Related Experiment Video
Updated: Jul 7, 2026
07:30
Optimization for Sequencing and Analysis of Degraded FFPE-RNA Samples
Published on: June 8, 2020
Proposal of new gene filtering method, BagPART, for gene expression analysis with small sample
Takashi Kawamura1, Hiro Takahashi, Hiroyuki Honda
1Department of Biotechnology, School of Engineering, Nagoya University, Nagoya, Japan.
Journal of Bioscience and Bioengineering
|February 26, 2008
Summary
The BagPART filtering method enhances gene expression analysis for small sample datasets by accurately selecting important genes. This novel approach improves binary classification accuracy, outperforming traditional methods.
Area of Science:
- Bioinformatics
- Computational Biology
- Machine Learning in Genomics
Background:
- Gene expression analysis often faces challenges with limited sample sizes compared to the number of genes.
- Small sample datasets are common in genomic studies, complicating accurate analysis.
Purpose of the Study:
- To introduce a novel gene selection and binary classification method for small sample datasets.
- To improve the accuracy of gene expression analysis in scenarios with limited samples.
Main Methods:
- Developed the BagPART filtering method, combining Bagging with Projective Adaptive Resonance Theory (PART).
- Utilized PART for effective gene selection.
- Applied the method to binary classification tasks.
Main Results:
- The BagPART method demonstrated significantly higher accuracy (p<10(-10)) in binary classification compared to conventional methods.
- The method is particularly effective for datasets with a small sample size relative to the number of genes.
Conclusions:
- The BagPART filtering method offers a robust solution for gene expression analysis with small sample sizes.
- This approach enhances the reliability and accuracy of identifying important genes and performing classification in genomics.
