Aggregated Data Ensemble Clustering - version c

Share:

Description

Function ADECc performs aggregated data ensemble clustering in which in every iteration the number of random samples taken is randomly set between m/2 and m-1 with m the total number of features. The number of features to sample can also be prespecified by the user. Further, each resulting dendrogram is cut numerous times into a different specific number of clusters.

Usage

1
2
3
ADECc(List, distmeasure = "tanimoto",normalize=FALSE,method=NULL,t = 10, 
r = NULL,nrclusters = seq(5, 25, 1), clust = "agnes", linkage = "ward",
alpha=0.625)

Arguments

List

A list of data matrices of the same type. It is assumed the rows are corresponding with the objects.

distmeasure

The distance measure to be used on the fused data matrix (character). Should be one of "tanimoto", "euclidean", "jaccard","hamming".

normalize

Logical. Indicates whether to normalize the distance matrices or not. This is recommended if different distance types are used. More details on normalization in Normalization.

method

A method of normalization. Should be one of "Quantile","Fisher-Yates", "standardize","Range" or any of the first letters of these names.

t

The number of iterations.

r

Optional. The number of features to take for the random sample.

nrclusters

A sequence of numbers of clusters to cut the dendrogram in.

clust

Choice of clustering function (character). Defaults to "agnes".

linkage

Choice of inter group dissimilarity (character). Defaults to "ward".

alpha

The parameter alpha to be used in the "flexible" linkage of the agnes function. Defaults to 0.625 and is only used if the linkage is set to "flexible"

Details

ADECc starts with the merging of the data matrices into one larger data matrix. Then, ensemble clustering is performed on the fused data. This comes down to repeatedly applying hierarchical clustering. A random sample of features is taken in each application. Further, variation is inserted by not splitting the dendrogram a single time into one specific number of clusters but multiple times and for a range of numbers of clusters. More information can be found in Fodeh et al. (2013).

Value

The returned value is a list with the following three elements.

AllData

Fused data matrix of the data matrices

S

The resulting co-association matrix

Clust

The resulting clustering

The value has class 'ADEC'. The Clust element will be of interest for further applications.

Note

For now, only hierarchical clustering with the agnes function is implemented.

Author(s)

Marijke Van Moerbeke

References

FODEH, J. S., BRANDT, C., LUONG, B. T., HADDAD, A., SCHULTZ, M., MURPHY, T., KRAUTHAMMER, M. (2013). Complementary Ensemble Clustering of Biomedical Data. J Biomed Inform. 46(3) pp.436-443.

See Also

ADEC,ADECa,ADECb

Examples

1
2
3
4
5
data(fingerprintMat)
data(targetMat)
L=list(fingerprintMat,targetMat)
MCF7_ADECc=ADECc(L,distmeasure="tanimoto",normalize=FALSE,method=NULL,t=10,r=NULL,
nrclusters=seq(5,25,1),clust="agnes",linkage="ward",alpha=0.625)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.