Plot of channel intensity marginal densities

ggplotMarginalDensities uses ggplot2 to draw plots of marginal densities of selected channels of a flowSet. If the flowSet contains several flowFrames, all events are concatenated together. By default, a pseudo Rsquare projection quality indicator, and the number of dimensions of the MDS projection are provided in sub-title

ggplotMarginalDensities(
  x,
  sampleSubset,
  channels,
  pDataForColour,
  pDataForGroup,
  nEventInSubsample = Inf,
  seed = NULL,
  transList
)

Arguments

x

a flowCore::flowSet (or a single flowCore::flowFrame)

sampleSubset

(optional) a logical vector, of size nrow(pData), which is by construction the nb of samples, indicating which samples to keep in the plot. Typically it is obtained through the evaluation of a logical condition on pData rows.

channels

(optional)

pDataForColour

(optional) which phenoData(fs) variable will be used as colour aesthetic. Should be a character.

pDataForGroup

(optional) which phenoData(fs) variable will be used as group aesthetic. Should be a character.

nEventInSubsample

how many event to take (per flowFrame of the flowSet).

seed

if not null, used in subsampling.

transList

a flowCore::transformList that will be applied before plotting.

Value

a ggplot object

Examples

library(CytoPipeline)

data(OMIP021Samples)

# estimate scale transformations 
# and transform the whole OMIP021Samples

transList <- estimateScaleTransforms(
    ff = OMIP021Samples[[1]],
    fluoMethod = "estimateLogicle",
    scatterMethod = "linearQuantile",
    scatterRefMarker = "BV785 - CD3")

OMIP021Trans <- CytoPipeline::applyScaleTransforms(
    OMIP021Samples, 
    transList)
    

# As there are only 2 samples in OMIP021Samples dataset,
# we create artificial samples that are random combinations of both samples

ffList <- c(
    flowCore::flowSet_to_list(OMIP021Trans),
    lapply(3:5,
           FUN = function(i) {
               aggregateAndSample(
                   OMIP021Trans,
                   seed = 10*i,
                   nTotalEvents = 5000)[,1:22]
           }))

fsNames <- c("Donor1", "Donor2", paste0("Agg",1:3))
names(ffList) <- fsNames

fsAll <- as(ffList,"flowSet")

flowCore::pData(fsAll)$grpId <- factor(c("D1", "D2", rep("Agg", 3)))
flowCore::pData(fsAll)$lbl <- paste0("S", 1:5)

# plot densities, all samples together
p <- ggplotMarginalDensities(fsAll)

# plot densities, per sample
p <- ggplotMarginalDensities(fsAll, pDataForGroup = "lbl")

# plot densities, per sample and coloured by group
p <- ggplotMarginalDensities(
    fsAll, 
    pDataForGroup = "lbl",
    pDataForColour = "grpId")