Nonparametric method for detecting imprinting effect using all members of general pedigrees with missing data

• Published in: Journal of human genetics Vol. 59; no. 10; pp. 541 - 548
• Main Authors: Zhang, Fangyuan, Lin, Shili
• Format: Journal Article
• Language: English
• Published: England Nature Publishing Group 01.10.2014
• Subjects: Asymmetry; Computer Simulation; Genetic Diseases, Inborn - genetics; Genomic Imprinting; Genotype; Genotypes; Humans; Missing data; Models, Genetic; Monte Carlo Method; Nonparametric statistics; Offspring; Pedigree; Power; Statistics, Nonparametric
• ISSN: 1434-5161; 1435-232X; 1435-232X
• DOI: 10.1038/jhg.2014.67

Imprinting effects can lead to parent-of-origin patterns in complex human diseases. For a diallelic marker locus, Pedigree Parental-Asymmetry Test (PPAT) and its extension MCPPAT using pedigrees allowing for missing genotypes are simple and powerful for detecting imprinting effects. However, these approaches only take affected offspring into consideration, thus not making full use of the data available. In this paper, we propose Monte Carlo Pedigree Parental-Asymmetry Test using both affected and unaffected (MCPPATu) offsprings, which allows for missing genotypes through Monte Carlo sampling. Simulation studies demonstrate that MCPPATu controls the empirical type I error rate well under the null hypotheses of no parent-of-origin effects. It is also demonstrated that the use of additional information from unaffected offspring and partially observed genotypes in the analysis can greatly improve the statistical power. Indeed, for common diseases, MCPPATu is much more powerful than MCPPAT when all genotypes are observed and the power improvement is even greater when there is missing data. For rarer diseases, there are still substantial power gains with the inclusion of unaffected offspring, although the gains are less impressive compared with those for more common diseases.