The analysis can be run in a local virtual docker environment. For that, first pull the image by typing:
docker pull aljoshoh/pheb_r3.5.3
Finally, run the image by:
docker run -it -p 3838:3838 aljoshoh/pheb_r3.5.3 /bin/bash
You should now be in the bash of the pheb docker environment (if you are not sure, check for the existence of /.dockerenv). Now type:
start_rstudio 3838
Now you can navigate to 127.0.0.1:3838 in a browser for running a virtual rstudio session. If you want to build your own container, enter the env directory (cd environment) and run docker build -f Dockerfile -t "aljoshoh/pheb_r3.5.3" . Then, the above command can be used to run the generated docker image.
If you are working on Windows 10 you will have to convert the line endings of the two bash scripts. Herefore natigate to
cd .. /usr/bin/and convert the line endings of your 2 bash scripts.
sed -i -e 's/\r$//' start_rstudio.sh sed -i -e 's/\r$//' rstudio_auth.sh
Now you can connect to a rstudio rsever session for costumization and running scripts manually. This will show you a link and password you can use to login on a webbrowser. You can nagivate to the link on a browser and login with username rstudio and the shown password.
Alternatively, you can use conda for installing its requirements. In the desired location, type:
git clone https://github.com/aljoshoh/pheb.git
cd pheb/
condo activate
./environment/make_env.sh
This will install a virtual conda environment called pheb_r3.5.3. It also contains an installation of Rstudio.
The chunks provided in workflow.Rmd can be used to run the different substeps of the analysis.