A Novel Method for Feature Gene Selection Based on Geodesic Distance
Selecting feature genes is very important to identify disease and benefit to make drug. Today, clustering methods have been successfully applied to feature gene selection. In this paper, we use geodesic distance, instead of Euclidean distance, to cluster similar genes, as Geodesic distance can show the true geometric structure of the genes. We select the genes with the highest Signal-to-Noise Ratio (SNR) at each cluster, which is called feature gene subset. In order to get the number of clusters, a support vector machine (SVM) is used to train the subset which minimizes bounds on leave-one-out error. Then we select the subset with the smallest number which has better classification performance. The result of our experiments shows our method can be succesfully used to select feature genes with clustering algorithm.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
Document Type: Research Article
Publication date: 2010-06-01
More about this publication?
- Journal of Computational and Theoretical Nanoscience is an international peer-reviewed journal with a wide-ranging coverage, consolidates research activities in all aspects of computational and theoretical nanoscience into a single reference source. This journal offers scientists and engineers peer-reviewed research papers in all aspects of computational and theoretical nanoscience and nanotechnology in chemistry, physics, materials science, engineering and biology to publish original full papers and timely state-of-the-art reviews and short communications encompassing the fundamental and applied research.
- Editorial Board
- Information for Authors
- Submit a Paper
- Subscribe to this Title
- Terms & Conditions
- Ingenta Connect is not responsible for the content or availability of external websites