Consensus_Cluster_Analysis: Consensus_Cluster_Analysis

In FanqianYin/omicstoolkits: Integrated bioinformatics toolkits for quantitative omics data analysis

Description

Consensus clustering result of multi-clustering algorithms.

Usage

Consensus_Cluster_Analysis(
  Exp,
  k,
  methods = "ALL",
  dist.method = "euclidean",
  consensus.method = "average",
  scale = TRUE,
  saveplot = FALSE,
  plot.dir.suffix = NULL,
  ConsensusClusterPlus.reps = 100
)

clustering.methods()

Consensus_Cluster_conMethod(res.con, consensus.method)

Arguments

Exp	data.frame or matrix with rownames represent samples, colnames represent features like proteins or genes.
k	number of clusters
methods	Charactor vector. Default "ALL" will perform all 20 clustering methods.
dist.method	Method for calculating dissimilarity. Default "euclidean", can be any methods supported by dist()
consensus.method	Method used for the final consensus cluster by hieraichical clustering. Default are set to "average". Alternatively, "complete" may get a diferent view of result, which you should try. It Can be set as any method supported by hclust.
scale	Logical value, whether to scale your input data. Default: TRUE.
saveplot	under developing
plot.dir.suffix	under developing
ConsensusClusterPlus.reps	Integer value, used by consensusClusterPlus.
res.con	result of Consensus_Cluster_Analysis().

Details

Consensus clustering result of multi-clustering algorithms.

Value

Consensus_Cluster_Analysis returns res.con–one list contains all cluster results.

clustering.methods returns all supported clutersing algorithms.

Consensus_Cluster_conMethod returns one new res.con with the new consensus.method.

Examples

#Use ALL supported clustering methods
res.con <- Consensus_Cluster_Analysis(iris[-5], 3, methods = "ALL")
#Use a subset of clustering methods. Note: not avaliable now.
#Under development: res.con <- Consensus_Cluster_Analysis(iris[-5], 3, methods = c("kmeans", "pam"))

#plot result:
plot_consensus_cluster.heatmap(res.con)
plot_consensus_cluster.pca(res.con)

#list all supported clustering algorithms:
clustering.methods()
all.methods <- unlist(clustering.methods)

#Change consensus cluster method:
res.con <- Consensus_Cluster_conMethod(res.con, "complete")

FanqianYin/omicstoolkits documentation built on Jan. 3, 2022, 5:57 p.m.