Biostatistics Advance Access published online on April 12, 2006
Biostatistics, doi:10.1093/biostatistics/kxj029
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1 Department of Statistics, Stanford University, Stanford, CA 94305-4065, USA
* To whom correspondence should be addressed. In many microarray studies, a cluster defined on one dataset is sought in an independent dataset. If the cluster is found in the new dataset, the cluster is said to be reproducible and may be biologically significant. Classifying a new datum to a previously defined cluster can be seen as predicting which of the previously defined clusters is most similar to the new datum. If the new data classified to a cluster are similar, molecularly or clinically, to the data already present in the cluster, then the cluster is reproducible and the corresponding prediction accuracy is high. Here we take advantage of the connection between reproducibility and prediction accuracy to develop a validation procedure for clusters found in datasets independent of the one in which they were characterized. We define a cluster quality measure called the in-group proportion (IGP) and introduce a general procedure for individually validating clusters. Using simulations and real breast cancer datasets, the in-group proportion is compared to four other popular cluster quality measures (homogeneity score, separation score, silhouette width, and WADP score). Moreover, simulations and the real breast cancer datasets are used to compare the four versions of the validation procedure which all use the in-group proportion, but differ in the way in which the null distributions are generated. We find the in-group proportion is the best measure of prediction accuracy, and one version of the validation procedure is the more widely applicable than the other three. An implementation of this algorithm is in a package called clusterRepro available through The Comprehensive R Archive Network (http://cran.r-project.org).
Received July 28, 2005
Revised February 28, 2006
Accepted March 8, 2006
Article
Are clusters found in one dataset present in another dataset?
Amy V. kapp 1 *
and
Robert Tibshirani 2
2 Department of Health Research & Policy and Department of Statistics, Stanford, University, Stanford, CA
Amy V. kapp, E-mail: akapp{at}stanford.edu
Abstract 
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. S. Ross, C. Hatzis, W. F. Symmans, L. Pusztai, and G. N. Hortobagyi Commercialized Multigene Predictors of Clinical Outcome for Breast Cancer Oncologist, May 1, 2008; 13(5): 477 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Lusa, L. M. McShane, J. F. Reid, L. De Cecco, F. Ambrogi, E. Biganzoli, M. Gariboldi, and M. A. Pierotti Challenges in Projecting Clustering Results Across Gene Expression Profiling Datasets J Natl Cancer Inst, November 21, 2007; 99(22): 1715 - 1723. [Abstract] [Full Text] [PDF]
|
|