The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Thursday 1 Mai
2014 at 11h00 in the Salle Delachaux (Biopôle 2, premier étage, Route de la Corniche 10, Lausanne, M2: Vennes).
The speaker will be Daniel Wegmannn
(University of Fribourg, Switzerland) who will speak about Model based inference from genomic data - Approximate and some less approximate attempts
Abstract: The rapidly increasing amount of available sequencing and genotyping data of huge samples triggers the desire to contrast or infer parameters from ever more complex models. Unfortunately, the likelihood
functions of such models are often unknown, intractable or simply computationally out of reach. Here I will present two major strategies in such cases through examples: approximate inference and full likelihood inference of approximate models. I will begin
by presenting the Approximate Bayesian Computation framework and show an application to GWAS data. I will then present two ongoing research projects where we use full likelihood inference of approximate models, firstly a novel approximate Wright-Fisher process
to infer parameters from genetic data obtained at different time points, and secondly an attempt to formulate a model to infer gene-trait associations along with population structure while taking linkage partly into account.
Hope to see you there!
Valentin
(for the organizers: Jérôme Goudet, Valentin Rousson, Frédéric Schütz)