dist2hclust: Hierarchially-cluster distance matrix

In NMikolajewicz/scMiko: scRNAseq analysis functions. Developed using Seurat framework.

Hierarchially-cluster distance matrix

Description

Hierarchially-cluster distance matrix using flashClust package.

Usage

dist2hclust(d.mat, method = "average", ...)

Arguments

d.mat	distance matrix
method	hierarchial-clustering method. Default is "average". See flashClust for options.
...	Additional arguments passed to flashClust flashClust package

Value

hclust object

Author(s)

Nicholas Mikolajewicz

See Also

flashClust

Examples

# Get expression matrix
which.data <- "scale"

# variable gene only matrix
use.var <- T
if (use.var){
  exp.mat <- getExpressionMatrix(so.query, only.variable = use.var, which.data = which.data, use.additional.genes = NA)
} else {
  exp.mat <- exp.mat.complete
}


# transpose expressio matrix (genes are columns)
t.exp.mat <- t(exp.mat)
datExpr <- as.matrix(t.exp.mat)
SubGeneNames=colnames(datExpr)

# capture output used to hide undesired print statement
print2hide <- capture.output(allowWGCNAThreads())

# transform matrix if necessary
if (min(datExpr) < 0) {
  datExpr.noz <- datExpr + abs(min(datExpr))
} else {
  datExpr.noz <- datExpr
}

# run WGCNA
output.all <- runWGCNA(datExpr.noz, cor.metric = "rho_p", soft.power = 2, use.TOM = T)

# unpack output
s.mat <- output.all[["s.mat"]] # similar matrix
a.mat <- output.all[["a.mat"]] # adjacency matrix
w.mat <- output.all[["w.mat"]] # topological overlap matix
d.mat <- output.all[["d.mat"]] # disimilarity matix

geneTree <- dist2hclust(d.mat)

NMikolajewicz/scMiko documentation built on June 28, 2023, 1:41 p.m.