The results of this work, carried out at the Department of Medical Genetics (DGM) of the University of Lausanne (UNIL) and the SIB Swiss Institute of Bioinformatics, were published in the February 20th 2014 issue of PLOS Genetics.

While rather successful –with over 12,000 detected correlations– genome-wide association studies (GWAS) usually find genetic effects of small amplitude. Even combined, these effects explain only a small fraction of the heritability of a phenotype (i.e. the part of the variation of a phenotype that is genetic in origin). This phenomenon is in part due to the complexity of the clinical phenotypes usually investigated in GWAS: the biological path leading from genetic origin to complex phenotype is long and difficult to uncover.

Between gene and phenotype

The study was led by Rico Rueedi, postdoctoral researcher at the DGM and at the SIB, Sven Bergmann, DGM principal investigator and SIB group leader, and Zoltán Kutalik, now principal investigator at the Institute for Social and Preventative Medicine of the CHUV and SIB group leader, and investigates a class of less complex phenotypes: metabolite concentrations.

These phenotypes, called metabotypes, are located between genes and clinical phenotypes. Consequently, not only are genetic effects on metabotypes stronger – and therefore easier to detect – but metabotypes can also provide insight into the biological mechanisms that underlie the correlations between genetic and phenotypic variations. "In this study, which involves data from 1,000 individuals of the CoLaus cohort, two metabolites emerge as intermediaries; the first between the FUT2 gene and Crohn's disease (one of the inflammatory bowel diseases), and the second between the SLC7A9 gene and chronic kidney disease," specifies Rico Rueedi, first author of the article published in PLOS Genetics.

Towards new biomarkers

In addition to providing biological context, the two metabolites may prove to be pre-symptomatic biomarkers for Crohn's disease and chronic kidney disease. "If so, these metabolites would allow to diagnose these diseases before they become clinically expressed, and to undertake preventative therapeutic measures," concludes Rico Rueedi.


Rico Rueedi, Sven Bergmann, Computational Biology Group
Zoltán Kutalik, Statistical Genetics Group