Related Experiment Video
Updated: Jun 7, 2026
07:35
Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
Published on: October 11, 2018
Multi-dimensional discovery of biomarker and phenotype complexes
Philip R O Payne1, Kun Huang, Kristin Keen-Circle
1Department of Biomedical Informatics, The Ohio State University, 3190 Graves Hall, 333 West 10th Avenue, 43210, Columbus, Ohio, USA. philip.payne@osumc.edu
BMC Bioinformatics
|November 4, 2010
Summary
Integrating gene networks, clinical features, and knowledge creates multi-dimensional disease models. This approach facilitates the discovery of novel biomarker-phenotype complexes for personalized medicine.
Area of Science:
- Biomedical Informatics
- Translational Research
- Systems Biology
Background:
- The need for integrated bio-molecular and phenotypic networks in personalized healthcare is growing.
- Existing methods for network development lack integrative approaches for systems-level disease modeling.
- Bridging gene networks, clinical features, and ontologies remains a significant challenge.
Purpose of the Study:
- To address the gap in integrative network approaches for disease modeling.
- To develop and demonstrate multi-modal methods for network induction.
- To identify significant biomarker-phenotype complexes across multiple granularity levels.
Main Methods:
- Utilized multi-modal approaches for network induction.
- Integrated orthogonal networks of genes, clinical features, and conceptual knowledge.
- Leveraged a large-scale data repository from the Chronic Lymphocytic Leukemia Research Consortium.
Main Results:
- Demonstrated computational tractability in linking diverse biological networks.
- Created multi-dimensional models of interrelated biomarkers and phenotypes.
- Identified systems-level models capable of supporting hypothesis discovery and testing.
Conclusions:
- Integrated network models are computationally feasible and valuable for research.
- Systems-level models can uncover novel, knowledge-anchored biomarker-phenotype associations.
- Proposed a conceptual model to guide the cross-linkage of network analysis results.
