AsexStats : an R package for computing stats of asexual genome evolution. Developed specifically for AsexGenomeEvol project

The project is hosted at : https://github.com/AsexGenomeEvol

Content of the package

I will categorize functions by type (project specific will be somehow marked as well)

loading functions

• filter_thetas - function for filtering thetas

plotting functions

• sex_legend - plots a sex/asex legend
• plot_log_hist - plots y-log histogram of sex/asex data

Misc

• ssplit - split a string into vector of substring
• make_data_frame - create a data.frame with predefined column names

data & variables

• timema_pairs - useful descriptors of species pairs (labels, divergence_times, ...)
• timemas - timema species labels stored in a list (codes, labels, ...)
• asex_blue, sex_red - colours used for asexual and sexual species

Install

• get the package (clone or download the latest code)
• open terminal in the folder of the package and run Rscript install.R,

or using devtools package, run R and type

library(devtools)
install_github("AsexGenomeEvol/AsexStats", build_vignettes=FALSE)

local installation

Users on clusters classically do not have a root access and therefore they won't be able to install R packages to the common folders of all the users. To access an R package without root access you need to install the package on some alternative place (like your home) and tell R that there will be an alternative place to look for packages.

The very beginning is the same, download this package, then create a place for your local R packages (/home/$USER/R in this example).

cd ~
git clone https://github.com/AsexGenomeEvol/AsexStats
mkdir -p /home/$USER/R
CMD INSTALL AsexStats -l /home/$USER/R

Now the package was installed to your home (~/R). To tell R about the alternative path you set an environmental variable by adding the line export R_LIBS=/home/$USER/R into your .bashrc file (the setting file of your bash)

echo "export R_LIBS=/home/$USER/R" >> ~/.bashrc

Development

• testing is done using testthat and it is integrated to installation (if test wont pass, installation fails)
• documentation is generated using devtools and roxygen2
• you can use make for rebuilding package - however, the Makefile is not fully portable (yet?), you need to modify path to your R package inside of the makefile.
• use branches and pull requests for development of the new features

Adding new data

• create an R script (*.R) in folder data-raw that will generate a *.rdata data file in data folder
• add the name of data/scripts in R/data.R (use one of previous entries as template)
• add the name to VARIABLES in the Makefile
• document and reinstall package
• before making a commit, check that data look as expected and check if the documentation of data (?mydata) looks as expected