wqspt_plot {wqspt} | R Documentation |
Plotting method for wqspt object
Description
Generates plots to help visualize and summarize WQS permutation test results.
Usage
wqspt_plot(
wqsptresults,
FixedPalette = FALSE,
InclKey = FALSE,
AltMixName = NULL,
AltOutcomeName = NULL,
ViridisPalette = "D",
StripTextSize = 14,
AxisTextSize.Y = 12,
AxisTextSize.X = 12,
LegendTextSize = 14,
LegendWidthIn = 0.4,
LegendHeightIn = 0.4,
PvalLabelSize = 5,
HeatMapTextSize = 5
)
Arguments
wqsptresults |
An object of class |
FixedPalette |
If TRUE, the heatmap color key for the mixture weights has categorical cutoffs with the following categories: <0.1, 0.1 - <0.2, 0.2 - <0.3, and >= 0.3. If false, the heatmap color key is continuous and dependent on the weight values. |
InclKey |
If TRUE, a horizontal heatmap legend is included at the bottom of the full plot. |
AltMixName |
Defaults to NULL. If not NULL, these are alternative names for the mixture components to be displayed on the heatmap y axis. |
AltOutcomeName |
Defaults to NULL. If not NULL, this is an alternative name for the outcome to be displayed on the heatmap x axis. |
ViridisPalette |
Color palette to be used for the viridisLite package-based coloring of the heatmap, with possible values from 'A' to 'E'. Defaults to 'D'. |
StripTextSize |
Text size for the plot strip labels. Defaults to 14. |
AxisTextSize.Y |
Text size for the y axis text. Defaults to 12. |
AxisTextSize.X |
Text size for the x axis text. Defaults to 12. |
LegendTextSize |
Text size for the legend text. Defaults to 14. |
LegendWidthIn |
Width of the horizontal bottom legend in inches. Defaults to 0.4. |
LegendHeightIn |
Height of the horizontal bottom legend in inches. Defaults to 0.4. |
PvalLabelSize |
The geom_text size for the permutation test p-value label. Defaults to 5. |
HeatMapTextSize |
The geom_text size for the mixture weight heatmap labels. Defaults to 5. |
Value
Returns a list with 4 objects.
FullPlot |
Two plots stacked vertically: (1) A forest plot of the beta WQS coefficient with the naive confidence intervals as well as the permutation test p-value and (2) a heatmap of the WQS weights for each mixture component. |
CoefPlot |
Forest plot of the beta WQS coefficient with the naive confidence intervals as well as the permutation test p-value. |
WtPlot |
A heatmap of the WQS weights for each mixture component. |
WtLegend |
A legend for the weights in the WtPlot heatmap, saved as a TableGrob object. |
Examples
library(gWQS)
# mixture names
PCBs <- names(wqs_data)[1:10] #10 of the original 34 for a quick example
#quick example WQSPT model
perm_test_res <- wqs_full_perm(formula = yLBX ~ wqs, data = wqs_data,
mix_name = PCBs, q = 10, b_main = 4,
b_perm = 4, b1_pos = TRUE, b_constr = FALSE,
niter = 3, seed = 16,
plan_strategy = "multicore",
stop_if_nonsig = FALSE)
#plot with continuous heatmap for mixture weights, omitting the heatmap
# legend on the plot
wqsptplot <- wqspt_plot(perm_test_res)
#plot the full plot with the WQS coefficient forest plot above & the mixture
# weight heatmap below
wqsptplot$FullPlot
#plot with binned heatmap for mixture weights
wqsptplot <- wqspt_plot(perm_test_res, InclKey = TRUE, LegendWidthIn = 0.8,
FixedPalette = TRUE)
#plot the full plot with the WQS coefficient forest plot above & the mixture
# weight heatmap below
wqsptplot$FullPlot