LAUSR

Numerous statistical methods have been developed to explore genomic imprinting and maternal effects, which are causes of parent-of-origin patterns in complex human diseases. However, most of them either only model one of these two confounded epigenetic effects, make strong yet unrealistic assumptions about the population to avoid over-parameterization, or are only applicable to study designs that require recruitment of difficult-to-obtain control families. In this paper, we develop a partial Likelihood method for detecting Imprinting and Maternal Effects for a Discordant SibPair design LIMEDSP utilizing all available sibship data without the need to recruit separate control families. By matching affected and unaffected probands and stratifying according to their familial genotypes, a partial likelihood component free of nuisance parameters can be extracted from the full likelihood. This alleviates the need to make assumptions about the population. Theoretical analysis shows that the partial maximum likelihood estimators based on LIMEDSP are consistent and asymptotically normally distributed. Based on the close-form formula for computing information, we compared a study design with more independent families versus one with larger families by keeping the total number of individuals needed to be genotyped fixed. We further carried out a simulation study, which demonstrates the robust property of LIMEDSP and shows that it is a powerful approach without resorting to collecting control families. To illustrate its practical utility, LIMEDSP was applied to a clubfoot dataset and the Framingham Heart Study. Statistica Sinica: Preprint doi:10.5705/ss.202016.0114