Structural Metrics Predict Site-Specific Evolutionary Rate in Membrane Bound Proteins

Membrane proteins are involved in many critical biological processes and mutations are linked to various diseases. We examined how the properties of the location of an amino acid residue within the protein structure dictates the rate at which it evolves. We tested 3 structural metrics: WCNSC, WCNCA and RSA for their effectiveness at predicting evolutionary rates within membrane proteins. WCNSC performed better than WCNCA in almost all cases and better than RSA in the majority. However, for some classes of proteins, especially those where the pore is a major feature, the effectiveness of WCN greatly diminished while that of RSA decreased by a lesser degree.