PARTITIONED ASSOCIATION-LINKAGE TEST - DISTINGUISHING NECESSARY FROM SUSCEPTIBILITY LOCI

Marker allele-disease association and linkage between a disease locus and a marker locus are two different phenomena. Linkage without eviden ce of association and association without evidence of linkage are poss ible observations. Linkage analysis uses marker loci and the phenomeno n of recombination to look for disease-related loci which are presumab ly major contributors to disease expression (''necessary'' loci). Howe ver, the phenomenon of association is more complex. One explanation fo r the existence of an association is that there is a ''necessary'' loc us in linkage disequilibrium with a marker locus, Another explanation is that the marker locus itself (or a closely linked locus in linkage disequilibrium with the marker) is a ''susceptibility'' locus, which i ncreases the probability of contracting the disease but is not necessa ry for disease expression, Although there are other possible explanati ons for the existence of an association, these two can lead to differe nt results when family data from a disease showing association are ana lyzed for linkage between the associated marker and the disease. If th e linkage disequilibrium hypothesis is correct, there will be evidence for linkage. If the susceptibility locus hypothesis is correct, there may be strong evidence against linkage. In this work, we explore a me thod that could indicate whether an association is due to a susceptibi lity locus or a necessary locus. We show that, by dividing families ba sed on the presence or absence of the associated marker allele in a ra ndomly chosen affected sib, calculating lod scores, and then calculati ng a heterogeneity statistic, we could distinguish whether linkage dat a came from a susceptibility locus or a necessary locus. (C) 1996 Wile y-Liss, Inc.