Imperial College London
Portrait Picture

DrCliveHoggart

Faculty of Medicine, Department of Infectious Disease

Honorary Research Officer
 
 
 
//

Contact

 

+44 (0)20 7594 3915c.hoggart

 
 
//

Location

 

231Norfolk PlaceSt Mary's Campus

//

Summary

 

Publications

Citation

BibTex format

@article{O'Reilly:2012:10.1371/journal.pone.0034861,
author = {O'Reilly, PF and Hoggart, CJ and Pomyen, Y and Calboli, FCF and Elliott, P and Jarvelin, M-R and Coin, LJM},
doi = {10.1371/journal.pone.0034861},
journal = {PLOS One},
title = {MultiPhen: Joint Model of Multiple Phenotypes Can Increase Discovery in GWAS},
url = {http://dx.doi.org/10.1371/journal.pone.0034861},
volume = {7},
year = {2012}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - The genome-wide association study (GWAS) approach has discovered hundreds of genetic variants associated with diseasesand quantitative traits. However, despite clinical overlap and statistical correlation between many phenotypes, GWAS aregenerally performed one-phenotype-at-a-time. Here we compare the performance of modelling multiple phenotypes jointlywith that of the standard univariate approach. We introduce a new method and software, MultiPhen, that models multiplephenotypes simultaneously in a fast and interpretable way. By performing ordinal regression, MultiPhen tests the linearcombination of phenotypes most associated with the genotypes at each SNP, and thus potentially captures effects hiddento single phenotype GWAS. We demonstrate via simulation that this approach provides a dramatic increase in power inmany scenarios. There is a boost in power for variants that affect multiple phenotypes and for those that affect only onephenotype. While other multivariate methods have similar power gains, we describe several benefits of MultiPhen overthese. In particular, we demonstrate that other multivariate methods that assume the genotypes are normally distributed,such as canonical correlation analysis (CCA) and MANOVA, can have highly inflated type-1 error rates when testing casecontrolor non-normal continuous phenotypes, while MultiPhen produces no such inflation. To test the performance ofMultiPhen on real data we applied it to lipid traits in the Northern Finland Birth Cohort 1966 (NFBC1966). In these dataMultiPhen discovers 21% more independent SNPs with known associations than the standard univariate GWAS approach,while applying MultiPhen in addition to the standard approach provides 37% increased discovery. The most associatedlinear combinations of the lipids estimated by MultiPhen at the leading SNPs accurately reflect the Friedewald Formula,suggesting that MultiPhen could be used to refine the definition of existing phenotypes or uncover novel heritablepheno
AU  - O'Reilly,PF
AU  - Hoggart,CJ
AU  - Pomyen,Y
AU  - Calboli,FCF
AU  - Elliott,P
AU  - Jarvelin,M-R
AU  - Coin,LJM
DO  - 10.1371/journal.pone.0034861
PY  - 2012///
SN  - 1932-6203
TI  - MultiPhen: Joint Model of Multiple Phenotypes Can Increase Discovery in GWAS
T2  - PLOS One
UR  - http://dx.doi.org/10.1371/journal.pone.0034861
UR  - http://hdl.handle.net/10044/1/28720
VL  - 7
ER  -
Download