POINTING THE WAY TOWARD QTL IN DAIRY CATTLE USING COMPASS

Department of Animal Sciences, 206 Edward R. Madigan Laboratory, 1201 W. Gregory Drive, Urbana, IL 61801 USA

One of the most challenging problems in modern genomics is the identification of candidate genes for QTL. In domestic animal species, QTL are mapped at such low resolution that literally thousands of genes may lie within a marker interval. Most investigators rely on the use of comparative mapping information to select candidate genes (comparative positional candidate approach), but this is complicated by the poor resolution of existing comparative maps and the paucity of mapped genes with known function. We have conducted a moderate resolution scan of the dairy cattle genome for QTL affecting milk yield, composition and health traits. Eight large half-sib families (n = 1076) in the Dairy Bull DNA Repository (DBDR) were used as a resource population. Data were analyzed using a granddaughter design. Genome coverage was estimated to be 2551 cM for 174 markers genotyped. QTL affecting milk yield (most strongly affecting fat percentage) were identified on BTA3 and BTA14. The process of identifying candidate genes for these QTL is proceeding using a strategy termed comparative mapping by annotation and sequence similarity (COMPASS, Ma et al., Mammal. Genome, 9:545). COMPASS uses bioinformatics to predict the map location of cattle ESTs and known genes on the basis of comparative mapping information. Map location is then confirmed and ESTs ordered with respect to QTL markers using radiation hybrid (RH) mapping. We have generated nearly 1000 ESTs and are successfully using COMPASS to predict their map location. Approximately 100 ESTs with known human orthologs have already been genotyped on the RH panel, and several fall into intervals containing QTL. When fully scaled and coupled with microarray technologies this process will permit identification of a refined set of candidate genes for QTL detected in the DBDR.