clusterTimeseries2: Cluster a processed time-series with k-means or flowClust

In raim/segmenTools: Tools for Functional Genome Exploration

Cluster a processed time-series with k-means or flowClust

Description

A wrapper for clustering a time-series object tset provided by processTimeseries, where specifically the DFT of a time-series and requested data transformation were calculated. The clustering is performed on the tset$dat matrix by kmeans or model-based by packages flowClust & flowMerge.

Usage

clusterTimeseries2(
  tset,
  K = 16,
  method = "flowClust",
  selected,
  parameters = c(iter.max = 1e+05, nstart = 100, B = 500, tol = 1e-05, lambda = 1, nu =
    4, nu.est = 0, trans = 1, merge = FALSE, randomStart = 0),
  nui.thresh = -Inf,
  ncpu = 1,
  verb = 1,
  ...
)

Arguments

tset	a timeseries processed by processTimeseries
K	selected cluster numbers, the argument centers of kmeans
method	string specifying the clustering algorithm to use, currently "kmeans" and "flowClust" are implemented
selected	a pre-selected cluster number which is then used as a start clustering for flowMerge (if option merge==TRUE)
parameters	named vector of parameters for clustering algorithms, currently ALL required parameters MUST be specified
nui.thresh	threshold correlation of a data point to a cluster center; if below the data point will be added to nuissance cluster 0
ncpu	number of cores available for parallel mode of flowClust. NOTE: parallel mode of flowClust is often non-functional. Alternatively, you can set options(mc.cores=ncpu) directly.
verb	level of verbosity, 0: no output, 1: progress messages
...	further parameters to flowClust or kmeans

Details

This function attempts to combine the previous separate clustering wrappers from package segmenTier, flowclusterTimeseries and k-mean's based clusterTimeseries, the latter of which is used for the segmenTier algorithm. Please see the corresponding help files for details on the clustering parameters in argument parameters.

raim/segmenTools documentation built on April 25, 2024, 10:14 a.m.