Up: Component summary Component

FlowMeans

Clusters rows in CSV files. The component is a wrapper for the R-package flowMeans. Clustering is based on k-means clustering but spherichal clusters are merged to find non-spherichal clusters.

Version	1.0
Bundle	flowand
Categories	FlowCytometry
Authors	Erkka Valo (erkka.valo@helsinki.fi)
Requires	flowMeans (R-package)
Source files	component.xml FlowMeans.r
Usage	Example with default values FlowMeans( //inList = CSVList, //in = CSV, channelsToCluster = "*", clusterIDColName = "cluster", iterMax = 50, mahalanobis = true, maxN = -1, nStart = 10, numC = -1, standardize = true )

Inputs

Name	Type	Mandatory	Description
inList	CSVList	Optional	CSV files to cluster.
in	Array<CSV>	Optional	CSV files to cluster.

Outputs

Name	Type	Description
clustersList	CSVList	Clustered CSV files.
clusters	Array<CSV>	Clustered CSV files.
report	Latex	Report of the clustering.

Parameters

Name	Type	Default	Description
channelsToCluster	string	"*"	The columns used to cluster the data in a comma-separated list. Default value "*" uses all columns. If all the columns are not present in a input csv file, the clustering is done with the subset of columns that are present.
clusterIDColName	string	"cluster"	The name of the column in the `clusters` output which contains the cluster IDs of the rows.
iterMax	int	50	The maximum number of iterations allowed.
mahalanobis	boolean	true	If TRUE (default) mahalanobis distance will be used. Otherwised, euclidean distance will be used.
maxN	int	-1	Maximum number of clusters. If set to a negative value (default) the value will be estimated automatically from the data.
nStart	int	10	The number of random sets used for initialization.
numC	int	-1	Number of clusters. If set to a negative value (default), the value will be estimated automatically.
standardize	boolean	true	If TRUE, the data will be transformed to the [0,1] interval for clustering. The transformed values are not put into the output files.

Generated 2019-02-08 07:42:08 by Anduril 2.0.0