Comparison of genetic association strategies in the presence of rare alleles

BMC Proceedings

Proceedings

Alain Empain¹  Lizzy De Lobel²  François Van Lishout³  Kristel Van Steen³  Jestinah M Mahachie John³  Tom Cattaert³ 

[1] Bioinformatics and Modeling, GIGA-R, University of Liege, Avenue de l’Hôpital 1, 4000, Liège, Belgium;Department of Applied Mathematics and Computer Science, Ghent University, Krijgslaan 281 S9, 9000, Ghent, Belgium;Systems and Modeling Unit, Montefiore Institute, University of Liege, Grande Traverse 10, 4000, Liège, Belgium;Bioinformatics and Modeling, GIGA-R, University of Liege, Avenue de l’Hôpital 1, 4000, Liège, Belgium;

关键词: Minor Allele Frequency;    Rare Variant;    Unrelated Individual;    Family Data;    Multiple Testing Corrective Method;   

DOI  :  10.1186/1753-6561-5-S9-S32

来源: Springer

【 摘 要 】

In the quest for the missing heritability of most complex diseases, rare variants have received increased attention. Advances in large-scale sequencing have led to a shift from the common disease/common variant hypothesis to the common disease/rare variant hypothesis or have at least reopened the debate about the relevance and importance of rare variants for gene discoveries. The investigation of modeling and testing approaches to identify significant disease/rare variant associations is in full motion. New methods to better deal with parameter estimation instabilities, convergence problems, or multiple testing corrections in the presence of rare variants or effect modifiers of rare variants are in their infancy. Using a recently developed semiparametric strategy to detect causal variants, we investigate the performance of the model-based multifactor dimensionality reduction (MB-MDR) technique in terms of power and family-wise error rate (FWER) control in the presence of rare variants, using population-based and family-based data (FAM-MDR). We compare family-based results obtained from MB-MDR analyses to screening findings from a quantitative trait Pedigree-based association test (PBAT). Population-based data were further examined using penalized regression models. We restrict attention to all available single-nucleotide polymorphisms on chromosome 4 and consider Q1 as the outcome of interest. The considered family-based methods identified marker C4S4935 in the VEGFC gene with estimated power not exceeding 0.35 (FAM-MDR), when FWER was kept under control. The considered population-based methods gave rise to highly inflated FWERs (up to 90% for PBAT screening).

【 授权许可】

Unknown   
© Mahachie John et al; licensee BioMed Central Ltd. 2011. This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.