PMC:4564992 / 3831-5430 JSONTXT

{"project":"2_test","denotations":[{"id":"26320892-16339052-2052496","span":{"begin":273,"end":275},"obj":"16339052"},{"id":"26320892-17549757-2052496","span":{"begin":273,"end":275},"obj":"17549757"},{"id":"26320892-19809476-2052496","span":{"begin":273,"end":275},"obj":"19809476"},{"id":"26320892-25078964-2052496","span":{"begin":273,"end":275},"obj":"25078964"},{"id":"26320892-24903222-2052497","span":{"begin":362,"end":364},"obj":"24903222"},{"id":"26320892-17701901-2052498","span":{"begin":429,"end":431},"obj":"17701901"},{"id":"26320892-8447318-2052499","span":{"begin":502,"end":504},"obj":"8447318"},{"id":"26320892-10364536-2052500","span":{"begin":1000,"end":1002},"obj":"10364536"},{"id":"26320892-20016592-2052501","span":{"begin":1127,"end":1128},"obj":"20016592"}],"text":"A variety of statistical methods have been used for the detection and estimation of parent-of-origin effects in humans. We focus here on methods designed for binary (disease) traits, rather than on methods that have been developed for the analysis of quantitative traits.9–12 A review of the most popular currently used approaches is given by Connolly and Heron.13 One intuitive approach, available in the software package PLINK,14 is to use an adaptation of the transmission disequilibrium test (TDT),15 whereby transmissions and non-transmissions of an allele of interest to an affected offspring are stratified according to parental origin. However, such TDT-like approaches generally have the disadvantage of discarding observations in which both parents and offspring are heterozygous (given that parental origin cannot be assigned in this case), of erroneously assuming the transmissions from two heterozygous parents are independent (which is not true in the presence of child-genotype effects16), and of being sensitive to (i.e., invalid in the presence of) maternal-genotype effects. In an Icelandic study, Kong et al.5 used a case-control version of this TDT-like approach and overcame these limitations by performing long-range phasing of SNP data to infer haplotypes and comparing the resulting haplotypes with those present in the closest relatives on the paternal and maternal sides (considered as “surrogate parents”) in order to infer parent of origin; they also excluded the possibility of maternal-genotype effects by evaluating the effects of the non-transmitted maternal alleles."}