Skip to contents

PCA Plot of MaxQuant Sample Columns

Source: R/pca_plot_maxquant.R
pca_plot_maxquant.Rd

This function performs principal component analysis on the measurement columns present in a processed MaxQuant subset and plots PC1 against PC2, with each sample column shown as a point. The input data frame should be the same processed subset that would be passed to volcano_plot_maxquant().

Usage

pca_plot_maxquant(
  df_subset = NULL,
  meas = "LFQ.intensity",
  groups = NULL,
  by_protein = FALSE,
  threshold_p = 0.05,
  threshold_fc = 1,
  vp_colours = NULL,
  x_label = NULL,
  y_label = NULL,
  xy_line = TRUE,
  fsize = 8,
  label_points = NULL,
  point_args = list(size = 3, alpha = 0.8),
  label_args = list(size = 3, max.overlaps = Inf, segment.alpha = 0.5, segment.size =
    0.2, colour = "black")
)

Arguments

df_subset

a data frame containing the subset of the MaxQuant data that will be used to generate the plot

meas

string indicating the measurement type (e.g. "LFQ.intensity", "Intensity", etc.)

groups

optional character vector used to relabel the experimental groups in the legend. If NULL, group names are deduced from the measurement column names.

by_protein

boolean indicating whether to perform PCA on the protein rows (default is FALSE). If TRUE, PCA will be performed on the protein rows rather than the sample columns.

threshold_p

numeric indicating the p-value threshold. Used if by_protein is TRUE to colour the points in the PCA plot based on significance.

threshold_fc

numeric indicating the fold change threshold (in log2 space, i.e. 1 is a 2-fold change, 2 is a 4-fold change, etc.). Used if by_protein is TRUE to colour the points in the PCA plot based on significance.

vp_colours

a named vector of colours for the PCA plot.

x_label

string specifying the label for the x-axis. If NULL, a default label showing the variance explained by PC1 will be generated.

y_label

string specifying the label for the y-axis. If NULL, a default label showing the variance explained by PC2 will be generated.

xy_line

boolean indicating whether to add a line to indicate x = y = 0

fsize

numeric indicating the font size to use for the plot (default is 8)

label_points

boolean indicating whether to label sample points (default is NULL and triggers a setting of FALSE if by_protein is TRUE, and TRUE if by_protein is is FALSE). If TRUE, sample points will be labelled with their sample names. If FALSE, no labels will be added.

point_args

a list of arguments to be passed to geom_point() for the points in the PCA plot (default is list(size = 3, alpha = 0.8)). If the user specifies additional arguments or changes a default argument, the defaults will be merged with the user-specified entries.

label_args

a list of arguments to be passed to ggrepel::geom_text_repel() for the labels in the PCA plot (default is list(size = 3, max.overlaps = Inf, segment.alpha = 0.5, segment.size = 0.2, colour = "black")). If the user specifies additional arguments or changes a default argument, the defaults will be merged with the user-specified entries.

Value

ggplot object containing the PCA plot

Examples

if (FALSE) { # \dontrun{
# load the MaxQuant data
df <- load_maxquant(file = "proteinGroups.txt", datadir = "inst/extdata")
# process the MaxQuant data to get the subset for plotting
df_subset <- process_maxquant(df, group1 = "WT", group2 = "Control")
# generate the PCA plot
pca_plot_maxquant(df_subset)
} # }