ggpmisc 0.3.6

Package ‘ggpmisc’ focuses mainly on plot annotations. The new features added in this version required quite minor code changes but add features that I hope will be found useful.

Changes from version 0.3.5 the most recent CRAN release, are:


Annotations using NPC coordinates

NPC or native plot coordinates, are very useful for annotations. On the other hand, they are not of any use for plotting actual data. The geometries using this kind of position coordinates defined in ‘ggpmisc’ make it possible to have different annotations in the different panels when using facets. Traditional annotations in ‘ggplot2’ do not use the data, but instead take as arguments constant values for the aesthetics, and consequently add identical annotations to all panels. Annotations in ‘ggplot2’ use data coordinates. In many cases, the desired position of annotations is unrelated to the data, but instead related to the native coordinates of the plotting viewport. Using NPC coordinates for annotations allows consistent positioning, which is very important from the graphical design perspective.

Starting from version 0.3.6  ‘ggpmisc’ exports a modified definition of annotate() from ‘ggplot2’. The modification adds support of the position aesthetics npcx and npcy retaining all other functionality unaltered. As a consequence geometries "text_npc""label_npc""table_npc", "plot_npc", and "grob_npc" can now be used as the first argument to annotate(). In addition single ggplots, data frames and grobs as well as lists of such objects are accepted arguments to label.


# plot to be inset
p <- ggplot(mtcars, aes(factor(cyl), mpg, colour = factor(cyl))) +
  stat_boxplot() +
  labs(y = NULL) +
  theme_bw(9) + theme(legend.position = "none")

# main plot with p as an inset
ggplot(mtcars, aes(wt, mpg, colour = factor(cyl))) +
  geom_point() +
  annotate("plot_npc", npcx = "left", npcy = "bottom", label = p) +
  expand_limits(y = 0, x = 0)

Tagging equations with group labels (experimental)

Starting from version 0.3.6  stat_poly_eq() supports use of grouping with equations and identifying them by using labels.  Previously use of  the color aesthetic was the only way of “linking” equations to plotted curves, which is frequently distracting or unavailable for printing. After some past failed attempts at implementing this, I recently realized that using a pseudo-aesthetic made implementation very easy and its use flexible and straightforward. The catch is that this relies on undocumented behavior of ‘ggplot2’ and will not necessarily work with future versions of ‘ggplot2’. Statistic stat_poly_eq() now copies grp.label from its input into its returned value. One can map any variable to the pseudo-aesthetic grp.label to achieve this. Values are passed to the output only if all values within the group are the same, otherwise grp.label is filled with NA. The signature of stat_poly_eq() remains unchanged.


my.formula <- y ~ x

ggplot(mtcars, aes(wt, mpg, 
                   linetype = factor(cyl), shape = factor(cyl),
                   grp.label = factor(cyl))) +
  geom_point() +
  stat_smooth(formula = my.formula, method = "lm", colour = "black") +
  stat_poly_eq(aes(label = stat(paste("bold(\"cyl\"~~", grp.label, 
                                      "*':')~~~", eq.label, sep = ""))), 
               formula = my.formula, label.x = "right", parse = TRUE) +

Marking and or labeling group centroids

The new stat_centroid() and stat_summary_xy(). stat_centroid() applies the same function to x and y and this function defaults to mean(). In the case of stat_summary_xy() the functions applied to x and y are passed as separate arguments, and they both default to simply copying their input.


Starting from version 0.3.6, statistic stat_poly_eq() can optionally generate character strings encoded in markdown suitable for geom_richtext() from package ‘ggtext’.

Acknowlegement: This update was encouraged by recent questions at stackoverflow. The tag [ggpmisc] is in use at stackoverflow for questions related to this package.

Documentation web site at includes all help pages, with output from all examples, and vignettes in HTML format.

NOTE: The new version of the package is on its way to CRAN.

Please raise issues concerning bugs or enhancements to this package through Bitbucket

learnrbook 1.0.1

Package documentation web site at:

This is the first version submitted to CRAN for the book as published in the R Series.

Versions starting from 1.0.0 are for the book as published in the R Series. Earlier versions were for various partial drafts of the book, as pre-published through LeanPub. The book was published a few of weeks ago, but shipping has started in the last few days.

NOTE: The updated package is on its way to CRAN.

Please raise issues concerning bugs or enhancements to this package through Bitbucket

photobiology 0.10.5

Package documentation web site at:

This update adds a new attribute to objects of class response_spct to enable storage of metadata to distinguish action spectra from response spectra. All other changes are tweaks to make use easier or are minor bug fixes, which do not add important new functionality.

Changes from version 0.10.4 the most recent CRAN release, are:


  • Implement attribute "response.type" to distinguish between response spectra and action spectra stored in response_spct objects.
  • Add methods setResponseType() and getResponseType().
  • Add method drop_user_cols() to remove user-defined columns from spectra.
  • Add method collect2mspct() and rename method uncollect() into uncollect2spct().
  • Add convenience function spct_metadata() to query the value of metadata attributes.
  • Revise add_attr2tb() expanding support to all metadata attributes.
  • Revise smooth_spct() methods adding new parameter wl.range.
  • Revise compare_spct() function to accept scaled and normalized spectra with a warning (instead of triggering an error).


  • Make the pre-computed color data private.
  • Revise rbindspct() to gracefully handle duplicate member names in its input.
  • Revise smooth_spct() methods to fix bug in handling of strength in "custom" method.

NOTE: The updated package is on its way to CRAN.

Please raise issues concerning bugs or enhancements to this package through Bitbucket

Performance of package ‘photobiology’

In recent updates I have been trying to remove performance bottlenecks in the package. For plotting spectra with ‘ggspectra’ an obvious performance bottleneck has been the computation of color definitions from wavelengths. The solution to this problem was to use pre-computed color definitions in the most common case, that of human vision. Many functions and operators as well as assignments were repeatedly checking the validity of spectral data. Depending on the logic of the code, several of these checks were redundant. It is now possible to enable and disable checks both internally and in users’ code. This has been used within the package to avoid redundant or unnecessary checks when the logic of the computations ensures that results are valid.

In addition changes in some of the ‘tidyverse’ packages like ‘tibble’, ‘dplyr’, ‘vctrs’ and ‘rlang’ seem to have also improved performance of ‘photobiology’ very significantly. If we consider the time taken to run the test cases as an indication of performance, the gain has been massive, with runtime decreasing to nearly 1/3 of what it was a few months ago. This happened in spite of an increase in the number of test cases from about 3900 to 4270. Currently the 4270 test cases run on my laptop in 23.4 s. Updates ‘rlang’ (0.4.7) and/or ‘tibble’ (3.0.3) appearing this week in CRAN seem to have reduced runtime by about 30% compared to the previous versions.

The take home message is that even though there is a small risk of package updates breaking existing scripts, there is usually an advantage in keeping your installed R packages and R itself up to date. If some results change after an update it is important to investigate which one is correct, as it is both possible that earlier bugs have been fixed or new ones introduced. When needed it is possible, although slightly more cumbersome, to install superseded versions from the source-package archive at CRAN, which keeps every single version of the packages earlier available through CRAN. With respect to R itself, multiple versions can coexist on the same computer so it is not necessary to uninstall the version currently in use to test another one, either older or newer.