Estimating Missing Heritability for Disease from Genome-wide Association Studies

Queensland Institute of Medical Research, 300 Herston Road, Herston, Queensland 4006, Australia.
The American Journal of Human Genetics (Impact Factor: 10.93). 03/2011; 88(3):294-305. DOI: 10.1016/j.ajhg.2011.02.002
Source: PubMed


Genome-wide association studies are designed to discover SNPs that are associated with a complex trait. Employing strict significance thresholds when testing individual SNPs avoids false positives at the expense of increasing false negatives. Recently, we developed a method for quantitative traits that estimates the variation accounted for when fitting all SNPs simultaneously. Here we develop this method further for case-control studies. We use a linear mixed model for analysis of binary traits and transform the estimates to a liability scale by adjusting both for scale and for ascertainment of the case samples. We show by theory and simulation that the method is unbiased. We apply the method to data from the Wellcome Trust Case Control Consortium and show that a substantial proportion of variation in liability for Crohn disease, bipolar disorder, and type I diabetes is tagged by common SNPs.

Download full-text


Available from: Sang Hong Lee,
  • Source
    • "Recent new mathematical methods enable the computation of the amount of risk of illness contingent on SNPs, which are usually determined by GWAS studies (Kendler, 2013). A corresponding study showed that a 41% risk for bipolar I disorder is defined by the GWAS data of Wellcome Trust Case Control Consortium (Lee et al., 2011). This is considerably more than in schizophrenia (14–24%) or somatic diseases, such as Crohn's disease (24%) and diabetes mellitus type 1 (31%). "
    [Show abstract] [Hide abstract]
    ABSTRACT: In bipolar disorders, there are unclear diagnostic boundaries with unipolar depression and schizophrenia, inconsistency of treatment guidelines, relatively long trial-and-error phases of treatment optimization, and increasing use of complex combination therapies lacking empirical evidence. These suggest that the current definition of bipolar disorders based on clinical symptoms reflects a clinically and etiologically heterogeneous entity. Stratification of treatments for bipolar disorders based on biomarkers and improved clinical markers are greatly needed to increase the efficacy of currently available treatments and improve the chances of developing novel therapeutic approaches. This review provides a theoretical framework to identify biomarkers and summarizes the most promising markers for stratification regarding beneficial and adverse treatment effects. State and stage specifiers, neuropsychological tests, neuroimaging, and genetic and epigenetic biomarkers will be discussed with respect to their ability to predict the response to specific pharmacological and psychosocial psychotherapies for bipolar disorders. To date, the most reliable markers are derived from psychopathology and history-taking, while no biomarker has been found that reliably predicts individual treatment responses. This review underlines both the importance of clinical diagnostic skills and the need for biological research to identify markers that will allow the targeting of treatment specifically to sub-populations of bipolar patients who are more likely to benefit from a specific treatment and less likely to develop adverse reactions. Copyright © 2015. Published by Elsevier B.V.
    European neuropsychopharmacology: the journal of the European College of Neuropsychopharmacology 01/2015; 58(3). DOI:10.1016/j.euroneuro.2014.12.006 · 4.37 Impact Factor
  • Source
    • "doi:10.1534/genetics.114.167916/-/DC1. et al. 2010; Brachi et al. 2011; Lee et al. 2011; Zuk et al. 2012). Heritability estimates are also of great relevance to plant breeders, as they give a measure for the breeding potential of a trait. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Heritability is a central parameter in quantitative genetics, both from an evolutionary and a breeding perspective. For plant traits heritability is traditionally estimated by comparing within and between genotype variability. This approach estimates broad-sense heritability, and does not account for different genetic relatedness. With the availability of high-density markers there is growing interest in marker based estimates of narrow-sense heritability, using mixed models in which genetic relatedness is estimated from genetic markers. Such estimates have received much attention in human genetics but are rarely reported for plant traits. A major obstacle is that current methodology and software assume a single phenotypic value per genotype, hence requiring genotypic means. An alternative that we propose here, is to use mixed models at individual plant or plot level. Using statistical arguments, simulations and real data we investigate the feasibility of both approaches, and how these affect genomic prediction with G-BLUP and genome-wide association studies. Heritability estimates obtained from genotypic means had very large standard errors and were sometimes biologically unrealistic. Mixed models at individual plant or plot level produced more realistic estimates, and for simulated traits standard errors were up to 13 times smaller. Genomic prediction was also improved by using these mixed models, with up to a 49% increase in accuracy. For GWAS on simulated traits, the use of individual plant data gave almost no increase in power. The new methodology is applicable to any complex trait where multiple replicates of individual genotypes can be scored. This includes important agronomic crops, as well as bacteria and fungi. Copyright © 2014, The Genetics Society of America.
    Genetics 12/2014; 199(2). DOI:10.1534/genetics.114.167916 · 5.96 Impact Factor
  • Source
    • "). GWAS have had some success in identifying genes for human disease traits, where LD levels are relatively high, population structures extensively studied and funding available for very large sample sizes. However, even in humans GWAS have failed to identify loci responsible for a high percentage of the heritability of many traits (Lee et al., 2011 "
    [Show abstract] [Hide abstract]
    ABSTRACT: Genetic research on the interactions of temperate tree hosts with pests and pathogens, and breeding for resistance or low susceptibility, are hindered by the long generation time and large size of tree species. Foresters need to be quick to exploit new technologies that may accelerate research and breeding programmes, and opportunistic in gaining maximum use from existing experimental tree plots. A fruitful approach may be to apply new genomic methods to the analysis of established provenance and progeny trials, seed orchards and clonal archives, where screening for pests and pathogens may occur. An important test case of this approach is underway in Britain with respect to ash (Fraxinus excelsior L.) and the search for resistance to the dieback caused by Hymenoscyphus fraxineus (T. Kowalski) Baral, Queloz, Hosoya, comb. nov. This review examines: (1) the use of field trials for pathogen and pest resistance in forest trees, (2) how field trials may support the application of genomic technologies to tree health issues, (3) the extent of the field trial resource in Britain, (4) issues that constrain the use and maintenance of field trials, (5) an outline of possible experimental designs, (6) the use of natural systems and (7) funding of long-term trials. Application of the latest technologies may be critically dependent on the availability of well-designed and maintained, long-term, field trials that produce invaluable resources and results for decades.
    Forestry 12/2014; 88(1). DOI:10.1093/forestry/cpu046 · 2.09 Impact Factor
Show more