Additional limitations of the clustering validation method figure of merit

ABSTRACT

Clustering analysis is an important exploratory tool that aids in the analysis and organization of genomic data. Each biological data set has different characteris, and the decision of which clustering method is appropriate and how many clusters are optimal on a dataset-by-dataset basis can be problematic. The Figure of Merit (FOM) is a quantitative clustering validation method designed to aid in these decisions. While FOM is useful, it does have limitations which must be considered when using it. This research shows that the FOM is biased toward Euclidean distance. Performing FOM analysis on clusters created by using Pearson's correlation coefficient as a similarity measure is shown to be non-optimal, and mathematically inadvisable. A new, correlation coefficient-biased version of the FOM has been developed, and preliminary results indicate that this new FOM is effectively biased toward clusters generated using the correlation coefficient.

REFERENCES

• 1: Copson, E. T. Metric Spaces. Cambridge University Press, London, 1968.
• 2: Eisen, M. B., Spellman, P. T., Brown, P. O. and Botstein, D. Cluster analysis and display of genome-wide expression patterns. Proc. Natl. Acad. Sei, 95, (1998), 14863--14868.
• 3: Electronic Source, Affymetrix, Inc. GeneChip Expression Analysis: Data Analysis Fundamentals, https://www.affymetrix.com/support/downloads/manuals/dat a_analysis_fundamentals_manual.pdf, Nov. 27, 2006.
• 4: Daxin Jiang , Chun Tang , Aidong Zhang, Cluster Analysis for Gene Expression Data: A Survey, IEEE Transactions on Knowledge and Data Engineering, v.16 n.11, p.1370-1386, November 2004 [doi>10.1109/TKDE.2004.68]
• 5: Olver, P. J. and Shakiban, C. Applied Linear Algebra. Pearson Prentice Hall, Upper Saddle River, NJ, 2006.
• 6: Pitts, C. G. C. Introduction to Metric Spaces. Oliver and Boyd, Edinburgh, 1972.
• 7: Peter Rousseeuw, Silhouettes: a graphical aid to the interpretation and validation of cluster analysis, Journal of Computational and Applied Mathematics, v.20 n.1, p.53-65, Nov. 1987 [doi>10.1016/0377-0427(87)90125-7]
• 8: Shamir, R. and Sharaa, R. Algorithmic approaches to clustering gene expression data. In Current Topic in Computational Biology. MIT Press, 2001.
• 9: Yeung, K. Y., Haynor, D. R. and Ruzzo, W. L. Validating Clustering for Gene Expression Data. Technical Report UW-CSE-00-01-01, University of Washington, Seattle, WA, 2000.
• 10: Yeung, K. Y., Haynor, D. R. and Ruzzo, W. L. Validating clustering for gene expression data. Bioinformatics, 17 (4), (2001), 309--318.
• 11: Ka Yee Yeung , Walter L. Ruzzo, Cluster analysis of gene expression data, University of Washington, 2001
• 12: Yeung, K. Y., Medvedovic, M. and Bumgarner, R. E. Clustering gene-expression data with repeated measurements. Genome Biology, 4 (5) R34, Epub 2003 Apr 25, http://expression.washington.edu/publications/kayee/yeunggb2003/.

INDEX TERMS

Primary Classification:
  J. Computer Applications
  J.3 LIFE AND MEDICAL SCIENCES
      Subjects: Biology and genetics

General Terms:
Algorithms, Measurement

Keywords:
Euclidean distance, FOM, K-means, Pearson's correlation coefficient, cluster analysis, cluster validation method, clustering, correlation coefficient, distance metric, figure of merit, gene expression, similarity measure