Approaching Inherited Disease on a Genomic Scale

We review current approaches that can extend our understanding of monogenic disease towards complex disease. Recent studies showed that currently established disease genes differ in their protein size, tissue specificity and the phylogenetic distribution of homologs. These characteristics can be explained by the fact that monogenic disease mutations must be sufficiently deleterious to produce a clearly recognizable phenotype, but also must not be lethal in an early embryonic stage. On the other hand, deletion of each gene in the human genome must be evolutionarily disadvantageous. For most genes, this disadvantage might manifest as an increased susceptibility to complex disease. Accordingly, mildly deleterious variants can be observed in a wide spectrum of genes. The phenotypic manifestation of these mildly deleterious variants might depend on somatic mutations, which cause the breakdown of compensating mechanisms in individual cells. At present, association studies are the most promising strategy for mapping complex disease phenotypes. However, these are restricted to the identification of common disease variants and often provide only marginally convincing statistical evidence. Novel computational strategies, which take prior biological knowledge into account, therefore might play a major role in the design and interpretation of large-scale association studies.