Related Experiment Video
Updated: Jul 8, 2026
07:35
Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
Signature Evaluation Tool (SET): a Java-based tool to evaluate and visualize the sample discrimination abilities of
Chih-Hung Jen1, Tsun-Po Yang, Chien-Yi Tung
1Microarray & Gene Expression Analysis Core Facility, VGH National Yang-Ming University Genome Research Center, Taipei, Taiwan. chjen2@ym.edu.tw
BMC Bioinformatics
|January 29, 2008
Summary
A new tool, Signature Evaluation Tool (SET), helps researchers identify smaller, more effective gene expression signatures for cancer research. It evaluates discrimination power and aids in clinical decision-making by refining gene sets.
Area of Science:
- Bioinformatics
- Computational Biology
- Genomics
Background:
- Gene expression signatures are crucial for cancer research but often contain too many genes for clinical use.
- Experimental validation of large gene sets is limited by resource availability.
- Clinical researchers need tools to assess the discriminatory power of candidate gene signatures.
Purpose of the Study:
- To introduce the Signature Evaluation Tool (SET) for evaluating gene expression signatures.
- To provide a flexible platform for refining gene signatures for clinical application.
- To aid in the identification of clinically relevant gene markers.
Main Methods:
- Developed a Java-based tool, SET, utilizing Golub's weighted voting algorithm.
- Incorporated visual presentation of prediction strength for individual array samples.
- Enabled iterative selection/de-selection of genes to re-evaluate signature discrimination power.
Main Results:
- SET can rapidly reduce large gene signatures to a manageable size.
- The tool allows re-evaluation and adjustment of signature discrimination power for user-defined or external signatures.
- SET can assess signature classification capability across multiple microarray datasets.
- Visualizing prediction strength helps re-evaluate uncertain or mis-grouped samples.
Conclusions:
- SET evaluates and visualizes the sample-discrimination ability of gene expression signatures.
- The tool functions as a filtration system for signature identification, bridging clinical analysis and prediction.
- SET's simplicity and flexibility make it valuable for marker identification in clinical research.
