Genetic Mapping Using the Theory of the Added Variable Plot in the Mixed Models

Abstract

Recently, one of the most important problems in genetics is the identification of genes associated with complex diseases. A useful design for this proposal corresponds to collect data from extended families and molecular markers platforms SNPs (Single Nucleotide polymorphism). These platforms represent points of reference strategically placed along the genome of the individual sand are high dimensional.Analysis of these data brings analytical challenges as the problem of multiple testing and selection of predictive variables. In this thesis, we propose a criterion for discriminating predictors of genetic effects due to random polygenic component and the residual component, under the framework of a linear mixed model. Also, considering that the individual effects of predictor variables is expected to be small,it issuggested a method for finding ordered subsets of these variables and study their simultaneous effect on the response variable under study. In this context, is used the theory of the added variable plot under a mixed model framework. The proposals are validated through a simulation study, which is based on structures of families involved in the Project “Baependi Heart Study” (FAPESP Process 2007/58150-7), whose objective is to identify genes associated with cardiovascular risk factors in the Brazilian population. This proposal is implemented by using the R statistical environment and for the simulation of genetic predictors is adopted the SimPed application.