New Ways in Quantitative Genetics: Results of QTL Analyses Revisited by Cross Validation and Bootstrapping* A.E. MELCHINGER 1 , S. CHANDRA 2 , and H.F. UTZ 1 1 Institute of Plant Breeding, Seed Science and Population Genetics, University of Hohenheim, 70593 Stuttgart, Germany 2 International Crops Research Institute for the Semi-Arid Tropics, Patancheru 502 324, India SUMMARY After a brief review on the progress in biometric methods of QTL analyses for populations from biparental crosses, we summarize trends obtained from experimental studies with crops and simulations. Furthermore, we describe the use of cross validation and bootstrapping in QTL mapping and employed these methods for re-analyses of published QTL experiments. In agreement with simulations, validation with an independent sample or cross validation confirmed the low precision and substantial bias of estimated QTL effects and p, the proportion of the genotypic variance explained by the detected QTL, obtained with the commonly used sample sizes. While bootstrapping might be a further alternative, at present we recommend cross validation with composite interval mapping based on the regression approach as a simple and computationally manageable procedure for obtaining unbiased estimates of QTL effects and a realistic assessment of the prospects of marker- assisted selection. INTRODUCTION About 20 years ago, Lewontin (1977) questioned the progress in Quantitative Genetics. His major criticism was that it could not provide answers to the following questions (cited in ___________________________________________________________________ * Presented at EUCARPIA Biometrics Conference, Paris, August 2000