WHOLE GENOME SCAN FOR QUANTITATIVE TRAIT LOCI AFFECTING GROWTH AND FEED EFFICIENCY TRAITS IN CHICKENS USING A THREE GENERATION DESIGN J. B. C. H. M. van Kaam1, J. A. M. van Arendonk1, M. A. M. Groenen1, H. Bovenhuis1, A. L. J. Vereijken2 'Wageningen Institute of Animal Sciences, Wageningen Agricultural University, P.O. Box 338, 6700 AH Wageningen, The Netherlands 2Euribrid B.V., P.O. Box 30, 5830 AA Boxmeer, The Netherlands SUMMARY Growth and feed efficiency traits in broilers were analysed to detect QTLs. Data was from 10 families of a cross between two broiler lines in a three generation design. The first two generations consisted of full sib animals, which were genotyped. The third generation consisted of half sibs, which were used to obtain observations. These observations were used to calculate trait values for animals in the second generation. Difference in variance of observations on both sexes was taken into account by standardising. A whole genome scan was performed using a linkage map consisting of 368 microsatellite markers on 24 autosomal linkage groups. QTL detection was done with an across family weighted regression interval mapping approach. The most likely position for a QTL affecting feed consumption and body weight at 48 days was found on chromosome I at 234-240 cM. A permutation test indicated a genomewise significance level of this QTL for feed consumption of 4%. Keywords: Chicken, Quantitative Trait Loci. Body weight, Three generation design, Regression INTRODUCTION The availability of a considerable number of DNA markers and a genetic linkage map (Groenen et al. 1997) for chickens facilitates genome wide scans for genes affecting quantitative traits (QTLs). Van der Beek et al. (1995) suggested the use of a three generation full sib-half sib design. In this design, genotypes are determined for first and second generation full sib animals while phenotypic information is collected on third generation half sib grandoffspring. The aim of this study was to detect and localise QTLs in a three generation design. For this purpose a regression approach was developed that takes the frill sib structure of tire design into account. This regression approach was based on tire method of Knott et al. (1994) for half sib designs. The method was applied to body weight at 23 and 48 days, feed consumption and feed efficiency. 310