dropletPlot: Plot a droplet classification with a colour-blind palette,...

In CRUKMI-ComputationalBiology/twoddpcr: Classify 2-d Droplet Digital PCR (ddPCR) data and quantify the number of starting molecules

Description

Plot an object comprising droplet amplitudes and their classification. If specified, centres of clusters can be marked, e.g. k-means clustering can take a set of centres as the initial centres of the algorithm, and the algorithm also outputs the final cluster centres. Limits to the axes can also be set for ease of comparison and consistency.

If a ggplot object is given as a parameter, this method will simply plot it with the pretty colours, centres and restrictions on the axes.

If a data.frame is given as a parameter, it should correspond to droplets with their classification.

If a ddpcrWell object is given as a parameter, plot the droplets in the well with its classification.

If a ddpcrPlate object is given as a parameter, plot the droplets from all wells with their classifications.

Usage

dropletPlot(
  droplets,
  ch1Label = "Ch1 Amplitude",
  ch2Label = "Ch2 Amplitude",
  ...
)

## S4 method for signature 'data.frame'
dropletPlot(
  droplets,
  ch1Label = "Ch1 Amplitude",
  ch2Label = "Ch2 Amplitude",
  cMethod = "None",
  mapping = aes_string(x = "Ch2.Amplitude", y = "Ch1.Amplitude", colour = cMethod),
  finalCentres = NULL,
  initialCentres = NULL,
  selectedCentre = NULL,
  pointSize = 1,
  plotLimits = list(x = c(1000, 9000), y = c(3000, 13500)),
  legendLabels = ddpcr$classesRain
)

## S4 method for signature 'ddpcrWell'
dropletPlot(
  droplets,
  ch1Label = "Ch1 Amplitude",
  ch2Label = "Ch2 Amplitude",
  cMethod = "None",
  mapping = aes_string(x = "Ch2.Amplitude", y = "Ch1.Amplitude", colour = cMethod),
  finalCentres = NULL,
  initialCentres = NULL,
  selectedCentre = NULL,
  pointSize = 1,
  plotLimits = list(x = c(1000, 9000), y = c(3000, 13500)),
  legendLabels = ddpcr$classesRain
)

## S4 method for signature 'ddpcrPlate'
dropletPlot(
  droplets,
  ch1Label = "Ch1 Amplitude",
  ch2Label = "Ch2 Amplitude",
  cMethod = "None",
  mapping = aes_string(x = "Ch2.Amplitude", y = "Ch1.Amplitude", colour = cMethod),
  finalCentres = NULL,
  initialCentres = NULL,
  selectedCentre = NULL,
  pointSize = 1,
  plotLimits = list(x = c(1000, 9000), y = c(3000, 13500)),
  legendLabels = ddpcr$classesRain
)

Arguments

droplets	An object corresponding to droplet amplitudes and their classifications. This can be in the form of: A data frame with columns Ch1.Amplitude, Ch2.Amplitude and a classification column (see the parameter cMethod). A ddpcrWell object. A ddpcrPlate object. A ggplot (gg) object. For example, this could be the output of ggplot.well or ggplot.plate. We should not need to use this unless we are writing new methods to plot new data types.
ch1Label	The label for the channel 1 target. Defaults to "Ch1 Amplitude".
ch2Label	The label for the channel 2 target. Defaults to "Ch2 Amplitude".
...	Other plotting parameters that depend on the object type of droplets.
cMethod	This should be the name or column number of droplets corresponding to the classification. This column should only have entries in "NN", "PN", "NP, "PP", "Rain" and "N/A". If "None", plots the droplets with all of them classified as N/A. Defaults to "None".
mapping	A list of aesthetic mappings to use for the plot. Defaults to ggplot2::aes_string(x="Ch2.Amplitude", y="Ch1.Amplitude", colours=cMethod). Not used if droplets is a ggplot object.
finalCentres	A data frmae of final centres to plot (e.g. those returned by the k-means or c-means algorithms). If NULL, nothing is plotted. Defaults to NULL.
initialCentres	A data frame of initial centres to plot (e.g. initial cluster centres used in the k-means). If NULL, nothing is plotted. Defaults to NULL. This parameter is useful for illustrative reasons.
selectedCentre	An initial centre to highlight. This should be either "NN", "NP", "PN" or "PP". If NULL, nothing is highlighted. Defaults to NULL. This parameter is useful for illustrative reasons.
pointSize	The size to draw each droplet. Defaults to 1.
plotLimits	A list of 2-element vectors with names x and y. These are used to fix the x and y limits of the plot, which is especially useful for comparing plots. Defaults to list(x=c(1000, 9000), y=c(3000, 13500)).
legendLabels	The character vector corresponding to the labels for the legend. The elements of the vector should correspond to the NN, NP, PN, PP, Rain and N/A classes, respectively. Defaults to ddpcr$classesRain.

Value

A ggplot object with all of the given information above.

Author(s)

Anthony Chiu, [e-mail]

Examples

## Get a data frame and relabel the "Cluster" column to the right form.
aWell <- KRASdata[["E03"]]
aWell$Cluster <- relabelClasses(aWell, classCol="Cluster")

## Plot the data frame.
dropletPlot(aWell, cMethod="Cluster")

## Plot a ddpcrWell object.
aWell <- ddpcrWell(well=KRASdata[["E03"]])
dropletPlot(aWell, cMethod="Cluster")

## Plot a ddpcrPlate object.
krasPlate <- ddpcrPlate(wells=KRASdata[c("E03", "H03", "C04", "F04")])
dropletPlot(krasPlate, cMethod="Cluster")

## Use K-means clustering to classify a single sample. Then plot the
## classification and final cluster centres.
aWell <- kmeansClassify(aWell)
centres <- clusterCentres(aWell, cMethod="kmeans")
dropletPlot(aWell, cMethod="kmeans", finalCentres=centres)

CRUKMI-ComputationalBiology/twoddpcr documentation built on Feb. 14, 2021, 9:18 p.m.