Genome metabolome integrated network analysis to uncover connections between genetic variants and complex traits: an application to obesity.

TitleGenome metabolome integrated network analysis to uncover connections between genetic variants and complex traits: an application to obesity.
Publication TypeJournal Article
Year of Publication2014
AuthorsValcárcel B, Ebbels TMD, Kangas AJ, Soininen P, Elliot P, Ala-Korpela M, Jarvelin M-R, De Iorio M
JournalJ R Soc Interface
Volume11
Issue94
Pagination20130908
Date Published05/2014
ISSN1742-5662
KeywordsAdolescent, Adult, Female, Gene Regulatory Networks, Genome, Human, Genome-Wide Association Study, Humans, Infant, Male, Metabolome, Obesity, Quantitative Trait Loci
Abstract

Current studies of phenotype diversity by genome-wide association studies (GWAS) are mainly focused on identifying genetic variants that influence level changes of individual traits without considering additional alterations at the system-level. However, in addition to level alterations of single phenotypes, differences in association between phenotype levels are observed across different physiological states. Such differences in molecular correlations between states can potentially reveal information about the system state beyond that reported by changes in mean levels alone. In this study, we describe a novel methodological approach, which we refer to as genome metabolome integrated network analysis (GEMINi) consisting of a combination of correlation network analysis and genome-wide correlation study. The proposed methodology exploits differences in molecular associations to uncover genetic variants involved in phenotype variation. We test the performance of the GEMINi approach in a simulation study and illustrate its use in the context of obesity and detailed quantitative metabolomics data on systemic metabolism. Application of GEMINi revealed a set of metabolic associations which differ between normal and obese individuals. While no significant associations were found between genetic variants and body mass index using a standard GWAS approach, further investigation of the identified differences in metabolic association revealed a number of loci, several of which have been previously implicated with obesity-related processes. This study highlights the advantage of using molecular associations as an alternative phenotype when studying the genetic basis of complex traits and diseases.

DOI10.1098/rsif.2013.0908
Alternate JournalJ R Soc Interface
PubMed ID24573330
PubMed Central IDPMC3973353
Grant List1RL1MH083268-01 / MH / NIMH NIH HHS / United States
5R01HL087679-02 / HL / NHLBI NIH HHS / United States
5R01MH63706:02 / MH / NIMH NIH HHS / United States
G0500539 / / Medical Research Council / United Kingdom
G0600705 / / Medical Research Council / United Kingdom