Mapping of multiple quantitative trait loci by simple regression in half-sib designs

Detection of QTL in outbred half-sib family structures has mainly been based on interval mapping of single QTL on individual chromosomes. Methods to account for linked and unlinked QTL have been developed, but most of them are only applicable in designs with inbred species or pose great demands on computing facilities. This study describes a strategy that allows for rapid analysis, involving multiple QTL, of complete genomes. The methods combine information from individual analyses after which trait scores for a specific linkage group are adjusted for identified QTL at other linkage groups. Regression methods are used to estimate QTL positions and effects; permutation tests are used to obtain empirical threshold values. The description of the methods is complemented by an example of the combined analysis of 28 bovine chromosomes and their associations with milk yield in Finnish Ayrshire cattle. In this example, the individual analysis revealed five suggestive QTL affecting milk yield. Following the strategy presented in this paper, the final combined analysis showed eight significant QTL affecting milk yield. This clearly demonstrates the potential gain of using the combined analysis. The use of regression methods, with low demands on computing resources, makes this approach very practical for total genome scans.