OV_Histotypes

Classification of Ovarian Cancer Histotypes

Generate RSF

The Report of Statistical Findings (RSF) can be generated by following the steps below:

1. Clone the repository to your local machine and open the project by navigating to the OV_Histotypes.Rproj R project file. RStudio will then be launched.
2. Run renv::restore() to restore all packages used in this project into a private library on your local machine. Note that there are two local packages otta and ottaOvca stored on the UBC TeamShare network drive I (\\files.ubc.ca\team\gpec\i) located at I:/cellar. Ensure you are first connected to UBC MyVPN (myvpn.ubc.ca) if running offsite, otherwise renv::restore() will work if you are connected to the eduroam WiFi network.
3. Restart the RStudio session and the Build tab should now have a Render Book button. Click that button to generate HTML and PDF formats of the RSF, or use the dropdown menu to generate a specific format only.
4. The report files are generated in _book/. From the Terminal, run quarto publish gh-pages to publish the updated RSF to GitHub Pages, hosted on the gh-pages branch.
5. RSF is generated at URL https://talhouklab.github.io/OV_Histotypes/

Classifier Pipeline

The classifier pipeline can be run by following the steps below:

1. Clone the repository to a desired path in a GSC server. We recommend using the gphost11 server as it uses the newest Rocky 9 OS and has R-4.6.0 installed, the same R version configured for this project.
2. Follow Step 2 of [Generate RSF] to install R package dependencies. You'll need to use sftp/scp/rsync to transfer the local packages otta and ottaOvca to the server, and place them at ~/Cellar of your root directory. This path is specified in the environment variable RENV_PATHS_CELLAR in .Renviron.
3. SSH to the numbers server to submit SLURM jobs. This is the job management system that allows us to submit a large number of scripts to worker nodes.
4. Open a new tmux session called histotypes via tmux -S ~/.socket new -s histotypes. Run the pipeline within the tmux session, otherwise the remote SSH session will likely time out after it has been idle for a long time.
5. Navigate to pipeline/README.md to see all the different pipelines and the order in which they should be run using make
6. The outputs will be in the output directory specified in pipeline/assets/params.sh. Parameters in params.sh can be modified as necessary to change inputs to the pipeline.