Article Abstract:
Three generalized constraint methods have been proposed. They were evaluated using simulated data. Results compare the power of methods with and without covariates for a single-gene model with age-dependent onset and for quantitative and qualitative gene-gene and gene-environment interaction models. Covariates can improve ability to detect linkage and can be very valuable when there are qualitative gene-environment interactions. The best approach usually is to assume no dominance variance exists and to get constrained estimates for covariate models under the assumption. Covariate models have been developed in the past as an extension to affected sib-pair methods in which covariate effects are jointly estimated with degree of excess allele sharing. The models can estimate differences in sib-pair allele sharing associated with measurable genes or environment. When an affected sib-pair model is generalized to allow for covariates that have an effect on allele sharing, new constraints and methods are needed.
User Contributions:
Comment about this article or add new information about this topic:
Article Abstract:
A simulation study has been made of effects of assignment of prior identity-by-descent probabilities to sib pairs in covariance-structure modeling of a quantitative-trait locus. The sib pairs are unselected. Sib-pair selection strategies intended for identification of the most informative sib pairs for the purpose of detecting a quantitative-trait locus (QTL) lead to a missing-data problem in genetic covariance-structure modeling of QTL effects. A proposed solution to the problem is assigning prior identity-by-descent (IBD probabilities to the unselected sib pairs. This has been investigated with two maximum-likelihood )approaches to estimation considered. Assignment of prior IBD probabilities brings serious estimation bias in the pi-hat approach. The null distribution of the log-likelihood ratio does not follow the expected null distribution in the pi-hat approach after selection. In the IBD mixture approach, the null distribution is not in agreement with expected results.
User Contributions:
Comment about this article or add new information about this topic:
Article Abstract:
Combined association sib-pair and linkage analysis has been carried out for quantitative traits. Extending maximum-likelihood variance-components procedures currently used for mapping quantitative-trait loci in sib pairs in order to allow for a simultaneous test of allelic association has been suggested. Modeling of the allelic means for a test of association is involved, as is modeling at the same time of the sib-pair covariance structure to test linkage. By partitioning the mean effect of a locus into within- and between-sibship components it controls for invalid association made on the basis of population admixture and stratification. Simulations of models of real and unfounded association are used to show the value of the method.
User Contributions:
Comment about this article or add new information about this topic: