kozakHeatmap: Make sequence region heatmap relative to scoring

View source: R/general_plots.R

kozakHeatmapR Documentation

Make sequence region heatmap relative to scoring

Description

Given sequences, DNA or RNA. And some score, ribo-seq fpkm, TE etc. Create a heatmap divided per letter in seqs, by how strong the score is.

Usage

kozakHeatmap(
  seqs,
  rate,
  start = 1,
  stop = max(nchar(seqs)),
  center = ceiling((stop - start + 1)/2),
  min.observations = ">q1",
  skip.startCodon = FALSE,
  xlab = "TIS",
  type = "ribo-seq"
)

Arguments

seqs

the sequences (character vector, DNAStringSet)

rate

a scoring vector (equal size to seqs)

start

position in seqs to start at (first is 1), default 1.

stop

position in seqs to stop at (first is 1), default max(nchar(seqs)), that is the longest sequence length

center

position in seqs to center at (first is 1), center will be +1 in heatmap

min.observations

How many observations per position per letter to accept? numeric or quantile, default (">q1", bigger than quartile 1 (25 percentile)). You can do (10), to get all with more than 10 observations.

skip.startCodon

startCodon is defined as after centering (position 1, 2 and 3). Should they be skipped ? default (FALSE). Not relevant if you are not doing Translation initiation sites (TIS).

xlab

Region you are checking, default (TIS)

type

What type is the rate scoring ? default (ribo-seq)

Details

It will create blocks around the highest rate per position

Value

a ggplot of the heatmap

Examples


## Not run: 
if (requireNamespace("BSgenome.Hsapiens.UCSC.hg19")) {
  txdbFile <- system.file("extdata", "hg19_knownGene_sample.sqlite",
                          package = "GenomicFeatures")
  #Extract sequences of Coding sequences.
  cds <- loadRegion(txdbFile, "cds")
  tx <- loadRegion(txdbFile, "mrna")

  # Get region to check
  kozakRegions <- startRegionString(cds, tx, BSgenome.Hsapiens.UCSC.hg19::Hsapiens
                                    , upstream = 4, 5)
  # Some toy ribo-seq fpkm scores on cds
  set.seed(3)
  fpkm <- sample(1:115, length(cds), replace = TRUE)
  kozakHeatmap(kozakRegions, fpkm, 1, 9, skip.startCodon = F)
}

## End(Not run)


Roleren/ORFik documentation built on Nov. 13, 2024, 10 p.m.