clusterTimeProfiles: Clustering of time profiles.

In MarenS2/pwOmics: Pathway-based data integration of omics data

Description

Soft clustering of time series data with Mfuzz R package [1]. Filtering of genes with low expression changes possible via min.std parameter. Expression values are standardized and undergo fuzzy c-means clustering based on minimization of weighted square error function (see [2]). Fuzzifier parameter m is estimated via mestimate function of [1] based on a relation proposed by Schwaemmle and Jansen [3]. The optimal number of clusters is determined via the minimum distance between cluster centroid using Dmin function of [3]. Be aware that the cluster number determination might be difficult especially for short time series measurements.

Usage

clusterTimeProfiles(dynConsensusNet, min.std = 0, ncenters = 12)

Arguments

dynConsensusNet	result of dynamic analysis: inferred net generated by consDynamicNet function.
min.std	threshold parameter to exclude genes with a low standard deviation. All genes with an expression smaller than min.std will be excluded from clustering. Default value is 0.
ncenters	integer specifying the maximum number of centers which should be tested in minimum distance between cluster centroid test; this number is used as initial number to determine the data-specific maximal cluster number based on number of distinct data points.

Value

output dataframe of mfuzz function.

References

1. L. Kumar and M. Futschik, Mfuzz: a software package for soft clustering of microarray data, Bioinformation, 2(1) 5-7, 2007.

2. Bezdak JC, Pattern Recognition with Fuzzy Objective Function Algorithms, Plenum Press, New York, 1981.

3. Schwaemmle and Jensen, Bioinformatics, Vol. 26 (22), 2841-2848, 2010.

Examples

## Not run: 
data(OmicsExampleData)
data_omics = readOmics(tp_prots = c(0.25, 1, 4, 8, 13, 18, 24), 
tp_genes = c(1, 4, 8, 13, 18, 24), OmicsExampleData,
PWdatabase = c("biocarta", "kegg", "nci", "reactome"), 
TFtargetdatabase = c("userspec"))
data_omics = readPhosphodata(data_omics, 
phosphoreg = system.file("extdata", "phospho_reg_table.txt", 
package = "pwOmics.newupdown")) 
data_omics = readTFdata(data_omics, 
TF_target_path = system.file("extdata", "TF_targets.txt", 
package = "pwOmics.newupdown"))
data_omics_plus = readPWdata(data_omics,  
loadgenelists = system.file("extdata/Genelists", package = "pwOmics.newupdown")) 

## End(Not run)
## Not run: 
data_omics_plus = identifyPR(data_omics_plus)
setwd(system.file("extdata/Genelists", package = "pwOmics.newupdown"))
data_omics = identifyPWs(data_omics_plus)
data_omics = identifyTFs(data_omics)
data_omics = identifyRsofTFs(data_omics, 
noTFs_inPW = 1, order_neighbors = 10)
data_omics = identifyPWTFTGs(data_omics)
statConsNet = staticConsensusNet(data_omics)
consDynNet = consDynamicNet(data_omics, statConsNet)
clusterTimeProfiles(consDynNet)

## End(Not run)

MarenS2/pwOmics documentation built on May 17, 2019, 9:10 a.m.