DataFrameQuantitativeTrack-class: Constructor for DataFrameQuantitativeTrack

DataFrameQuantitativeTrack-classR Documentation

Constructor for DataFrameQuantitativeTrack

Description

DataFrameQuantitativeTrack creates and IGV track for bed objects imported using rtracklayer

Usage

DataFrameQuantitativeTrack(
  trackName,
  quantitativeData,
  color = "blue",
  trackHeight = 50,
  autoscale,
  min = NA_real_,
  max = NA_real_,
  visibilityWindow = 1e+05
)

Arguments

trackName

A character string, used as track label by igv, we recommend unique names per track.

quantitativeData

A base R data.frame

color

A CSS color name (e.g., "red" or "#FF0000")

trackHeight

track height, typically in range 20 (for annotations) and up to 1000 (for large sample vcf files)

autoscale

Autoscale track to maximum value in view

min

Sets the minimum value for the data (y-axis) scale. Usually zero.

max

Sets the maximum value for the data (y-axis) scale. This value is ignored if autoscale is TRUE

visibilityWindow

Maximum window size in base pairs for which indexed annotations or variants are displayed. Defaults: 1 MB for variants, whole chromosome for other track types.

Details

Detailed description goes here

Value

A DataFrameQuantitativeTrack object

See Also

DataFrameAnnotationTrack

GRangesQuantitativeTrack

GRangesAnnotationTrack

DataFrameAnnotationTrack

DataFrameQuantitativeTrack

GRangesAnnotationTrack

GRangesQuantitativeTrack

GenomicAlignmentTrack

UCSCBedAnnotationTrack

UCSCBedGraphQuantitativeTrack

VariantTrack

igvAnnotationTrack

Examples

base.loc <- 88883100
tbl.blocks <- data.frame(chrom=rep("chr5", 3),
                  start=c(base.loc, base.loc+100, base.loc + 250),
                  end=c(base.loc + 50, base.loc+120, base.loc+290),
                  score=runif(3),
                  stringsAsFactors=FALSE)

track.blocks <- DataFrameQuantitativeTrack("blocks", tbl.blocks, autoscale=TRUE)

locs <- seq(from=base.loc, length.out=1000)
tbl.wig <- data.frame(chrom=rep("chr5", 1000), start=locs-1, end=locs,
                      score=runif(n=1000, min=-1, max=1))
track.wig <- DataFrameQuantitativeTrack("wig", tbl.wig, autoscale=FALSE,
                                        min=min(tbl.wig$score), max=max(tbl.wig$score),
                                        color="random")
if(interactive()){
   igv <- igvR()
   setGenome(igv, "hg38")
   setBrowserWindowTitle(igv, "DataFrameQuantitativeTrack demo")
   displayTrack(igv, track.blocks)
   roi <- sprintf("%s:%d-%d", tbl.blocks$chrom[1],
                  min(tbl.blocks$start)-1000, max(tbl.blocks$end) + 1000)
   showGenomicRegion(igv, roi)
   displayTrack(igv, track.wig)
   }


paul-shannon/IGV documentation built on Nov. 17, 2024, 11:55 p.m.