Perturbation and candidate analysis to combat overfitting of gene expression microarray data. (1st January 2011)

• Record Type: Journal Article
• Authors: Mathur, Ravi; David Schaffer, J.; Land, Walker H.; Heine, John J.; Hernandez, Jonathan M.; Yeatman, Timothy
• Is Part Of: International journal of computational biology and drug design. Volume 4:Number 4(2011)
• Page Start: 307
• Page End: 315
• Publication Date: 2011-01-01
• Subjects: Az value -- colorectal cancer -- cross-validation -- DNA microarray -- GA -- genetic algorithm -- noise perturbation -- overfitting -- perturbation -- ROC curve -- support vector machines; Computational biology -- Periodicals; Drugs -- Design -- Periodicals; 570.285
• DOI: 10.1504/IJCBDD.2011.044443
• Languages: English
• ISSNs: 1756-0756

Abstract

Analysis of gene expression microarray datasets presents the high risk of over-fitting (spurious patterns) because of their feature-rich but case-poor nature. This paper describes our ongoing efforts to develop a method to combat over-fitting and determine the strongest signal in the dataset. A GA-SVM hybrid along with Gaussian noise (manual noise gain) is used to discover feature sets of minimal size that accurately classifies the cases under cross-validation. Initial results on a colorectal cancer dataset shows that the strongest signal (modest number of candidates) can be found by a binary search.