DIYABC-RF GUI is currently paused. Feel free to contact us (by submitting an issue) if you want to take over the development for this project. Pull request are also welcome to fix bug or implement missing functionalities. You can still use the command line version of the software which is still maintained/supported. Check the documentation page here. Thanks
Disclaimer:
DIYABC-RF GUIis under final development stage. You may still encounter a few bugs. Please check ongoing issues or fill a new one here if you encounter any problem.
We provide a graphical user interface (GUI) for the DIYABC-RF software [1], called DIYABC-RF GUI.
Note:
DIYABC-RF GUIreplaces the old interfaceDIYABC V2.1which is not maintained anymore.
Please check the project website for additional information and detailed documentation.
DIYABC-RF GUI is available as a standalone application, or as a shiny web app implemented in the diyabcGUI R package .
You can either install and run the standalone app, or install the diyabcGUI R package and run DIYABC-RF GUI as a standard shiny app, c.f. below.
DIYABC-RF GUI provides a set of tools implementing Approximate Bayesian Computation (ABC) combined with supervised machine learning based on Random Forests (RF), for model choice and parameter inference in the context of population genetics analysis.
DIYABC-RF GUI (and the package diyabcGUI) is a user-friendly interface for command-line softwares diyabc and abcranger, which are elementary bricks of the DIYABC-RF pipeline.
The DIYABC-RF GUI software is edited by the DIYABC-RF Core team.
DIYABC-RF Core team: François-David Collin, Ghislain Durif, Louis Raynal, Mathieu Gautier, Renaud Vitalis, Eric Lombaert, Jean-Michel Marin, Arnaud Estoup
The Windows DIYABC-RF GUI standalone app is based on DesktopDeployR by Wyming Lee Pang (https://github.com/wleepang/DesktopDeployR).
See the dedicated file for detailed copyright and licensing information.
zipprogram
For Windows users:
-
Please download the latest release of
DIYABC-RF GUIat https://github.com/diyabc/diyabcGUI/releases/latest and unzipDIYABC-RF_GUI_<latest_version>.zip -
To launch
DIYABC-RF GUI, runDIYABC-RF_GUI.batin the previously extracted directory (either by double-clicking it or in a terminal, you can also create a shortcut to run it by right-clicking on it). -
It will open a new tab in your web browser and you can use
DIYABC-RF GUIas a web app.
Important: you should not forget to quit the app when you are done with the dedicated button (otherwise some background related processes will remains active). Repeat steps 2 and 3 to launch again the application.
A log file for DIYABC-RF GUI is available in your user-specific directory for temporary files, generally
C:\Users\<username>\AppData\Local\Temp\DIYABC-RF_GUI_<timestamp>.log.
If you want to open multiple DIYABC-RF projects, you need to simultaneously open multiple instances of DIYABC-RF GUI (i.e. step 2 and 3).
At the moment, the standalone app is not available for Linux and MacOS users. Nonetheless, Linux and MacOS users can install the diyabcGUI package, c.f. below, and
run the DIYABC-RF GUI as a standard shiny app.
Note: if encountering instability in the standalone app, we recommend to install and use the
shinyapp available in thediyabcGUIR package, c.f. below.
- Install
devtoolspackage (if not installed on your system)
install.packages("devtools")Note: if you encounter any issue when installing
devtools, please check the next section.
- Install
diyabcGUIpackage
devtools::install_github(
"diyabc/diyabcGUI",
subdir = "R-pkg"
)- The first time after installation, you need to download required binary files (e.g.
diyabcandabcrangercommand line tools) by running
library(diyabcGUI)
diyabcGUI::dl_all_latest_bin()Note: you can run this command from time to time to update the required binary files in case new versions were released.
- Launch the interface
library(diyabcGUI)
diyabcGUI::diyabc()The function diyabc() will launch DIYABC-RF GUI as a standard shiny web app, that you will be able to use either in your web browser or in the Rstudio shiny app viewer.
To run simultaneously mutliple instances of DIYABC-RF GUI, e.g. to simultaneously manage and run multiple projects, you just need to run several times the function diyabc() from R (this is not possible from RStudio).
You may encounter some issue when installing devtools, please check the official devtools page.
Following devtools recommandations, make sure you have a working development environment.
- Windows: Install Rtools.
- Mac: Install Xcode from the Mac App Store.
- Linux: Install a compiler and various development libraries (details vary across different flavors of Linux).
For Ubuntu users here is a guide to install devtools requirement (users of other Linux distributions may still find it useful).
As a shiny app, DIYABC-RF GUI can be installed and run from a Shiny server. To do so, you just need (on Unix system, please adapt for Windows server) to:
- install the
diyabcGUIpackage on your system, c.f. above - manage the file access rights so that the Shiny server has access to the R package installation directory
- Create a symbolic link to the directory given by the R command
system.file("application", package = "diyabcGUI")inside thesite_dirfolder configured in/etc/shiny-server/shiny-server.conf(by default/srv/shiny-server), e.g.:
ln -s /path/to/R_LIBS/diyabcGUI/application /srv/shiny-server/diyabc- DIYABC-RF GUI is now available on your server at
https://my.shiny.server.address/diyabc
Please see the dedicated directory for instructions about the standalone building.
[1] Collin F-D, Durif G, Raynal L, Gautier M, Vitalis R, Lombaert E., Marin J-M, Estoup A., 2021, Extending Approximate Bayesian Computation with Supervised Machine Learning to infer demographic history from genetic polymorphisms using DIYABC Random Forest. Molecular Ecology Resources, Wiley/Blackwell, 21(8), pp. 2598–2613. <doi/10.1111/1755-0998.13413> <hal-03229207>