medv function

Clustering Method of Medvedovic

Based on a posterior similarity matrix of a sample of clusterings medv obtains a clustering by using 1-psm as distance matrix for hierarchical clustering with complete linkage. The dendrogram is cut at a value h close to 1.

medv(psm, h=0.99)

Arguments

• psm: a posterior similarity matrix, usually obtained from a call to comp.psm.
• h: The height at which the dendrogram is cut.

Returns

vector of cluster memberships.

References

Medvedovic, M. Yeung, K. and Bumgarner, R. (2004) Bayesian mixture model based clustering of replicated microarray data, Bioinformatics, 20 , 1222-1232.

Author(s)

Arno Fritsch, arno.fritsch@tu-dortmund.de

See Also

comp.psm for computing posterior similarity matrix, maxpear, minbinder, relabel

for other possibilities for processing a sample of clusterings.

Examples

data(cls.draw1.5) 
# sample of 500 clusterings from a Bayesian cluster model 
tru.class <- rep(1:8,each=50) 
# the true grouping of the observations
psm1.5 <- comp.psm(cls.draw1.5)
medv1.5 <- medv(psm1.5)
table(medv1.5, tru.class)