Graph Constrained Discriminant Analysis: a new method for the integration of a graph into a classification process

Vincent~Guillemot^a,b, Arthur Tenenhaus^a,b Laurent Le Brusquet^b and Vincent Frouin^a

^a CEA, Laboratoire d'Exploration Fonctionnelle des Génomes, 2 rue Gaston Crémieux - 91000 Evry

^b Department of Signal and Electronic Systems, Supélec, Plateau de Moulon, 3 rue Joliot-Curie, Gif sur Yvette, 91192, France

R package for Windows R package for Linux Real Datasets

Abstract

Integrating gene regulatory networks (GRNs) into the classification process of DNA microarrays is an important issue in bioinformatics, both because this information has a true theoretical interest and because it helps in the interpretation of the final classifier. We present a method called graph-constrained discriminant analysis (gCDA), which aims to integrate the information contained in one or several GRNs into a classification procedure.

Using simulated and real microarray datasets, we show that gCDA is able to integrate information into a correct graph, which results in an improvement of the classification performance. When the integrated graph includes erroneous information, the performance of the computed classifier is only slightly worse. The gCDA framework also allows the classification process to take into account as many a priori graphs as there are classes in the dataset.

The gCDA procedure was applied to simulated data and to three publicly available microarray datasets. gCDA shows an improvement in classification performance when compared to state-of-the-art classification methods.

R package for Windows R package for Linux Real Datasets