Class prediction in microarray studies based on activation of pathways

Authors:
Henryk Maciejewski
Affiliations:
Institute of Computer Engineering, Control and Robotics, Wroclaw University of Technology, Wrocław, Poland
Venue:
HAIS'11 Proceedings of the 6th international conference on Hybrid artificial intelligent systems - Volume Part I
Year:
2011

Citing 6
Cited 1

A global test for groups of genes: testing association with a clinical outcome

Bioinformatics
Analyzing gene expression data in terms of gene sets

Bioinformatics
Quality of Feature Selection Based on Microarray Gene Expression Data

ICCS '08 Proceedings of the 8th international conference on Computational Science, Part III
A First Study on the Use of Coevolutionary Algorithms for Instance and Feature Selection

HAIS '09 Proceedings of the 4th International Conference on Hybrid Artificial Intelligence Systems
Model Instability in Microarray Gene Expression Class Prediction Studies

Computer Aided Systems Theory - EUROCAST 2009
Designing fusers on the basis of discriminants – evolutionary and neural methods of training

HAIS'10 Proceedings of the 5th international conference on Hybrid Artificial Intelligence Systems - Volume Part I

Feature selection based on activation of signaling pathways applied for classification of samples in microarray studies

ICAISC'12 Proceedings of the 11th international conference on Artificial Intelligence and Soft Computing - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a novel approach to building sample classifiers based on microarray gene expression studies. This approach differs from standard methods in the way features are selected. Standard methods focus on features (genes) with most differential expression between classes of samples compared, while the proposed approach takes into account apriori domain knowledge of relationships between features, available e.g., in the form of pathway or gene-ontology databases. Features for classification are then selected on the basis of activation of pathways (gene sets) rather than mutually unrelated genes with very high individual predictive power. Performance of the proposed method is illustrated on the basis of sample microarray studies.