Pattern Recognition in Bioinformatics

Volume 5265 of the series Lecture Notes in Computer Science pp 250-261

Gene Selection for Microarray Data by a LDA-Based Genetic Algorithm

  • Edmundo Bonilla HuertaAffiliated withLERIA, Université d’Angers
  • , Béatrice DuvalAffiliated withLERIA, Université d’Angers
  • , Jin-Kao HaoAffiliated withLERIA, Université d’Angers

* Final gross prices may vary according to local VAT.

Get Access


Gene selection aims at identifying a (small) subset of informative genes from the initial data in order to obtain high predictive accuracy. This paper introduces a new wrapper approach to this difficult task where a Genetic Algorithm (GA) is combined with Fisher’s Linear Discriminant Analysis (LDA). This LDA-based GA algorithm has the major characteristic that the GA uses not only a LDA classifier in its fitness function, but also LDA’s discriminant coefficients in its dedicated crossover and mutation operators. The proposed algorithm is assessed on a set of seven well-known datasets from the literature and compared with 16 state-of-art algorithms. The results show that our LDA-based GA obtains globally high classification accuracies (81%-100%) with a very small number of genes (2-19).


Linear discriminant analysis genetic algorithm gene selection classification wrapper