To understand the genetic architecture of cardiovascular diseases, it's commonly used the construction of the genetic map of these diseases. However, this procedure has certain difficulties, from the definition and collection of the data to its analysis. In general, for these studies are collected data from individuals and their families, in which many variables are evaluated phenotypic associated with the diseases of interest, and genotypic data, using molecular markers platforms of high dimensionality. The data used in this work are part of a larger project of InCor / USP (Heart Institute - Sao Paulo University) in which data were collected from families in the city of Baependi - MG, which include information to genotype SNPs (Affymetrics 6.0, about 1 million markers) and many of the phenotype individuals in two different years, 2006 and 2012. In this project we propose a methodology for analyzing this type of data. Initially, latent variables are constructed from a subset of SNPs using of item response theory (IRT). Then the combination of the subset of SNPs with phenotypes cardiovascular is done by adjusting the mixed model that include the latent constructed. The space of SNPs is then optimized for larger subsets of association by using genetic algorithms.
News published in Agência FAPESP Newsletter about the scholarship: