preprocessIntervals: Preprocess intervals

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

Description

Optimize intervals for copy number calling by tiling long intervals and by including off-target regions. Uses scanFa from the Rsamtools package to retrieve GC content of intervals in a reference FASTA file. If provided, will annotate intervals with mappability and replication timing scores.

Usage

preprocessIntervals(
  interval.file,
  reference.file,
  output.file = NULL,
  off.target = FALSE,
  average.target.width = 400,
  min.target.width = 100,
  min.off.target.width = 20000,
  average.off.target.width = 2e+05,
  off.target.padding = -500,
  mappability = NULL,
  min.mappability = c(0.5, 0.1, 0.7),
  reptiming = NULL,
  average.reptiming.width = 1e+05,
  exclude = NULL,
  off.target.seqlevels = c("targeted", "all"),
  small.targets = c("resize", "drop")
)

Arguments

interval.file	File specifying the intervals. Interval is expected in first column in format CHR:START-END. Instead of a file, a GRanges object can be provided. This allows the use of BED files for example. Note that GATK interval files are 1-based (first position of the genome is 1). Other formats like BED files are often 0-based. The import function will automatically convert to 1-based GRanges.
reference.file	Reference FASTA file.
output.file	Optionally, write GC content file.
off.target	Include off-target regions.
average.target.width	Split large targets to approximately this size.
min.target.width	Make sure that target regions are of at least this specified width. See small.targets.
min.off.target.width	Only include off-target regions of that size
average.off.target.width	Split off-target regions to that
off.target.padding	Pad off-target regions.
mappability	Annotate intervals with mappability score. Assumed on a scale from 0 to 1, with score being 1/(number alignments). Expected as GRanges object with first meta column being the score. Regions outside these ranges are ignored, assuming that mappability covers the whole accessible genome.
min.mappability	double(3) specifying the minimum mappability score for on-target, off-target, and chrY regions in that order. The chrY regions are only used for sex determination in ‘PureCN’ and are therefore treated differently. Requires mappability.
reptiming	Annotate intervals with replication timing score. Expected as GRanges object with first meta column being the score.
average.reptiming.width	Tile reptiming into bins of specified width.
exclude	Any target that overlaps with this GRanges object will be excluded.
off.target.seqlevels	Controls how to deal with chromosomes/contigs found in the reference.file but not in the interval.file.
small.targets	Strategy to deal with targets smaller than min.target.width.

Value

Returns GC content by interval as GRanges object.

Author(s)

Markus Riester

References

Talevich et al. (2016). CNVkit: Genome-Wide Copy Number Detection and Visualization from Targeted DNA Sequencing. PLoS Comput Biol.

Examples

reference.file <- system.file("extdata", "ex2_reference.fa", 
    package="PureCN", mustWork = TRUE)
interval.file <- system.file("extdata", "ex2_intervals.txt", 
    package="PureCN", mustWork = TRUE)
bed.file <- system.file("extdata", "ex2_intervals.bed", 
    package="PureCN", mustWork = TRUE)
preprocessIntervals(interval.file, reference.file, 
    output.file="gc_file.txt")

intervals <- import(bed.file)
preprocessIntervals(intervals, reference.file, 
    output.file="gc_file.txt")

PureCN documentation built on Nov. 8, 2020, 5:37 p.m.