed using either C atoms or the center of mass of amino acid sites to account for side chain orientation. Validation on 14 functionally characterized proteins of different size, conservation level and kingdom/domain of species has demonstrated the relevance of CONSTRUCT.