Analysis of the relationship between functional divergence and the propensity of duplicated genes to be involved in human Mendelian diseases

Abstract

In recent studies, it has been observed that genes that have been duplicated during the course of vertebrate evolution are overrepresented among those genes that cause Mendelian diseases. My objective was to determine whether measures of functional divergence are correlated with the propensity of duplicated genes to be involved in Mendelian disease. To test this, I used a phylogeny-based maximum-likelihood mixture-model prediction program, FunDi, that accounts for functional divergence in phylogenetic trees. I then conducted a statistical analysis of the data, measuring the Rho value of functional divergence weight and branch lengths values, using a Pearson correlation test and two-sided Wilcoxon-Mann-Whitney U-test. Statistically significant correlation was found for the relationship between the length of the branch in the phylogenetic tree separating disease-associated genes and its orthologs from the rest of the gene family and the propensity for a gene to be involved in autosomal recessive disorders. Optimization with FunDi, which accounts for functional divergence in its model, resulted in shorter branch lengths. Unfortunately, no statistical significance was found between the analyzed gene categories for the Rho value. Therefore, I conclude that while some measures of evolution and functional divergence, such as the internal branch length between groups, may be correlated with disease-association, direct measures of functional divergence measured in this study do not explain the propensity of duplicated genes to be involved in Mendelian diseases.