plotTest {IWTomics} | R Documentation |
The function plotTest
creates graphical representations of the Interval-Wise
Testing results present in the test
slot of a "IWTomicsData"
object.
In particular, it draws a heatmap of the adjusted p-value functions at each possible
scale (with the different maximum interval lengths considered), a plot of the adjusted
p-value curve at the chosen scale threshold and a plot of the feature measurements in
the region datasets tested (aligned curves or pointwise quantile curves).
plotTest(regionsFeatures, alpha=0.05, scale_threshold=NULL, nlevel=100, type="boxplot", N_regions=pmin(lengthRegions(regionsFeatures),10), probs=c(0.25,0.5,0.75), average=TRUE, size=TRUE, id_features_subset=idFeatures(regionsFeatures), col=1+seq_len(nRegions(regionsFeatures)), ask=TRUE, xlab="Windows", ylim=NULL, ...)
regionsFeatures |
|
alpha |
level of the hypothesis test used to select and display significant
results. Default |
scale_threshold |
threshold on the test scale (maximum interval length for
the p-value adjustment) to be used in the adjusted p-value plot. Can be either
a scalar (the same length for all features) or a vector (a length for each feature)
or a list of vectors (a vector for each test). If |
nlevel |
number of desired color levels for the adjusted p-value heatmap. |
type |
type of plot to be drawn in addition to the adjusted p-value heatmap and plot. Possible types are:
|
N_regions |
number of regions to be randomly chosen (for each region dataset) in
|
probs |
probabilities corresponding to the quantiles to be drawn in |
average |
if |
size |
if |
id_features_subset |
vector with the identifiers of the features to be plotted. |
col |
vector of plotting colors for the different region datasets. |
ask |
if |
xlab |
a title for the x axis. |
ylim |
the y limits of the feature measurement plot. |
... |
additional plot parameters. |
No value returned. The function produces a graphical output.
Marzia A Cremona, Alessia Pini, Francesca Chiaromonte, Simone Vantini
A Pini and S Vantini (2017). Interval-Wise Testing for functional data. Journal of Nonparametric Statistics.
IWTomicsData
for "IWTomicsData"
class, constructors, accessors and methods;
IWTomicsTest
function to perform the Interval-wise Testing;
plotSummary
to draw a summary plot of the test results.
data(regionsFeatures_center) ## One sample test for 'control' regions and feature 'ftr1' regionsFeatures_test=IWTomicsTest(regionsFeatures_center, id_region1='control',id_features_subset='ftr1') ## Plotting the results of the one sample test plotTest(regionsFeatures_test,col=5) ## Set the scale threshold to 25 plotTest(regionsFeatures_test,col=5,scale_threshold=25) ## Plot curves instead of boxplots, do not report sample size in each position plotTest(regionsFeatures_test,col=5,type='curves',size=FALSE) ## Two sample test for 'elem1', 'elem2' and 'elem3' vs 'control' regions regionsFeatures_test=IWTomicsTest(regionsFeatures_center, id_region1=c('elem1','elem2','elem3'), id_region2=c('control','control','control')) ## Plotting the results of the two sample test plotTest(regionsFeatures_test) ## Set scale thresholds for the different features plotTest(regionsFeatures_test,scale_threshold=c(25,30)) ## Plot only results regarding feature 'ftr2', setting scale thresholds for each test plotTest(regionsFeatures_test,id_features_subset='ftr2', scale_threshold=list(test1=10,test2=20,test3=30)) ## Report also 0.05 and 0.95 quantiles in the boxplots plotTest(regionsFeatures_test,id_features_subset='ftr2', scale_threshold=list(test1=10,test2=20,test3=30), probs=c(0.05,0.25,0.5,0.75,0.9))