Related Experiment Video
Updated: May 4, 2026
07:35
Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
7.0K
Multi-resolution-test for consistent phenotype discrimination and biomarker discovery in translational bioinformatics
Henry Han1, Xiao-Li Li, See-Kiong Ng
1Department of Computer and Information Sciences, Fordham University, New York 48105, USA , Quantitative Proteomics Center, Columbia University, New York 10027, USA.
Journal of Bioinformatics and Computational Biology
|December 31, 2013
Summary
A new Multi-Resolution-Test (MRT-test) algorithm accurately distinguishes complex disease phenotypes using omics data. This method identifies subtle gene expression patterns for robust clinical diagnosis and biomarker discovery.
Area of Science:
- Translational bioinformatics
- Systems biology
- Molecular diagnostics
Background:
- Accurate discrimination of complex disease phenotypes from omics data is challenging for clinical diagnosis.
- Existing methods struggle to consistently identify subtle gene and protein expression patterns indicative of different clinical conditions.
Purpose of the Study:
- To introduce a novel feature selection algorithm, Multi-Resolution-Test (MRT-test), for accurate and consistent phenotype discrimination in omics data.
- To develop a method for effective biomarker discovery enabling linear separation of high-dimensional data.
- To create a network marker synthesis (NMS) algorithm for deciphering molecular mechanisms of tumorigenesis.
Main Methods:
- Developed the Multi-Resolution-Test (MRT-test) feature selection algorithm.
- Applied MRT-test with state-of-the-art classifiers for complex disease diagnosis.
- Designed the network marker synthesis (NMS) algorithm using seed biomarkers.
Main Results:
- MRT-test achieved significantly accurate and consistent phenotype discrimination across various omics data.
- The algorithm effectively captures subtle data behaviors crucial for clinical-level diagnosis.
- Exceptional diagnostic results were obtained, demonstrating MRT-test's advantage in molecular diagnostics.
- MRT-test based diagnosis generated consistent and robust clinical-level phenotype separation for diverse diseases.
- The NMS algorithm provided biologically meaningful insights into the genetic basis of complex diseases.
Conclusions:
- MRT-test is a powerful tool for accurate and consistent molecular diagnostics in complex diseases.
- The combination of MRT-test and NMS offers a systems-level approach to understanding disease mechanisms.
- This methodology holds significant promise for advancing translational research and clinical applications.
