20/20+ can be downloaded on github.

Package requirements


Upgrading the package dependency of probabilistic2020 to version >=1.0.6 is recommended.

Once downloaded, please add the directory to your PATH variable. Because 20/20+ internally uses the random forest package in R, you will both need R and the randomForest library installed. Once R is installed, you can install the random forest package:

> install.packages("randomForest")

If you do not have permission to install randomForest on the system wide R, you can install in your local user directory by creating an ~/.Renviron file as the following:


Where, in this case, the R libraries will be installed in the ~/Rlibs directory.

20/20+ also requires the following python packages:

  • numpy
  • scipy
  • pandas>=0.17
  • scikit-learn
  • rpy2
  • probabilistic2020
  • matplotlib (optional)

To install these packages via pip you can use the following command:

$ pip install -r requirements.txt

Note, the above command won’t install the optional matplotlib package. Matplotlib is used for creating plots, if you want plotting output than issue the following command to install matplotlib:

$ pip install matplotlib

If you want the exact version 20/20+ was tested on use the requirements_dev.txt file and python 2.7. The probabilistic2020 python package is used to generate features used by 20/20+ from the mutations in MAF format.

Check your PATH variable

Make sure that you have add the 20/20+ directory to your PATH variable. If you have done this correctly, the following command should print the location of the script.

$ which