tests/test_plotting_functions.R

In BoutrosLab.plotting.general: Functions to Create Publication-Quality Plots

### TEST-SUITE #####################################################################################
# Description: This script runs the plotting code in BoutrosLab.plotting.general for testing purposes
# The plots are stored in variables and plotted together in a single multiple in order to save plot-generation time

warning(paste(
    'BPG will discontinue this test file in v8.0.0.',
    'Instead, tests will use the standard testthat format.',
    'For more information, see https://testthat.r-lib.org/'
    ));

### LOAD LIBRARY ###################################################################################
library(BoutrosLab.plotting.general);

### PLOTTING FUNCTIONS #############################################################################
test.data <- data.frame(
	x = 1:10,
	y = LETTERS[1:10],
	z = rnorm(10),
	type = rep(LETTERS[1:2], 5),
	stringsAsFactors = TRUE
	);

# testing legend.grob
covariate.legend <- list(
	legend = list(
		colours = default.colours(2),
		labels = levels(test.data$type)[1:2]
		)
	);

legend.grob2 <- legend.grob(
	legends = covariate.legend
	);

# testing covariates.grob
covariate.colours1 <- as.character(test.data$type);
covariate.colours1[covariate.colours1 == 'A'] <- default.colours(2)[1];
covariate.colours1[covariate.colours1 == 'B'] <- default.colours(2)[2];

# create an object to draw the covariates from
covariates1 <- list(
	rect = list(
		col = 'black',
		fill = covariate.colours1,
		lwd = 1.5
		)
	);

covariates.grob1 <- covariates.grob(
	covariates = covariates1,
	ord = c(1:10),
	side = 'top',
	size = .8
	);

barplot <- create.barplot(
	formula = y ~ x,
	data = test.data,
	legend = list(
		bottom = list(
			fun = covariates.grob1
			),
		right = list(
			fun = legend.grob2
			),
		inside = list(
			fun = draw.key,
			args = list(
				key = get.corr.key(
					x = test.data$x,
					y = test.data$y,
					label.items = c('spearman', 'spearman.p', 'kendall', 'beta1')
					)
				),
			x = 0.5,
			y = 0.5
			)
		),
	description = 'testing metadata'
	);

# this isn't printed anywhere
dend <- create.dendrogram(
	x = data.frame(test.data$z, rnorm(10))
	);

# # This is the slowest running function
# heatmap <- create.heatmap(
# 	x = as.matrix(test.data$z),
# 	clustering.method = 'none'
# 	);

dotmap <- create.dotmap(
	x = test.data$z
	);

hexbin <- create.hexbinplot(	
	formula = z ~ x,
	data = test.data
	);

# Also using generate.at.final()
density <- create.densityplot(
	x = list(
		a = test.data$x, 
		b = test.data$z
		),
	xat = c(0, 1.1, 1.5, 1, 6, 4),
	yat = c(0.34, 0.38, 0.7),
	type = c('g', 'l'),
	resolution = 100
	);

manhattanplot <- create.manhattanplot(
	formula = x ~ y,
	data = test.data
	);

polygonplot <- create.polygonplot(	
	formula = NA ~ x,
	data = test.data,
	max = test.data$x,
	min = test.data$z,
	xlimits = c(0, 10),
	ylimits = c(-2, 10)
	);

qqcomp <- create.qqplot.comparison(
	x = list(test.data$x, test.data$z)
	);

scatter <- create.scatterplot(
	formula = x ~ z,
	data = test.data
	);

seg <- create.segplot(
	formula = y ~ x + z,
	data = test.data
	);

violin <- create.violinplot(
	formula = y ~ z,
	data = test.data
	);

strip <- create.stripplot(
	formula = z ~ y,
	data = test.data
	);

boxplot <- create.boxplot(
	formula = x ~ y,
	data = test.data
	);

# these do not easily join the multiplot
histogram <- create.histogram(
	x = test.data$z
	);

qqfit <- create.qqplot.fit(
	x = test.data$z
	);

# Plot everything in one plot for quicker running time
create.multiplot(
	file = NULL,
	plot.objects = list(polygonplot, manhattanplot, hexbin, dotmap, density, qqcomp, scatter, seg, violin, barplot, strip, boxplot)
	);

### HELPER FUNCTIONS ###############################################################################

get.line.breaks(1:10);

scientific.notation(1234, 2, type = 'list');
scientific.notation(0, 2);
scientific.notation(c(1234, 1234), 1);

# This returns the same thing as when type = 'expression' - should something be changed?
scientific.notation(c(1234, 1234), 1, type = 'list');

display.statistical.result(1234);

### COLOUR FUNCTIONS ###############################################################################
default.colours(12, is.greyscale = FALSE);
default.colours(5, palette.type = 'chromosomes', is.greyscale = FALSE);
default.colours(5, palette.type = 'seq');
default.colours(c(4, 4), palette.type = c('seq', 'div'));

# display.colours(default.colours(3));
# display.colours(grey(0.5));

force.colour.scheme(c('stopgain snv', 'splicing'), 'annovar.annotation');