Detection of phenotypes in microarray data using force-directed placement transforms

Authors:
Dragana Veljkovic Perez;Kay A. Robbins
Affiliations:
Department of Computer Science, The University of Texas at San Antonio;Department of Computer Science, The University of Texas at San Antonio
Venue:
MLDM'11 Proceedings of the 7th international conference on Machine learning and data mining in pattern recognition
Year:
2011

Citing 5
Cited 0

Information Retrieval

Information Retrieval
Biclustering of Expression Data

Proceedings of the Eighth International Conference on Intelligent Systems for Molecular Biology
Biclustering Algorithms for Biological Data Analysis: A Survey

IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB)
Performance of feature-selection methods in the classification of high-dimension data

Pattern Recognition
Force Feature Spaces for Visualization and Classification

ICMLA '08 Proceedings of the 2008 Seventh International Conference on Machine Learning and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Distinct cancer phenotypes are observable within a single type of cancer, corresponding to patients with different disease subtypes, prognosis, and treatment response. Analysis of correlations between genes and patients allows detection of gene sets that differentiate between these cancer phenotypes. We investigate the effect of force-directed placement transforms on bicluster-based feature selection for phenotype and marker detection. The transforms incorporate class-based metadata directly into the dataset topology and sharpen differences between classes. By incorporating important external clinical information such as disease status and using the transform for detection, the approach captures complex structure not visible from direct analysis of the data. When applied to model microarray data, the transform is shown to increase the quality of feature detection. On real microarray data, the transform offers higher sample enrichments and provides an alternative view of phenotypes not visible without the transform.