Evaluating Clustering Algorithms for Gene Expression Data
Author Information
Author(s): Susmita Datta, Somnath Datta
Primary Institution: University of Louisville
Hypothesis
Can we systematically evaluate clustering algorithms for gene expression data using functional classes?
Conclusion
Functional information from gene ontology databases can help select the best clustering algorithm for gene expression data.
Supporting Evidence
- Three clustering algorithms showed significant performance in producing biologically meaningful clusters.
- UPGMA with correlation produced the most homogeneous biological clusters.
- Fanny and Diana also performed well under certain conditions.
Takeaway
This study helps scientists choose the best way to group genes based on their activity, using information about what those genes do.
Methodology
The study introduces two performance measures, the Biological Homogeneity Index (BHI) and Biological Stability Index (BSI), to evaluate clustering algorithms on gene expression data.
Limitations
The study may not account for all possible clustering algorithms or biological contexts.
Participant Demographics
The study analyzed gene expression data from breast cancer patients and yeast.
Statistical Information
P-Value
.028
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website