Oligogenic approaches to the predisposition of alcohol dependence: A genome-wide search on 255 families

Using data from a two stage family study (totally 255 families, 1102 genotyped subjects) we searched for an ethnicity-independent oligogenic configuration of interacting loci that explained the observed susceptibility to alcohol dependence. Our analyses were carried out under the constraint of reproducibility as during the adaptive search one set of families served as "training" samples while the second set of families were used as independent "test" samples. Based on a genetic similarity function that quantifies the genetic distances d(xi,xj) between feature vectors xi, xj made up of the allelic patterns of any two subjects i, j at n loci l ₁, l₂, .. l_n, our signal detection algorithm looked for significant differences between the parent-offspring similarities and the between-sib similarities of affected sib pairs under the constraint that no such differences showed up for the unaffected sib pairs. The question of unknown population admixture was addressed by means of a quantitative concept of ethnicity in combination with cluster-analytical methods. We found a reproducible configuration of 14 ethnicity-independent loci on chromosomes 1, 2, 4, 9, 12, 13, 14, 16, 18 and 20 as being vulnerability-related, and 5 ethnicity-independent loci on chromosomes 1, 5, 14 and 19 as being protective with respect to the clinical syndrome of alcohol dependence. None of these loci appeared to be, by itself, necessary or sufficient for the observed susceptibility to alcohol dependence.