The common genetic variants identified through genome-wide association studies explain only a small proportion of the genetic risk for complex diseases. issues. To address these issues, we use the weighted-sum pooling method to test the joint association of multiple rare and common variants within a gene. The proposed method is applied to the Genetic Association Workshop 17 (GAW17) simulated mini-exome data to analyze multiple traits. Because of the nature of the GAW17 simulation model, improved power was buy SCH 900776 (MK-8776) not observed for multiple-trait analysis compared to single-trait analysis. However, multiple-trait analysis did not result in a substantial loss of power because of the screening of multiple qualities. We conclude that this method would be useful for identifying pleiotropic genes. Background The common disease/common variant hypothesis buy SCH 900776 (MK-8776) claims that common variants contribute considerably to common diseases [1,2]. Following this hypothesis, genome-wide association studies possess successfully recognized associations with common variants. However, such common variants explain only a small proportion of the phenotypic variance. Many of the as yet undetected common variants may have small effect sizes; consequently they are not expected to contribute significantly to the missing heritability. An alternative theory, the common disease/rare variant hypothesis, argues that a large number of rare variations with moderate to high penetrances account for genetic susceptibility to common disease [1]. Recently, deep-resequencing studies of candidate genes have offered some evidence assisting the common disease/rare variant hypothesis [3]. Although numerous statistical methods have been developed to detect associations with common variants for common diseases, these methods are inefficient for rare variants because of the small quantity of observations for each single rare variant. One feasible method for rare variant analysis is definitely to pool buy SCH 900776 (MK-8776) multiple rare buy SCH 900776 (MK-8776) variants within a gene or region and to test their joint effect. This category of methods has been examined by Dering et al. [4]. Some genetic association studies examine a qualitative trait, such as the case-control status and some additional correlated quantitative qualities. For example, a genetic study of diabetes may examine the diabetic status and additional related phenotypes, such as body mass index and additional lipid profiles. Similarly, a glaucoma study may explore the related endophenotypes, such as central corneal thickness, intraocular pressure, and maximum vertical cup-to-disc percentage. One of the ways to analyze these data is definitely to perform single-trait analyses separately. An alternative way is to perform a multiple-trait analysis, which potentially offers improved power to determine the pleiotropic variants for these qualities [5,6]. Univariate test statistics or = (denote the available traits. Presume that the gene offers genotyped single-nucleotide polymorphisms (SNPs), including both common and rare ones. In the first step, the genetic score of the gene for an individual is determined using the weighted sum of all SNPs within the gene. Second, a univariate test is performed to establish the association of genetic scores with all the traits separately. Then, a gene-level association test using the linear or quadratic combination of single-trait univariate statistics is constructed for multiple qualities. Finally, the optimal subset of qualities is selected for multiple-trait analysis. The details of the various steps are explained in what follows. Gene score using weighted sum The weighted-sum gene score assigns different weights to each variant based on the estimated allele frequencies [8]. The score for gene for individual is given by: (1) where is the number of small alleles for SNP in individual is the total number of small alleles for SNP in all individuals. In the original Rabbit Polyclonal to OR2A42 study [8], the allele frequencies were estimated only for the control subjects. Because multiple-trait analysis needs to analyze multiple quantitative qualities as well as the disease status, in the present study we estimate the allele frequencies using all individuals. Association test.