NetPathMiner (version 1.8.0)

pathCluster: 3M Markov mixture model for clustering pathways

Description

3M Markov mixture model for clustering pathways

Usage

pathCluster(ybinpaths, M, iter = 1000)

Arguments

ybinpaths

The training paths computed by pathsToBinary.

M

The number of clusters.

iter

The maximum number of EM iterations.

Value

A list with the following items:

h

The posterior probabilities that each path belongs to each cluster.

labels

The cluster membership labels.

theta

The probabilities of each gene for each cluster.

proportions

The mixing proportions of each path.

likelihood

The likelihood convergence history.

params

The specific parameters used.

References

Mamitsuka, H., Okuno, Y., and Yamaguchi, A. 2003. Mining biologically active patterns in metabolic pathways using microarray expression profiles. SIGKDD Explor. News l. 5, 2 (Dec. 2003), 113-121.

See Also

Other Path clustering & classification methods: pathClassifier; pathsToBinary; plotClassifierROC; plotClusterMatrix, plotClusterProbs, plotClusters; plotPathClassifier; plotPathCluster; predictPathClassifier; predictPathCluster

Examples

## Prepare a weighted reaction network.
	## Conver a metabolic network to a reaction network.
 data(ex_sbml) # bipartite metabolic network of Carbohydrate metabolism.
 rgraph <- makeReactionNetwork(ex_sbml, simplify=TRUE)

	## Assign edge weights based on Affymetrix attributes and microarray dataset.
 # Calculate Pearson's correlation.
	data(ex_microarray)	# Part of ALL dataset.
	rgraph <- assignEdgeWeights(microarray = ex_microarray, graph = rgraph,
		weight.method = "cor", use.attr="miriam.uniprot", bootstrap = FALSE)

	## Get ranked paths using probabilistic shortest paths.
 ranked.p <- pathRanker(rgraph, method="prob.shortest.path",
					K=20, minPathSize=8)

	## Convert paths to binary matrix.
	ybinpaths <- pathsToBinary(ranked.p)
	p.cluster <- pathCluster(ybinpaths, M=2)
	plotClusters(ybinpaths, p.cluster)