PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Gene selection via the BAHSIC family of algorithms
Le Song, Justin Bedo, karsten borgwardt, Arthur Gretton and Alex Smola
Bioinformatics Volume 23, Number 13, i490-i498, 2007.


Motivation: Identifying significant genes among thousands of sequences on a microarray is a central challenge for cancer research in bioinformatics. The ultimate goal is to detect the genes that are involved in disease outbreak and progression. A multitude of methods have been proposed for this task of feature selection, yet the selected gene lists differ greatly between different methods. To accomplish biologically meaningful gene selection from microarray data, we have to understand the theoretical connections and the differences between these methods. In this article, we define a kernel-based framework for feature selection based on the Hilbert–Schmidt independence criterion and backward elimination, called BAHSIC. We show that several well-known feature selectors are instances of BAHSIC, thereby clarifying their relationship. Furthermore, by choosing a different kernel, BAHSIC allows us to easily define novel feature selection algorithms. As a further advantage, feature selection via BAHSIC works directly on multiclass problems. Results: In a broad experimental evaluation, the members of the BAHSIC family reach high levels of accuracy and robustness when compared to other feature selection techniques. Experiments show that features selected with a linear kernel provide the best classification performance in general, but if strong non-linearities are present in the data then non-linear kernels can be more suitable. Availability: Accompanying homepage is

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Theory & Algorithms
ID Code:3113
Deposited By:Arthur Gretton
Deposited On:19 December 2007