filterIntervals: Remove low quality intervals

In lima1/PureCN: Copy number calling and SNV classification using targeted short read sequencing

Remove low quality intervals

Description

This function determines which intervals in the coverage files should be included or excluded in the segmentation. It is called via the fun.filterIntervals argument of runAbsoluteCN. The arguments are passed via args.filterIntervals.

Usage

filterIntervals(
  normal,
  tumor,
  log.ratio,
  seg.file,
  filter.lowhigh.gc = 0.001,
  min.coverage = 15,
  min.total.counts = 100,
  min.targeted.base = 5,
  min.mappability = c(0.6, 0.1),
  min.fraction.offtarget = 0.05,
  normalDB = NULL
)

Arguments

normal	Coverage data for normal sample.
tumor	Coverage data for tumor sample.
log.ratio	Copy number log-ratios, one for each interval in the coverage file.
seg.file	If not NULL, then do not filter intervals, because data is already segmented via the provided segmentation file.
filter.lowhigh.gc	Quantile q (defines lower q and upper 1-q) for removing intervals with outlier GC profile. Assuming that GC correction might not have been worked on those. Requires interval.file.
min.coverage	Minimum coverage in both normal and tumor. Intervals with lower coverage are ignored. If a normalDB is provided, then this database already provides information about low quality intervals and the min.coverage is set to min.coverage/10000.
min.total.counts	Exclude intervals with fewer than that many reads in combined tumor and normal.
min.targeted.base	Exclude intervals with targeted base (size in bp) smaller than this cutoff. This is useful when the same interval file was used to calculate GC content. For such small targets, the GC content is likely very different from the true GC content of the probes.
min.mappability	double(2) specifying the minimum mappability score for on-target, off-target in that order.
min.fraction.offtarget	Skip off-target regions when less than the specified fraction of all intervals passes all filters
normalDB	Normal database, created with createNormalDatabase.

Value

logical(length(log.ratio)) specifying which intervals should be used in segmentation.

Author(s)

Markus Riester

Examples

normal.coverage.file <- system.file("extdata", "example_normal.txt.gz",
    package = "PureCN")
normal2.coverage.file <- system.file("extdata", "example_normal2.txt.gz",
    package = "PureCN")
normal.coverage.files <- c(normal.coverage.file, normal2.coverage.file)
normalDB <- createNormalDatabase(normal.coverage.files)

tumor.coverage.file <- system.file("extdata", "example_tumor.txt.gz",
    package = "PureCN")
vcf.file <- system.file("extdata", "example.vcf.gz",
    package = "PureCN")
interval.file <- system.file("extdata", "example_intervals.txt",
    package = "PureCN")

# The max.candidate.solutions, max.ploidy and test.purity parameters are set to
# non-default values to speed-up this example.  This is not a good idea for real
# samples.
ret <-runAbsoluteCN(normal.coverage.file = normal.coverage.file,
    tumor.coverage.file = tumor.coverage.file,
    genome = "hg19", vcf.file = vcf.file, normalDB = normalDB,
    sampleid = "Sample1", interval.file = interval.file,
    args.filterIntervals = list(min.targeted.base = 10), max.ploidy = 4,
    test.purity = seq(0.3, 0.7, by = 0.05), max.candidate.solutions = 1)

lima1/PureCN documentation built on Sept. 17, 2024, 5:48 a.m.