Subject 2: "Identification of genes affecting complex traits by mixed model methodology"

BACKGROUND

Identifying the genes influencing complex traits (e.g. diseases or other phenotypes of interest affected by multiple interacting genetic and environmental factors) is one of the hottest topics in modern biology. Indeed, the identification of genetic risk factors paves the way towards novel diagnostic as well as therapeutic/intervention strategies both in medicine and agriculture. In the medical field, the identification of genes influencing susceptibility to complex traits reveals novel targets for drug development by the pharmaceutical industry, and allows for the implementation of strategies towards personalized medicine. In addition, understanding the molecular architecture of complex traits is of major fundamental interest if we want to gain a better understanding of how natural selection is shaping life. Due to recent advances in marker genotyping technology, which allow for the genotyping of hundreds of thousands of Single Nucleotide Polymorphisms (SNPs) the genes can now directly be located by means of so-called "whole genome association studies".

RESPONSIBILITIES

In the mixed model methodology the Quantitative Trait Loci (QTL) effects of individual chromosomes are modeled as "random" effects sampled from multivariate normal distributions. The covariances between individual QTL effects are estimated conditional on flanking marker data using either pedigree information (i.e. linkage analysis) and/or population information (i.e. linkage disequilibrium analysis). Recently, this approach has been extended to allow for the simultaneous analysis of (i) multiple traits, and (ii) multiple QTL. The research project will first aim at improving the statistical models exploited by the existing methods in several respects: simultaneous handling of a large number of QTL effects, mutation and recombination, epistatic interactions, polygenic background effects, and coping with high density markers. The second part of the project will aim at extending the methods developed in part 1 to estimate genome wide "relative risk" in humans, "breeding values" in animals. The developments will be exploited and validated in the context of real-life datasets of medical and agronomic importance.

REQUIREMENTS

The ideal candidate would have a Master degree (or equivalent) in Mathematics or in Agronomic Engineering, with a strong interest in computational statistics and genetics.

LOCALE

The thesis project will be carried out in the Molecular and Factorial Genetics Unit of the CBIG/GIGA Centre of the University of Liège, under the supervision of Prof. Michel Georges, and in close collaboration with the Bioinformatics and Modeling Unit, under the supervision of Prof. Louis Wehenkel.

COMPENSATION

Full funding is available for a period of up to four years.

HOW TO APPLY

Applicants should send a short CV and motivation letter by email to Prof. Michel Georges (Michel.Georges[at]ulg.ac.be), from whom further information can be obtained.

DEADLINE

2006

MORE INFO

http://www.montefiore.ulg.ac.be/services/stochastic/new/
http://www.fmv.ulg.ac.be/genmol/Department/Molecular_Genetics/Molecular_Genetics.htm
http://www.montefiore.ulg.ac.be
http://www.giga.ulg.ac.be
http://www.ulg.ac.be