A novel approach to the clustering of microarray data via nonparametric density estimation
2011

New Method for Clustering Microarray Data

publication Evidence: moderate

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)

10.1186/1471-2105-12-49

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication