Likelihood-Based Methods to Detect Maternal-Fetal Genotype Effects in Families

• Activity Number: 16
• Type: Topic Contributed
• Date/Time: Sunday, August 1, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Janet Suzanne Sinsheimer*+ and Erica Childs and Christina Palmer and Kenneth Lange
• Companies: University of California, Los Angeles and University of California, Los Angeles and University of California, Los Angeles and University of California, Los Angeles
• Keywords: genetics ; likelihood ; gene by gene interactions ; missing data

Abstract

Because studies can collect large, complex families, we extend the maternal-fetal genotype (MFG) incompatibility test, a model based likelihood approach of jointly detecting maternal and offspring genotype effects, to allow for any family structure. We replace our nuclear-family mating type approach with a pedigree likelihood and calculate MFG incompatibility along with the Mendelian transmission probabilities. In order to allow arbitrary family structures, we make a slightly more stringent assumption of random mating with respect to the locus of interest. We find that the EMFG test provides improved accuracy of parameter estimation and dramatic gains in power in the presence of missing genotypes over the original nuclear family test.