Structural decomposition of genetic diversity in families with alcohol dependence

The Collaborative Study on the Genetics of Alcoholism (COGA) was launched in 1989 in the United States and is a six-center program to detect and map susceptibility genes for alcohol dependence and related phenotypes. The first-wave COGA sample comprises data on 987 individuals from 105 families (diagnosed probands and relatives) and includes the genotypes of 291 marker loci at an average intermarker distance of 13.8 cM. Based on a genetic similarity function that enables one to quantify the genetic distances d(xi,xj) between high-dimensional feature vectors xi, xj made up by the allelic patterns of individuals i, j with respect to loci 11, 12, .. In, we have investigated the sib-sib similarities which are expected to exceed the value of "0.5" in affected sib pairs if the region of interest (ROI) contains markers close to diseasecausing genes. The reference value "0.5" was derived by evaluating the parent-offspring similarities. Additionally, we constructed the eigenvectors that optimally represent the genetic variation ("diversity") associated with the high-dimensional feature vectors. It turned out that (1) typically 3-4 eigenvectors explained two thirds of the genetic variation inherent to the 10-15 polymorphic markers of each chromosome, and (2) several marker configurations on chromosomes 1, 3, 7, 15, and 17 reproducibly discriminated (P ≤ 0.01) index cases and unaffected sibs on the one hand, and affected and unaffected sibs on the other, while no such differences were found between index cases and affected sibs.