Abstract
Current genetic investigations are performed both on the basis of a rational and biologically based choice of candidate genes and through genome wide scans. Nonetheless, lack of replication is a common problem in psychiatric genetics as well as in other genetic fields. There are a number of reasons for this inconsistency, among them a well known but poorly considered issue is gene coverage. The aim of the present paper is to focus on this well known and defectively deemed bias, especially when a candidate gene approach is chosen. The rational and the technical feasibility of this proposal are discussed as well as a survey of current investigations. The known consistent methodology to fix this bias is also discussed.