Individual human genomes frequently contain variants that have epistatic interactions
AbstractThe availability of thousands of individual genomes provides many opportunities to understand genetic variation and the relationship to phenotype, particularly disease. However, this remains challenging as it is often difficult to identify if a non-synonymous variant alters protein structure and function. Many computational methods have been developed but they typically interpret individual variants in isolation, despite the possibility of variant-variant interactions. Here, we combine the genetic variation data present in the 1000 genome project with protein structural data to identify variant-variant interactions within individual human genomes. We find more than 4,000 combinations of variants that located close in 3D dimensional structure and more than 1,200 in protein-protein interfaces. Many variant combinations include amino acid changes that are compensatory such as maintaining charges or functional groups, thus supporting that these are coevolutionary events. This highlights the need for variant interpretation and precision medicine to consider the gestalt effects of variants.