ABSTRACT
In many QTL mapping studies, observations on several traits are recorded together with the genotypes of markers. Very often, such traits are correlated and there are common chromosome regions that affect multiple traits. Although the statistical methods described in previous chapters can be applied to each trait one-by-one, such approach does not take into account the correlation among multiple traits and loses power in detecting QTL with pleiotropic effects. A QTL has pleiotropic effect if it affects several traits simultaneously. Statistical methods especially for mapping multiple traits are demanded. Many methods for multi-trait QTL mapping have been developed in the literature, ranging from simple extensions of single-trait approaches to sophisticated multi-trait approaches designed specifically for multi-trait QTL mapping. In this chapter, we discuss some of these methods. In § 8.1, we deal with methods based on single-QTL-models, including one-trait-at-a-time approaches, meta-trait methods and multivariate composite interval mapping. In § 8.2, we consider more sophisticated approaches which are based on multiQTL models such as multivariate sparse partial least square regression, multivariate sequential procedures, and multivariate Bayesian approaches.
