detectActiveSignatures: Get a list of signatures active in a sample

In morrislab/TrackSigFreq: TrackSig - cancer evolutionary trajectory estimation

Description

detectActiveSignatures determines what signatures are active above an activity threshold, from a list of signature definintions. To do this, a multinomial mixture model of signature activities is fit to the collection of all mutations in the sample.

Usage

detectActiveSignatures(
  vcfFile,
  cnaFile = NULL,
  purity = 1,
  threshold = 0.05,
  prior = NULL,
  binSize = 100,
  referenceSignatures = alex_merged,
  refGenome = BSgenome.Hsapiens.UCSC.hg19::BSgenome.Hsapiens.UCSC.hg19
)

Arguments

vcfFile	path to variant calling format (vcf) file
cnaFile	path to copy number abberation (cna) file. If not provided, all copy numbers default to 2.
purity	number between 0 and 1 of the percentage of cells in the sample that are cancerous
threshold	minimum activity level that signature must have to be detected
prior	prior on the likelihood of observing a given signature (must match signatures present in referenceSignatures)
binSize	number of mutations per bin (default 100)
referenceSignatures	dataframe containing definitions of mutational signatures.
refGenome	BSgenome to use as reference

Value

Names of signatures active in sample.

morrislab/TrackSigFreq documentation built on July 5, 2021, 6:33 a.m.