Multi-sample copy-number heatmap

Builds a seq_plot() with a single track arranging per-sample CN calls as a heatmap: samples are placed on the y-axis (top to bottom according to sample_order), genomic windows along the x-axis, and each bin is a tile coloured by CN state.

Usage

seq_cn_heatmap(
  data,
  windows,
  sample_col = NULL,
  cn_col = NULL,
  state_colors = NULL,
  sample_order = NULL,
  bins = NULL,
  track_height = 3,
  track_id = NULL,
  ...
)

Arguments

data

A GRanges (long format) or numeric matrix.

windows

A GRanges of genomic view windows.

sample_col

Name of the mcols column giving sample identity. Auto-detected from c("sample", "sample_id", "Sample", "id") when NULL.

cn_col

Name of the mcols column giving integer CN state. Auto- detected from the same candidates as seq_copynumber().

state_colors

Named character vector keyed by CN state string. Defaults to the seq_copynumber() palette.

sample_order

Character vector of sample names in display order (top to bottom). When NULL, samples are sorted alphabetically.

bins

Optional GRanges giving the genomic position of each matrix column; required when data is a matrix.

track_height

Relative track height.

track_id

Character track_id for the generated track.

...

Additional arguments forwarded to seq_track().

Value

A SeqPlot object.

Details

Data may be supplied either as a long-format GRanges with one row per (sample, bin) carrying sample and cn mcols columns, or as a numeric matrix (rows = samples, cols = bins) plus a bins GRanges giving the genomic position of each column (passed via ...). Matrix input requires the matrix argument to have sample names as rownames.

Examples

library(GenomicRanges)
gr <- GRanges("chr1", IRanges(seq(1, 1e6, by = 2e4), width = 2e4),
              sample = rep(c("S1","S2","S3"), length.out = 50),
              cn     = sample(0:4, 50, replace = TRUE))
win <- GRanges("chr1", IRanges(1, 1e6))
seq_cn_heatmap(gr, windows = win)