TFE 2010-2011 (final year project)

Efficient genome-wide two-locus epistasis testing in disease association testing with quantitative traits

Analyzing the combined effects of genes and/or environmental factors on the development of complex diseases is a great challenge from both the statistical and computational perspective, even using a relatively small number of genetic and non-genetic exposures. Several data mining methods have been proposed for interaction analysis, among them, the Model-Based Multifactor Dimensionality Reduction (MB-MDR, Calle et al 2008). This is a relatively new MDR-based technique that is able to unify the best of both non-parametric and parametric worlds. It was developed to address some of the remaining concerns that go along with an MDR-analysis. One of the major advantages is that it is able to identify multiple significant epistasis model at the time, via a permutation strategy to assess significance, while maintaining a global 5% significance level. Moreover, it has recently been shown that MB-MDR adequately controls false positive rates (Mahachie et al 2010 –in submission).

The topic of this thesis is to compare the performance of the MB-MDR technique to the method of Wei et al (2010). The latter method was particularly developed to address the poorly explored issue of the control of false positive rate in the mapping of pair-wise epistatic quantitative trait loci. First, dig into the MDR and MB-MDR literature to fully understand the dimensionality reduction aspect and the pros and cons of the MB-MDR approach. Second, read Wei et al (2010) and try to implement their strategy. Third, apply MB-MDR and the Wei method to simulated data with 500 replicates. These data will be provided to you, so that you can focus on extracting the results and to summarize the power and false positive rate performance of both methods. Fourth, discuss your findings.

It is of utmost importance to be able to control the false positive findings in epstasis screens, especially in the light of next generation sequencing efforts (Figure 1). Depending on the progress made in this project, the work may lead to a genuine scientific publication.

See van_steen_5_2011.doc for references and figures.

Renseignements, Promoteur: