| The goal of this dissertation is to address the issue of how to meaningfully find quantitative trait loci (QTL) for correlated traits. It has been shown in the literature that a joint QTL analysis of multiple traits can have more power and be more precise than single trait QTL analysis when traits are correlated. Phenotypic correlation arises from environmental correlation, genetic correlation, or both. We wish to characterize the extent of the genetic correlation among traits.;First, we use a canonical transformation, in the form of principal component analysis, to combine many correlated traits into one, and apply single trait QTL analysis to it. We analyzed two different data sets: one from Saccharomyces cerevisiae, and another from eucalyptus. The traits analyzed in both data sets were gene expression levels generated in microarray experiments.;Subsequently, we implemented a novel multiple trait mapping method based on Multiple Interval Mapping to functionally related clusters previously studied in Saccharomyces cerevisiae. Treating RNA abundance as a phenotypic trait, we quantified the extent of the phenotypic variance due to genetic variance, and found additional QTL, previously undetected, which were functionally related to the clusters being studied.;The last part of our research contains a study of QTL for individual amino acid biosynthetic pathways of Saccharomyces cerevisiae. In the first part of this chapter, we look at the QTL topology for all individual amino acid biosynthetic pathways, finding a major transcriptional regulatory region for traits in these pathways. In the second part, we look at the QTL topology of some individual amino acid biosynthetic pathways in detail, paying close attention to the pleiotropic QTL in each of them. |
