Mapping Gene Associations in Human Mitochondria using Clinical Disease Phenotypes
For each gene group, we calculated the fraction of genes (y-axis) that interact with k other genes (x-axis). The gene group fractions were calculated for interactions to all human genes (A), all disease genes (B), and all mitochondrial genes (C). The color codes of the gene groups with their respective regression line slopes and p-values (in parenthesis) are shown in the table (D), with all results including correlation coefficients listed in Table S7. The fraction analysis of all gene interactions (A) showed a higher network connectivity for non-disease genes (CG) than for disease genes (DG), as indicated by the larger absolute value of the negative DG regression slope. In contrast, the connectivity to disease genes is relatively higher for DG than for CG suggesting a tendency for DG to interact with each other (B). Human orthologs to mouse essential genes had the highest network connectivity, while the mitochondrial gene groups had the highest tendency to interact to other mitochondrial genes (C).