SPADE.FCSToTree: Cluster and build minimum spanning tree from data in FCS...

In spade: SPADE -- An analysis and visualization tool for Flow Cytometry

Description

Hierarchically cluster observations in a set of FCS files and build a minimum spanning tree connecting those clusters.

Usage

SPADE.FCSToTree(infilenames, outfilename, graphfilename, clusterfilename,
		  cols = NULL, k = 200, arcsinh_cofactor=NULL, transforms=flowCore::arcsinhTransform(a=0, b=0.2), 
		  desired_samples = 50000, comp=TRUE)

Arguments

infilenames	Vector of FCS file names that should be used as input
outfilename	Name of FCS file to write subset of cells used for clustering along with their cluster assignment
graphfilename	Name of file to write gml graph description
clusterfilename	Name of file to write table of cluster centers
cols	Usually a vector of strings specifying the columns to be used in the density calculation, e.g., c("(Cd110)D","(Cs111)D"). Strings will be matched against the parameter names extracted from the FCS file. The default=NULL will use all parameters.
k	Desired number of clusters. Algorithm might create between [k/2,3k/2] clusters.
arcsinh_cofactor	DEPRECATED. Cofactor used in arcsinh transform asinh(data/arcsinh_cofactor) of data.
transforms	Transform object to apply to data. A single transform object will be applied to all channels. To apply different transforms to specific channels use a named vector of transform objects (where names are parameter names).
desired_samples	Desired number of samples to be used in clustering. Usually leave at default.
comp	Apply compensation matrix if present in SPILL or SPILLOVER keywords

Value

None.

Note

Underlying implementations have been parallelized with OpenMP. Set OMP_NUM_THREADS in environment to control the number of threads used. Implementation can be very memory intensive.

Author(s)

Michael Linderman

See Also

SPADE.downsampleFCS

Examples

	## Load two-parameters sample data included in package
	#data_file_path = paste(installed.packages()["spade","LibPath"],"spade","extdata","SimulatedRawData.fcs",sep=.Platform$file.sep)

	#output_dir <- tempdir()
	#
	## Compute and annotate FCS file with density
	#density_file_path <- paste(output_dir,.Platform$file.sep,basename(data_file_path),".density.fcs",sep="")
	#SPADE.addDensityToFCS(data_file_path, density_file_path, cols=c("marker1","marker2"))

	## Downsample FCS file based on density
	#downsample_file_path <- paste(output_dir,.Platform$file.sep,basename(data_file_path),".density.fcs",sep="")
	#SPADE.downsampleFCS(density_file_path, downsample_file_path)

	## Create tree from downsampled FCS file
	#cells_file_path <- paste(output_dir,"clusters.fcs",sep="")
	#clust_file_path <- paste(output_dir,"clusters.table",sep="")
	#graph_file_path <- paste(output_dir,"mst.gml",sep="")
	#SPADE.FCSToTree(downsample_file_path, cells_file_path, graph_file_path, clust_file_path, cols=c("marker1","marker2"))

spade documentation built on Oct. 5, 2016, 4:37 a.m.