623 – Genetic Association Studies
Estimating Genotype-Specific Call Rates from Offspring-Parents Trio Data
Sponsor: Section on Statistics in Epidemiology
Keywords: genotype, missingness, nonrandom, transmission disequilibrium, bias
Zhaoxia Yu
University of California at Irvine
An important question in genetic association studies is whether missingness of genotype data depends on the underlying true genotypes. We propose a statistical method to estimate the three call rates at a single nucleotide polymorphism (SNP) using case-parents data. A likelihood ratio test is derived to test whether the call rates are all the same. The application of our methods to a genome-wide association study demonstrates that rare genotypes are usually more difficult to call and this type of nonrandom missingness introduces systematic bias that tends to estimate the common allele to be the risk allele.
