Rare variant density across the genome and across populations

Gene-Environment Interactions in Asthma: Genetic and Epigenetic Effects

Over the past three decades, a large number of genetic studies have been aimed at finding genetic variants associated with the risk of asthma, applying various genetic and genomic approaches including linkage analysis, candidate gene polymorphism studies, and genome-wide association studies (GWAS). However, contrary to general expectation, even single nucleotide polymorphisms (SNPs) discovered by GWAS failed to fully explain the heritability of asthma. Thus, application of rare allele polymorphisms in well defined phenotypes and clarification of environmental factors have been suggested to overcome the problem of 'missing' heritability. Such factors include allergens, cigarette smoke, air pollutants, and infectious agents during pre- and post-natal periods. The first and simplest interaction between a gene and the environment is a candidate interaction of both a well known gene and environmental factor in a direct physical or chemical interaction such as between CD14 and endotoxin or between HLA and allergens. Several GWAS have found environmental interactions with occupational asthma, aspirin exacerbated respiratory disease, tobacco smoke-related airway dysfunction, and farm-related atopic diseases. As one of the mechanisms behind gene-environment interaction is epigenetics, a few studies on DNA CpG methylation have been reported on subphenotypes of asthma, pitching the exciting idea that it may be possible to intervene at the junction between the genome and the environment. Epigenetic studies are starting to include data from clinical samples, which will make them another powerful tool for re-search on gene-environment interactions in asthma.

Identification of rare variants in Alzheimer's disease

Much progress has been made in recent years in identifying genes involved in the risk of developing Alzheimer's disease (AD), the most common form of dementia. Yet despite the identification of over 20 disease associated loci, mainly through genome wide association studies (GWAS), a large proportion of the genetic component of the disorder remains unexplained. Recent evidence from the AD field, as with other complex diseases, suggests a large proportion of this "missing heritability" may be due to rare variants of moderate to large effect size, but the methodologies to detect such variants are still in their infancy. The latest studies in the field have been focused on the identification of coding variation associated with AD risk, through whole-exome or whole-genome sequencing. Such variants are expected to have larger effect sizes than GWAS loci, and are easier to functionally characterize, and develop cellular and animal models for. This review explores the issues involved in detecting rare variant associations in the context of AD, highlighting some successful approaches utilized to date.

Novel CYP2B6 enzyme variants in a Rwandese population: functional characterization and assessment of in silico prediction tools

Cytochrome P450 CYP2B6 is a highly polymorphic enzyme that metabolizes numerous drugs, pesticides, and environmental toxins. Sequence analysis of a Rwandese population identified eight functionally uncharacterized nonsynonymous variants c.329G>T (p.G110V), c.341T>C (p.I114T), c.444G>T (p.E148D), c.548T>G (p.V183G), c.637T>C (p.F213L), c.758G>A (p.R253H), c.835G>C (p.A279P), and c.1459C>A (p.R487S), and five novel alleles termed CYP2B6*33 to CYP2B6*37 were assigned. Recombinant expression in COS-1 cells and functional characterization using the antidepressant bupropion and the antiretroviral efavirenz (EFV) as substrates demonstrated complete or almost complete loss-of-function for variants p.G110V, p.I114T, p.V183G, and p.F213L, whereas p.E148D, p.R253H, p.A279P, and p.R487S variants were functional. The data were used to assess the predictive power of eight online available functional prediction programs for amino-acid changes. Although none of the programs correctly predicted the functionality of all variants, substrate docking simulation analyses indicated similar conformational changes by all four deleterious mutations within the enzyme's active site, thus explaining lack of enzymatic function for both substrates. Because low-activity alleles of CYP2B6 are associated with impaired EFV metabolism and adverse drug response, these results are of potential utility for personalized treatment strategies in HIV/AIDS therapy.

Empirical power of very rare variants for common traits and disease: results from sanger sequencing 1998 individuals

The optimal study design for identifying rare variants associated with common disease is not yet clear and researchers have to decide whether to prioritize lower sequencing coverage on larger sample sizes, or higher coverage on smaller sample sizes. High-coverage sequencing affords several advantages, such as genotype accuracy and improved identification of very rare variants, but this comes at increased cost. However, the magnitude of the contribution of very rare variants to the statistical power of gene-based association tests is unknown. By using Sanger sequence data on seven genes from 1998 subjects with simulated phenotypes, we provide evidence that excluding very rare variants, in general, reduces the statistical power of rare variant association tests only modestly. However, if the probability of being causal and the effect size of the causal variants are inversely related to the minor allele frequency, then very rare variants do contribute to some power, however the absolute power remains low. As very rare variants constitute the majority of variants identified in sequencing studies, these findings suggest that careful attention need to be placed on the plausible relationship that exist between very rare variants and common disease.

Analysis of exome sequences with and without incorporating prior biological knowledge

Next-generation sequencing technology provides new opportunities and challenges in the search for genetic variants that underlie complex traits. It will also presumably uncover many new rare variants, but exactly how these variants should be incorporated into the data analysis remains a question. Several papers in our group from Genetic Analysis Workshop 17 evaluated different methods of rare variant analysis, including single-variant, gene-based, and pathway-based analyses and analyses that incorporated biological information. Although the performance of some of these methods strongly depends on the underlying disease model, integration of known biological information is helpful in detecting causal genes. Two work groups demonstrated that use of a Bayesian network and a collapsing receiver operating characteristic curve approach improves risk prediction when a disease is caused by many rare variants. Another work group suggested that modeling local rather than global ancestry may be beneficial when controlling the effect of population structure in rare variant association analysis.