| The research project concerns the statistical aspects of genetic mapping with multivariate techniques. Within the area of behavioral genetics and genetic epidemiology the statistical approaches are mainly univariate in the sense that they deal with one phenotype per family member or twin at the time. The possibilities that multivariate twin data offer for determination of causality and for causal analysis of comorbidity and covariance are rarely utilized. The same holds for genetic linkage, association, and linkage disequilibrium (LD) studies. In theory, the presence of more than one quantitative trait locus (QTL) within a linkage region could be tested for with the use of multivariate statistical techniques. Several statistical issues need to be dealt with before this will work in practice. For the application of multivariate linkage and LD models multivariate STR marker, SNP, and phenotype data collected in Dutch twin families are available. |
