Search

Complex trait variation is likely to be explained by the combined effects of genes, environmental factors, and gene × environment (G × E) interaction. The authors introduce a novel 2-step method for detecting a G × E interaction in a genome-wide association study (GWAS) of case-parent trios. The method utilizes 2 sources of G × E information in a trio sample to construct a screening step and a testing step. Across a wide range of models, this 2-step procedure provides substantially greater power to detect G × E interaction than a standard test of G × E interaction applied genome-wide. For example, for a disease susceptibility locus with minor allele frequency of 15%, a binary exposure variable with 50% prevalence, and a GWAS scan of 1 million markers in 1,000 case-parent trios, the 2-step method provides 87% power to detect a G × E interaction relative risk of 2.3, as compared with only 25% power using a standard G × E test. The method is easily implemented using standard software. This 2-step scan for G × E interaction is independent of any prior scan that may have been conducted for genetic main effects, and thus has the potential to uncover new genes in a GWAS that have not been previously identified. [ABSTRACT FROM PUBLISHER]

Paraoxonase 1 (PON1) protects against oxidative modification of low density lipoproteins. The PON1 gene has 3 polymorphisms considered strong determinants of PON1 levels: Q192R and L55M in the coding region and C-108T in the promoter region. PON1 levels are also influenced by smoking. The authors hypothesized that PON1 variants could increase the risk of vascular thrombosis, leading in turn to placental insufficiency and small-for-gestational-age (SGA) birth. The author compared PON1 variants and haplotypes between 448 newborn SGA cases and 431 newborn controls from Montréal, Québec, Canada (1998–2000) and studied the effects of interaction with maternal smoking. Transmission of the variants in case-parent trios was used as validation of the case-control results; the authors combined case-control and family data to analyze the associations of variants with SGA birth. In the case-control analysis, the T allele from C-108T increased the risk of SGA birth (additive odds ratio = 1.30, 95% confidence interval (CI): 1.06, 1.59), and the TRL haplotype (T from C-108T, R from Q192R, and L from L55M) was associated with an odds ratio of 1.51 (95% CI: 1.07, 2.15). Among smokers, the CRL haplotype was protective (odds ratio = 0.48, 95% CI: 0.28, 0.82). Case-parent trio results were compatible with case-control results. [ABSTRACT FROM PUBLISHER]

For most associations of common single nucleotide polymorphisms (SNPs) with common diseases, the genetic model of inheritance is unknown. The authors extended and applied a Bayesian meta-analysis approach to data from 19 studies on 17 replicated associations with type 2 diabetes. For 13 SNPs, the data fitted very well to an additive model of inheritance for the diabetes risk allele; for 4 SNPs, the data were consistent with either an additive model or a dominant model; and for 2 SNPs, the data were consistent with an additive or recessive model. Results were robust to the use of different priors and after exclusion of data for which index SNPs had been examined indirectly through proxy markers. The Bayesian meta-analysis model yielded point estimates for the genetic effects that were very similar to those previously reported based on fixed- or random-effects models, but uncertainty about several of the effects was substantially larger. The authors also examined the extent of between-study heterogeneity in the genetic model and found generally small between-study deviation values for the genetic model parameter. Heterosis could not be excluded for 4 SNPs. Information on the genetic model of robustly replicated association signals derived from genome-wide association studies may be useful for predictive modeling and for designing biologic and functional experiments. [ABSTRACT FROM PUBLISHER]

In genetic association studies, investigators compare allele or genotype frequencies in unrelated case and control subjects or examine preferential allele transmissions from parents to affected offspring. In many genetic case-control studies, the collection of DNA material extends to relatives such as parents of cases. Thus, case-control and case-parent trio association analyses are possible. Whereas the goal of collecting genetic information from family members in a study initially designed as a case-control study is to enrich the genetic analysis, increase power, or address concern about population structure bias, methods of combining genetic data from unrelated case and control subjects with genetic trio data from the same study population are not well known. A number of hybrid approaches have been developed that utilize such data together. In this paper, the authors describe key features of genetic case-control and case-parent trio studies and review commonly used methods of genetic analysis for case-parent trio designs. In addition, they provide a pragmatic review of statistical methods and available software for existing hybrid approaches that combine various components of case-control and genetic trio data. The application of all methods is illustrated using a candidate gene study of childhood leukemia that included case-control subjects and their parents. [ABSTRACT FROM PUBLISHER]