Dataset Information

Analysis of case-control association studies with known risk variants.

ABSTRACT: The question of how to best use information from known associated variants when conducting disease association studies has yet to be answered. Some studies compute a marginal P-value for each Several Nucleotide Polymorphisms independently, ignoring previously discovered variants. Other studies include known variants as covariates in logistic regression, but a weakness of this standard conditioning strategy is that it does not account for disease prevalence and non-random ascertainment, which can induce a correlation structure between candidate variants and known associated variants even if the variants lie on different chromosomes. Here, we propose a new conditioning approach, which is based in part on the classical technique of liability threshold modeling. Roughly, this method estimates model parameters for each known variant while accounting for the published disease prevalence from the epidemiological literature.We show via simulation and application to empirical datasets that our approach outperforms both the no conditioning strategy and the standard conditioning strategy, with a properly controlled false-positive rate. Furthermore, in multiple data sets involving diseases of low prevalence, standard conditioning produces a severe drop in test statistics whereas our approach generally performs as well or better than no conditioning. Our approach may substantially improve disease gene discovery for diseases with many known risk variants.LTSOFT software is available online http://www.hsph.harvard.edu/faculty/alkes-price/software/.

SUBMITTER: Zaitlen N

PROVIDER: S-EPMC3381970 | biostudies-literature | 2012 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Analysis of case-control association studies with known risk variants.

Zaitlen Noah N Pasaniuc Bogdan B Patterson Nick N Pollack Samuela S Voight Benjamin B Groop Leif L Altshuler David D Henderson Brian E BE Kolonel Laurence N LN Le Marchand Loic L Waters Kevin K Haiman Christopher A CA Stranger Barbara E BE Dermitzakis Emmanouil T ET Kraft Peter P Price Alkes L AL

Bioinformatics (Oxford, England) 20120503 13

<h4>Motivation</h4>The question of how to best use information from known associated variants when conducting disease association studies has yet to be answered. Some studies compute a marginal P-value for each Several Nucleotide Polymorphisms independently, ignoring previously discovered variants. Other studies include known variants as covariates in logistic regression, but a weakness of this standard conditioning strategy is that it does not account for disease prevalence and non-random ascer ...[more]

PMID: 22556366

Similar Datasets

Project description:BackgroundX-ray repair cross-complementing group 1 (XRCC1) is one of the DNA repair genes encoding a scaffolding protein that participate in base excision repair (BER) pathway. However, studies on the association between polymorphisms in this gene and glioma have yielded conflicting results. This meta-analysis was performed to derive a more precise estimation between XRCC1 polymorphisms (Arg399Gln, Arg194Trp, and Arg280His) and glioma risk.MethodsData were collected from several electronic databases, with the last search up to November 28, 2012. Meta-analysis was performed by critically reviewing 9 studies for Arg399Gln polymorphism (3146 cases and 4296 controls), 4 studies for Arg194Trp polymorphism (2557 cases and 4347 controls), and 4 studies for Arg280His polymorphism (1936 cases and 2895 controls). All of the statistical analyses were performed using the software programs STATA (version 11.0).ResultsThe combined results showed that Arg399Gln polymorphism was significantly associated with glioma risk (Gln/Gln versus Arg/Arg: OR = 1.52, 95% CI = 1.03-2.23; recessive model: OR = 1.32, 95% CI = 1.01-1.73; additive model: OR = 1.21, 95% CI = 1.00-1.47), whereas Arg194Trp/Arg280His polymorphisms were all not significantly associated with glioma risk. As for ethnicity, Arg399Gln polymorphism was associated with increased risk of glioma among Asians (Gln/Gln versus Arg/Arg: OR = 1.78, 95% CI = 1.29-2.47; Arg/Gln versus Arg/Arg: OR = 1.28, 95% CI = 1.05-1.56; recessive model: OR = 1.59, 95% CI = 1.16-2.17; dominant model: OR = 1.36, 95% CI = 1.13-1.65; additive model: OR = 1.32, 95% CI = 1.15-1.52), but not among Caucasians. Stratified analyses by histological subtype indicated that the Gln allele of Arg399Gln polymorphism showed borderline association with the risk of glioblastoma among Caucasians. However, no evidence was observed in subgroup analyses for Arg194Trp/Arg280His polymorphisms.ConclusionsOur meta-analysis suggested that Arg399Gln polymorphism was associated with increased risk of glioma among Asians and borderline increased risk for glioblastoma among Caucasians, whereas Arg194Trp/Arg280His polymorphisms might have no influence on the susceptibility of glioma in different ethnicities.

Project description:Several studies have inspected the relationship between rs735482 polymorphism and the risk of some human cancers, but the findings remain controversial. We designed this meta-analysis to validate the association between rs735482 polymorphism and cancer risk. All articles were published before September 1, 2018 and searched in Pubmed, Embase, Web of Science, China National Knowledge Infrastructure, WangFang, and Chinese BioMedical databases, STATA 12.0 software was used for statistical analysis, which provides reasonable data and technical support for this article. A total of 10 studies were included in the meta-analysis, including 2652 cancer cases and 3536 rs735482 polymorphic controls. Data were directly extracted from these studies and odds ratios with 95% confidence intervals were computed to estimate the strength of the association. By pooling all eligible studies, the rs735482 polymorphism showed no significant association with susceptibility of several cancers in all the five genetic models (the allelic model: OR = 1.019, 95% CI: 0.916-1.134, P = .731). In addition, another adjusted OR data showed a significant increased risk between the rs735482 and susceptibility of several cancers (the codominant model BB vs AA: OR = 1.353, 95% CI: 1.033-1.774, P = .028) and the stratification analysis by ethnicity indicated the rs735482 is associated with an increased risk of cancer in Chinese group (BB vs AA, OR = 1.391, 95% CI = 1.054-1.837, P = .020; AB+BB vs AA OR = 1.253, 95% CI = 1.011-1.551, P = .039). However, the ERCC1 rs735482 is associated with a decreased risk of cancer in Italian group (AB vs AA, OR = 0.600, 95% CI = 0.402-0.859, P = .012; AB + BB vs AA, OR = 0.620, 95% CI = 0.424-0.908, P = .014). The results of this meta-analysis do not support the association between rs735482 polymorphism and cancer risk. But stratified analysis showed that rs735482 significantly increased the risk of cancer in Chinese while decreased the risk of cancer in Italian. Because of the limited number of samples, larger and well-designed researches are needed to estimate this association in detail.

Dataset Information

Analysis of case-control association studies with known risk variants.

Publications

Analysis of case-control association studies with known risk variants.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets