Conceptual thinking for in silico prioritization of candidate disease genes.

Prioritization of most likely etiological genes entails predicting and defining a set of characteristics that are most likely to fit the underlying disease gene and scoring candidates according to their fit to this "perfect disease gene" profile. This requires a full understanding of the disease phenotype, characteristics, and any available data on the underlying genetics of the disease. Public databases provide enormous and ever-growing amounts of information that can be relevant to the prioritization of etiological genes. Computational approaches allow this information to be retrieved in an automated and exhaustive way and can therefore facilitate the comprehensive mining of this information, including its combination with sets of empirically generated data, in the process of identifying most likely candidate disease genes.