The Sensitivity of Spectral Clustering Applied to Gene Expression Data

In large scale numerical datasets which are to be ordered or clustered it is important to be able to measure the sensitivity of the orderings or clusterings. In this paper we combine statistical methods with techniques from numerical linear algebra to obtain an efficient numerical method to measure...

Full description

Saved in:
Bibliographic Details
Main Authors: Spence, A., Stoyanov, Z., Vass, J.K.
Format: Conference Proceeding
Language:English
Subjects:
Breast cancer
Computational efficiency
Gene expression
Large-scale systems
Linear algebra
Matrix decomposition
Sensitivity analysis
Singular value decomposition
Statistical analysis
Thumb
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In large scale numerical datasets which are to be ordered or clustered it is important to be able to measure the sensitivity of the orderings or clusterings. In this paper we combine statistical methods with techniques from numerical linear algebra to obtain an efficient numerical method to measure the sensitivity of spectral clustering. The method is applied to answer a question about the expression of genes MYH11 and NDE1 in breast cancer samples.
ISSN:2151-7614
2151-7622
DOI:10.1109/ICBBE.2007.347