Problems with Genome-Wide Association Studies

University of Alabama at Birmingham, Birmingham, Alabama, United States
Science (Impact Factor: 31.48). 07/2007; 316(5833):1840-2. DOI: 10.1126/science.316.5833.1840a
Source: PubMed
1 Follower
  • Source
    • "Unfortunately, for those common diseases typically regarded as complex (i.e. involving more than a single loci in the determination of phenotype) this approach has yielded limited success [10,11]. The last decade has seen a gradual acknowledgment of disease complexity and greater focus on strategies for the detection of complex disease associations within clinical data [1,12-14]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Background The statistical genetics phenomenon of epistasis is widely acknowledged to confound disease etiology. In order to evaluate strategies for detecting these complex multi-locus disease associations, simulation studies are required. The development of the GAMETES software for the generation of complex genetic models, has provided the means to randomly generate an architecturally diverse population of epistatic models that are both pure and strict, i.e. all n loci, but no fewer, are predictive of phenotype. Previous theoretical work characterizing complex genetic models has yet to examine pure, strict, epistasis which should be the most challenging to detect. This study addresses three goals: (1) Classify and characterize pure, strict, two-locus epistatic models, (2) Investigate the effect of model ‘architecture’ on detection difficulty, and (3) Explore how adjusting GAMETES constraints influences diversity in the generated models. Results In this study we utilized a geometric approach to classify pure, strict, two-locus epistatic models by “shape”. In total, 33 unique shape symmetry classes were identified. Using a detection difficulty metric, we found that model shape was consistently a significant predictor of model detection difficulty. Additionally, after categorizing shape classes by the number of edges in their shape projections, we found that this edge number was also significantly predictive of detection difficulty. Analysis of constraints within GAMETES indicated that increasing model population size can expand model class coverage but does little to change the range of observed difficulty metric scores. A variable population prevalence significantly increased the range of observed difficulty metric scores and, for certain constraints, also improved model class coverage. Conclusions These analyses further our theoretical understanding of epistatic relationships and uncover guidelines for the effective generation of complex models using GAMETES. Specifically, (1) we have characterized 33 shape classes by edge number, detection difficulty, and observed frequency (2) our results support the claim that model architecture directly influences detection difficulty, and (3) we found that GAMETES will generate a maximally diverse set of models with a variable population prevalence and a larger model population size. However, a model population size as small as 1,000 is likely to be sufficient.
    BioData Mining 06/2014; 7:8. DOI:10.1186/1756-0381-7-8 · 1.54 Impact Factor
  • Source
    • "However, in the second association analysis, the comparison of the SR and the HEP and CIR groups did not yield significant results, even when merging the first- and second-stage samples in a meta-analysis. This could have been due to variation in the sampling cohort, environmental interactions, inadequate statistical power, or gene interactions [1, 44-49]. Furthermore, information on factors important for the progression of liver disease was lacking in the samples analyzed, such as data on alcohol consumption [50]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: CD8+ T cells are key factors mediating hepatitis B virus (HBV) clearance. However, these cells are killed through HBV-induced apoptosis during the antigen-presenting period in HBV-induced chronic liver disease (CLD) patients. Interferon-inducible protein 6 (IFI6) delays type I interferon-induced apoptosis in cells. We hypothesized that single nucleotide polymorphisms (SNPs) in the IFI6 could affect the chronicity of CLD. The present study included a discovery stage, in which 195 CLD patients, including chronic hepatitis B (HEP) and cirrhosis patients and 107 spontaneous recovery (SR) controls, were analyzed. The genotype distributions of rs2808426 (C > T) and rs10902662 (C > T) were significantly different between the SR and HEP groups (odds ratio [OR], 6.60; 95% confidence interval [CI], 1.64 to 26.52, p = 0.008 for both SNPs) and between the SR and CLD groups (OR, 4.38; 95% CI, 1.25 to 15.26; p = 0.021 and OR, 4.12; 95% CI, 1.18 to 14.44; p = 0.027, respectively). The distribution of diplotypes that contained these SNPs was significantly different between the SR and HEP groups (OR, 6.58; 95% CI, 1.63 to 25.59; p = 0.008 and OR, 0.15; 95% CI, 0.04 to 0.61; p = 0.008, respectively) and between the SR and CLD groups (OR, 4.38; 95% CI, 1.25 to 15.26; p = 0.021 and OR, 4.12; 95% CI, 1.18 to 14.44; p = 0.027, respectively). We were unable to replicate the association shown by secondary enrolled samples. A large-scale validation study should be performed to confirm the association between IFI6 and HBV clearance.
    03/2013; 11(1):15-23. DOI:10.5808/GI.2013.11.1.15
    • "In other words, partitioning of germplasm collections before sampling ensures that both the genetic and ecological properties of germplasm collections are fully represented in core collections (Brown 1995; van Hintum et al. 2000). Determination of genetic structures in germplasm collections is, also, a crucial aspect of association studies (Wang et al. 2005; Shriner et al. 2007) and general agreement exists among researchers that incorporating population structure into statistical models used in association mapping is necessary to avoid false positives (Pritchard et al. 2000b; Flint-Garcia et al. 2003; Zhu et al. 2008). "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, we used 20 morphological traits (during two consecutive growing seasons) and 11 microsatellite markers to assess the morphological and molecular variability and structure of the almond (Prunus dulcis (Mill.) D.A. Webb). Seventy one promising Iranian genotypes and three foreign reference cultivars (Ferragnes, Supernova, and Touno) were evaluated in this study. Kernel/shell ratio, kernel width/thickness ratio, softness of shell, nut weight, and kernel thickness were highly variable. Strong positive and, occasionally, negative correlations were detected among nut and kernel traits. Morphological traits were categorized by principle-components analysis (PCA) into 6 components which explained 88.1 % of the total variation. On the basis of the first two PCA axes, the 2-dimensional PCA plot grouped the samples according to their phenotypic characteristics. The results from molecular analyses (including a Bayesian clustering approach and a molecular phylogenetic network) did not correspond to morphological groupings. In this paper we report, for the first time, morphological and molecular variability and genetic structure of Iranian almond germplasm. Our results showed that model-based cluster analysis (using Structure software) was very appropriate for study of genetic relationships among almond accessions and can be used for study of the genetic structure of Prunus germplasm as well.
    Plant Systematics and Evolution 12/2012; 298(10). DOI:10.1007/s00606-012-0691-8 · 1.15 Impact Factor
Show more