BootList: BootList

In budczies/CNVPanelizer: Reliable CNV detection in targeted sequencing applications

Description

Performs a hybrid bootstrapping subsampling procedure similar to random forest. It bootstraps the reference samples and subsamples the amplicons associated with each gene. Returns a distribution of sample/reference ratios for each gene and sample of interest combination.

Usage

    BootList(geneNames, sampleMatrix, refmat, replicates)

Arguments

geneNames	A vector of gene names, with one entry for each sequenced amplicon.
sampleMatrix	A vector or matrix of the read counts from the sample of interest. In the case of a matrix columns represent samples and rows amplicons.
refmat	A matrix of the read counts obtianed from the reference samples. Columns represent reference samples and rows amplicons.
replicates	How many bootstrap replicates should be performed.

Value

Returns a list of numeric matrices: For each matrix a row represent a gene while each column represents a bootstrapping/subsampling iteration.

Author(s)

Thomas Wolf, Cristiano Oliveira

Examples

data(sampleReadCounts)
data(referenceReadCounts)
## Gene names should be same size as row columns
geneNames <- row.names(referenceReadCounts)

ampliconNames <- NULL

normalizedReadCounts <- CombinedNormalizedCounts(sampleReadCounts,
                                                 referenceReadCounts,
                                                 ampliconNames = ampliconNames)

# After normalization data sets need to be splitted again to perform bootstrap
samplesNormalizedReadCounts = normalizedReadCounts["samples"][[1]]
referenceNormalizedReadCounts = normalizedReadCounts["reference"][[1]]

# Should be used values above 10000
replicates <- 10

# Perform the bootstrap based analysis
bootList <- BootList(geneNames,
         samplesNormalizedReadCounts,
         referenceNormalizedReadCounts,
         replicates = replicates)

budczies/CNVPanelizer documentation built on May 29, 2019, 1:41 p.m.