Article Preview
TopGene expression analysis and classification has received significant attention in recent years. One of the main challenges in classifying gene expression data is that the number of genes is typically much higher than the number of analysed samples. Also, is it not clear which genes are important and which can be omitted without reducing the classification performance.
Many pattern classification techniques have been employed to analyse microarray data. Golub et al. (1999) used a weighted voting scheme, while Fort and Lambert-Lacroix (2005) employed partial least squares and logistic regression techniques and Furey et al. (2000) ap-plied support vector machines. Dudoit et al. (2002) investigated nearest neighbour classifiers, discriminant analysis, classification trees and boosting, while Statnikov et al. (2005) explored several support vector machine techniques, nearest neighbour classifiers, neural networks and probabilistic neural networks. Midelfart et al. (2002) evaluated several rough set-based classification approaches for gastric cancer diagnosis from gene expression data, while Driscoll et al. (2003) employed genetic programming for analysing gene expression data and Kim and Cho (2003) used evolutionary algorithms to design artificial neural networks in the context of colon cancer identification.
In several of these studies it has been found that no one classification algorithm is performing best on all datasets and that hence the exploration of several classifiers is useful. Similarly, no universally ideal gene selection method has yet been found as several works (Liu et al., 2002; Statnikov et al., 2005) have shown.