estimateSignatures: Estimate number of signatures based on cophenetic correlation...

In PoisonAlien/maftools: Summarize, Analyze and Visualize MAF Files

Estimate number of signatures based on cophenetic correlation metric

Description

Estimate number of signatures based on cophenetic correlation metric

Usage

estimateSignatures(
  mat,
  nMin = 2,
  nTry = 6,
  nrun = 10,
  parallel = 4,
  pConstant = NULL,
  verbose = TRUE,
  plotBestFitRes = FALSE
)

Arguments

mat	Input matrix of diemnsion nx96 generated by trinucleotideMatrix
nMin	Minimum number of signatures to try. Default 2.
nTry	Maximum number of signatures to try. Default 6.
nrun	numeric giving the number of run to perform for each value in range. Default 5
parallel	Default 4. Number of cores to use.
pConstant	A small positive value to add to the matrix. Use it ONLY if the functions throws an non-conformable arrays error
verbose	Default TRUE
plotBestFitRes	plots consensus heatmap for range of values tried. Default FALSE

Details

This function decomposes a non-negative matrix into n signatures. Extracted signatures are compared against 30 experimentally validated signatures by calculating cosine similarity. See http://cancer.sanger.ac.uk/cosmic/signatures for details.

Value

a list with NMF.rank object and summary stats.

See Also

plotCophenetic extractSignatures trinucleotideMatrix

Examples

## Not run: 
laml.maf <- system.file("extdata", "tcga_laml.maf.gz", package = "maftools")
laml <- read.maf(maf = laml.maf)
laml.tnm <- trinucleotideMatrix(maf = laml, ref_genome = 'BSgenome.Hsapiens.UCSC.hg19', prefix = 'chr',
add = TRUE, useSyn = TRUE)
library("NMF")
laml.sign <- estimateSignatures(mat = laml.tnm, plotBestFitRes = FALSE, nMin = 2, nTry = 3, nrun = 2, pConstant = 0.01)

## End(Not run)

PoisonAlien/maftools documentation built on Nov. 5, 2024, 4:12 p.m.