polyCluster: polyCluster pipeline

In syspremed/polyClustR: polyClustR: A pipeline to reconcile clustering solutions

Description

Runs the entire polyCluster pipeline.

Usage

polyCluster(filename, clusterAlg = c("hc", "pm", "km", "nmf"), maxK = 7,
  reps = 100, phenoFile = NULL, ref = "", nmfData = NULL,
  interactive = FALSE, seed = NULL)

Arguments

filename	Character string. File name of preprocessed and normalized expression values with genes in rows and samples in columns
clusterAlg	Character vector. Any combination of "hc" (hierarchical clustering), "pm" (partitioning around medoids), "km" (k-means) or "nmf" (nonnegative matrix factorization)
maxK	Integer > 2. The maximum number of clusters of samples to evaluate.
reps	Integer. The number of resampling (for "hc", "pm" and "km") or starting seed ("nmf") iterations.
phenoFile	Character string. File name of known phenotypes, with sample name in first column and phenotype in second column.
ref	Character string. Reference with which to name the output of the analysis.
nmfData	Character string. File path to the result of a previous NMF clustering by this function, in order to save time.
interactive	Logical. If FALSE, networks will be plotted with an automatic layout. If TRUE, a users can customise the layout through tkplot.
seed	Numeric. Value of seed to start clustering for reproducible results.

Value

Provides output from the whole pipeline in the current working directory.

Examples

library(polyClustR)

# Example breast dataset is provided with installation
exampleGE <- system.file('extdata', 'chin_breast_example_GE.txt', package = 'polyClustR')
exampleKnownSubtypes <- system.file('extdata', 'chin_breast_example_known_subtypes.txt', package = 'polyClustR')

polyCluster(exampleGE, clusterAlg = c('hc', 'km'), phenoFile = exampleKnownSubtypes, ref = 'test_run')
# Output is written to the current working directory.

syspremed/polyClustR documentation built on May 31, 2019, 12:49 a.m.