Related Experiment Video
Updated: Jun 26, 2025
05:22
Analyzing Multifactorial RNA-Seq Experiments with DiCoExpress
Published on: July 29, 2022
3.5K
Clustering algorithm based on DINNSM and its application in gene expression data analysis
Zongjin Li1, Changxin Song2, Jiyu Yang3
1Department of Computer, Qinghai Normal University, Xining, China.
Summary
A new method, Dual-Index Nearest Neighbor Similarity Measure (DINNSM), improves gene co-expression module identification. DINNSM offers more accurate biological insights than traditional similarity measures.
Area of Science:
- Bioinformatics
- Computational Biology
- Systems Biology
Background:
- Effective similarity measurement is vital for biologically meaningful clustering.
- Existing methods struggle with the complexity and interactions in biological systems.
Purpose of the Study:
- To develop a novel similarity measurement method for enhanced gene module discovery.
- To improve the biological relevance of gene expression data clustering.
Main Methods:
- Proposed the Dual-Index Nearest Neighbor Similarity Measure (DINNSM) algorithm.
- Calculated gene similarity using Pearson or Spearman correlation.
- Constructed a nearest-neighbor table and reconstructed the similarity matrix.
Main Results:
- DINNSM outperformed five common similarity measures on five gene expression datasets.
- Clustering results using DINNSM showed superior performance.
- Demonstrated improved accuracy in identifying gene co-expression modules.
Conclusions:
- DINNSM provides more accurate biological insights into gene interactions.
- Facilitates the identification of more biologically relevant gene co-expression modules.
