Determining the Underlying Distributions of Change in Free Energy Change for Pathogenic and Benign Protein Mutations

Abstract

A mutation in a patient's genome can affect a protein in that patient’s body, resulting in either no change in the health of the patient or a disease experienced by the patient. Assigning terminology, the mutations can therefore be referred to as benign or pathogenic, respectively. When these benign or pathogenic mutations occur, there is an associated change in change in free energy (ΔΔG) when the protein folds, which essentially means the act of the protein folding can become more or less stabilizing. The questions we were interested in are the following: are pathogenic protein mutations stabilizing or destabilizing when compared to benign protein mutations and is there a difference between ΔΔG distributions for benign and pathogenic mutations. In order to analyze the distribution of the ΔΔG’s, we looked at both data from a previous study and data obtained from an extensive literature search for pathogenic mutations found in patients who exhibit a disease. We found that there appears to be a statistical difference between the distribution of benign ΔΔG’s and pathogenic ΔΔG’s when organizing proteins by general function and that pathogenic mutations appear to be more destabilizing than benign mutations. Furthermore, pathogenic distributions appear better described by two gaussians, or a bimodal distribution, whereas benign distributions are adequately described by a single gaussian. Pathogenic distributions also appear to have greater range and variance. While the causes are not yet entirely understood, these results can play a role in understanding what, if any, role ΔΔG has on the pathogenicity of a mutation and could be one day used alongside other methods to generate a model that can help predict the pathogenicity of an arbitrary mutation.