A modiefed parallel version of code imported from https://github.com/Bioconductor-mirror/ConsensusClusterPlus 1.11.1
Usage
findModules.consensusCluster(
d = NULL,
maxK = 100,
reps = 100,
pItem = 0.8,
pFeature = 1,
clusterAlg = "kmeans",
innerLinkage = "average",
distance = "pearson",
changeCDFArea = 0.001,
nbreaks = 20,
seed = 123456789.12345,
weightsItem = NULL,
weightsFeature = NULL,
corUse = "everything",
verbose = F
)Arguments
- d
Optional. A matrix where columns=items/samples and rows are features. For example, a gene expression matrix of genes in rows and microarrays in columns. OR ExpressionSet object. (Default = NULL)
- maxK
Optional. An integer value. maximum cluster number to evaluate. (Default = 100)
- reps
Optional. An integer value. number of subsamples. (Default = 100)
- pItem
Optional. A numerical value. proportion of items to sample. (Default = 0.8)
- pFeature
Optional. A numerical value. proportion of features to sample. (Default = 1)
- clusterAlg
Optional. A character value. cluster algorithm. "hc" heirarchical (hclust) or "km" for kmeans. (Default = "kmeans")
- innerLinkage
Optional. A heirarchical linkage method for subsampling. (Default = "average")
- distance
Optional. A character value. sample distance measures: "pearson","spearman", or "euclidean". (Default = "pearson")
- changeCDFArea
Optional. Minimum spline distance for seq(2,`maxK`, length.out = `nbreaks`) (Default = 0.001)
- nbreaks
Optional. Number of breaks to use in seq(2,`maxK`,length.out = `nbreaks`) this becomes the kGrid argument in run.consensus.cluster. (Default = 20)
- seed
Optional, A numerical value. Sets random seed for reproducible results. (Default = 123456789.12345)
- weightsItem
Optional. A numerical vector. weights to be used for sampling items. (Default = NULL)
- weightsFeature
Optional. AN umerical vector. weights to be used for sampling features. (Default = NULL)
- corUse
Optional. Use all cores avaiable. (Default = "Everything")
- verbose
Optional. A boolean when set to TRUE, prints messages to the screen to indicate progress. This is useful for large datasets.(Default = FALSE)