A combinatorial partitioning method to identify multi-genic multi-locus models that predict quantitative trait variability.

摘要

Recent advances in genome research have accelerated the process of locating candidate genes and the variable sites within them, and have simplified the task of genotype measurement. The development of statistical and computational strategies to utilize information on hundreds, soon thousands, of variable loci to investigate the relationships between genome variation and phenotypic variation has not kept pace, particularly for quantitative traits that do not follow simple Mendelian patterns of inheritance. We present here a combinatorial partitioning method (CPM) that simultaneously considers many genes, each containing multiple variable loci, to identify partitions of multi-locus genotypes that predict interindividual variation in quantitative trait levels. A hallmark of this method is its reliance on observed data to identify subsets of multi-locus genotypes that have similar phenotypic distributions, accomplished without conditioning the analysis on a pre-specified genetic model. We illustrate this method with an application to plasma triglycerides, HDL cholesterol, Total cholesterol, and ApoE levels collected on 188 males and 241 females ages 20 to 60 years, ascertained without regard to health status from Rochester, MN. Genotype information for 18 diallelic loci in six coronary heart disease candidate gene regions APOA1-C3-A4, APOB, APOE, LDLR, LPL, and PON1 was considered. We found that many combinations of loci are involved in sets of genotypic partitions that predict a significant amount of trait variability, and that the most predictive sets of genotypic partitions show strong non-additivity between loci. These results suggest that traditional methods of building multi-locus models that rely on statistically significant marginal, single locus effects, are unlikely to identify combinations of loci that best predict trait variability. The CPM offers a strategy for exploring the high dimensional genotype state space for explanations of quantitative trait variation in the population at large that does not require an a priori specification of a genetic model.