Section: New Results

A Method to Combine Combinatorial Optimization and Statistics to Mine High-Throughput Genotyping Data

Participants : J. Hamon, C. Dhaenens, J. Jacques.

In collaboration with Gènes Diffusion, we are interested in high-throughput genotyping data in order to select a subset of genes explaining a trait of interest. We suggest to study these high-throughput data combining combinatorial optimization and statistical methods. A first method based on an ILS (Iterated Local Search) and using a statistical criterion to calculate the fitness was suggested and compared with classical statistical approaches.