compareARI {coseq}R Documentation

Pairwise comparisons of ARI values among a set of clustering partitions

Description

Provides the adjusted rand index (ARI) between pairs of clustering paritions.

Usage

compareARI(object, ...)

## S4 method for signature 'coseqResults'
compareARI(object, K = NULL, parallel = FALSE,
  BPPARAM = bpparam(), plot = TRUE, ...)

## S4 method for signature 'matrix'
compareARI(object, parallel = FALSE, BPPARAM = bpparam(),
  plot = TRUE, ...)

## S4 method for signature 'data.frame'
compareARI(object, parallel = FALSE,
  BPPARAM = bpparam(), plot = TRUE, ...)

Arguments

object

Object of class coseqResults or RangedSummarizedExperiment, or alternatively a n x M data.frame or matrix containing the clustering partitions for M different models

...

Additional optional parameters for corrplot

K

If NULL, pairwise ARI values will be calculated among every model in object x. Otherwise, K provides a vector of cluster numbers identifying a subset of models in x.

parallel

If FALSE, no parallelization. If TRUE, parallel execution using BiocParallel (see next argument BPPARAM). Note that parallelization is unlikely to be helpful unless the number of observations n in the clustering partitions or the number of models M are very large.

BPPARAM

Optional parameter object passed internally to bplapply when parallel=TRUE. If not specified, the parameters last registered with register will be used.

plot

If TRUE, provide a heatmap using corrplot to visualize the calculated pairwise ARI values.

Value

Matrix of adjusted rand index values calculated between each pair of models.

Author(s)

Andrea Rau

Examples

## Simulate toy data, n = 300 observations
set.seed(12345)
countmat <- matrix(runif(300*4, min=0, max=500), nrow=300, ncol=4)
countmat <- countmat[which(rowSums(countmat) > 0),]
conds <- rep(c("A","B","C","D"), each=2)

## Run the Normal mixture model for K = 2,3,4
run_arcsin <- coseq(object=countmat, K=2:4, iter=5, transformation="arcsin",
                    model="Normal")
run_arcsin

## Plot and summarize results
plot(run_arcsin)
summary(run_arcsin)

## Compare ARI values for all models (no plot generated here)
ARI <- compareARI(run_arcsin, plot=FALSE)

## Compare ICL values for models with arcsin and logit transformations
run_logit <- coseq(object=countmat, K=2:4, iter=5, transformation="logit",
                   model="Normal")
compareICL(list(run_arcsin, run_logit))

## Use accessor functions to explore results
clusters(run_arcsin)
likelihood(run_arcsin)
nbCluster(run_arcsin)
ICL(run_arcsin)

## Examine transformed counts and profiles used for graphing
tcounts(run_arcsin)
profiles(run_arcsin)

## Run the K-means algorithm for logclr profiles for K = 2,..., 20
run_kmeans <- coseq(object=countmat, K=2:20, transformation="logclr",
                    model="kmeans")
run_kmeans
[Package coseq version 1.4.0 Index]