Identification of differentially expressed genes: a comparison between wrapper and filter methods - CentraleSupélec Access content directly
Conference Papers Year :

Identification of differentially expressed genes: a comparison between wrapper and filter methods

Vincent Guillemot
Laurent Le Brusquet
Gilles Fleury
  • Function : Author

Abstract

In order to analyze transcriptomic data along with prior information in a classification context, we are studying the generation of a prediction rule. Transcriptomic data classically present the expression of thousands of genes (variables) with most often only a hundred experiments (individuals). Consequently classical classification methods are not adapted. The solution usually adopted to address this issue is to sort the variables by their relevance, in order to select those which will be used in the classification process.
No file

Dates and versions

hal-00257928 , version 1 (20-02-2008)

Identifiers

  • HAL Id : hal-00257928 , version 1

Cite

Vincent Guillemot, Vincent Frouin, Laurent Le Brusquet, Gilles Fleury. Identification of differentially expressed genes: a comparison between wrapper and filter methods. Proceedings of the International Workshop on Machine Learning in Systems Biology, Sep 2007, Evry, France. ⟨hal-00257928⟩
62 View
0 Download

Share

Gmail Facebook Twitter LinkedIn More