New Method for Clustering Microarray Data
Author Information
Author(s): De Bin Riccardo, Risso Davide
Primary Institution: Department of Statistical Sciences, University of Padova, Padova, Italy
Hypothesis
Can a nonparametric density estimation approach improve clustering of microarray data?
Conclusion
The proposed algorithm is a simple and effective tool for the clustering of microarray data, in an unsupervised setting.
Supporting Evidence
- The algorithm shows promising results in both simulated and real data.
- Gene filtering improves the clustering performance significantly.
- The method outperforms traditional clustering techniques like k-means and Mclust.
Takeaway
This study introduces a new way to group similar gene expressions, which helps in understanding diseases like cancer better.
Methodology
The study uses a three-step procedure: gene filtering, dimensionality reduction, and clustering in reduced space using a nonparametric model-based approach.
Limitations
The method may still be affected by the high dimensionality of data and the choice of principal components.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website