Related Experiment Video
Updated: May 15, 2026
09:40
Characterization of Functionally Associated miRNAs in Glioblastoma and their Engineering into Artificial Clusters for Gene Therapy
Published on: October 4, 2019
Subtyping glioblastoma by combining miRNA and mRNA expression data using compressed sensing-based approach
Wenlong Tang1, Junbo Duan, Ji-Gang Zhang
1Department of Biomedical Engineering, Tulane University, New Orleans, LA, USA. wyp@tulane.edu.
EURASIP Journal on Bioinformatics & Systems Biology
|January 15, 2013
Summary
Combining mRNA and miRNA genomic data significantly improves glioblastoma multiforme (GBM) subtype classification accuracy. This integrated approach enhances diagnostic potential and biomarker discovery for personalized cancer treatment.
Area of Science:
- Genomics
- Bioinformatics
- Cancer Research
Background:
- Accurate disease subtyping is crucial for personalized medicine.
- Single genomic data types often yield insufficient accuracy for complex diseases like glioblastoma multiforme (GBM).
- Advances in genomic technologies provide multiple data types (e.g., mRNA, miRNA) for analysis.
Purpose of the Study:
- To investigate if combining multiple genomic data types improves GBM subtype classification accuracy.
- To develop and apply a novel multi-class compressed sensing-based detector (MCSD) for GBM subtyping.
Main Methods:
- Utilized genome-wide mRNA and miRNA expression data from The Cancer Genome Atlas (TCGA) project.
- Developed a multi-class compressed sensing-based detector (MCSD) for classification.
- Trained MCSD on TCGA data and validated on an independent GBM patient cohort.
Main Results:
- Classification accuracy reached 69.1% with miRNA data and 52.7% with mRNA data alone.
- Combining both mRNA and miRNA expression data achieved a significantly higher accuracy of 90.9%.
- Identified potential biomarkers through integrated analysis, with some validated in existing literature.
Conclusions:
- Integrating multiple genomic data types substantially enhances GBM subtype classification accuracy.
- The proposed MCSD method demonstrates effectiveness in multi-modal genomic data analysis.
- Combined genomic analysis aids in identifying diagnostic biomarkers and advancing personalized cancer therapies.
