Multi-allelic TDT simulation study of indigenous fijian families with psychosis

For identifying disease genes of modest effect, instead of "classical" linkage analysis, methods that test jointly for linkage and disequilibrium may be more promising, particu-larly in isolated genetic populations. We have collected 140 psychotic indigenous Fijian probands with available parents, or first degree relatives to reconstruct parental genotypes when one or both parents are unavailable. To assess the feasibility of detecting linkage in the presence of association, we perform power studies by simulating marker data within the collected pedigree structures using the SLINK program. We have assumed linkage (with 10 equally frequent alleles) in order to test the power of the sample under various modes of inheritance. In the first phase of genotyping this sample, we anticipate investigating 10 candidate regions, and thus we correct for multiple testing (Bonferroni correction). Preliminary data demonstrate this dataset has reasonable power (75% - assuming a p-value threshold of 0.005) to detect the presence of strong disequilibrium when data is simulated using a dominant genetic model (penetrances: 0.001, 0.1, 0.1; disease allele frequency 0.05) and analyzed with a multi-allelic TDT test.