The code reproduces mRNA-based stemness indices (mRNAsi) from "Machine Learning Identifies Stemness Features Associated with Oncogenic Dedifferentiation" by Malta, Sokolov, et al.
Prerequisites:
- Install Docker
- Sign up for a synapse.org account and ensure that you have access to:
- syn2701943 - normalized mRNA data from PCBC (used to learn the stemness signature)
- syn4976369 - normalized and batch-corrected PanCan33 data
Download all relevant data with
docker run --rm -v "$PWD":/data artemsokolovdh/pancanstem /app/get-data.R <Synapse Username> '<Synapse Password>'
where
--rm
will clean up the container after it finishes running-v "$PWD":/data
makes the current working directory to be/data/
inside the containerartemsokolovdh/pancanstem
is the container, currently hosted on DockerHub- The source code of
/app/get-data.R
can be found in this GitHub repo - Important:
<Synapse Password>
must be enclosed by quotes to ensure that special characters are handled properly
Once the script finishes, you will find two folders, PCBC
and PanCan33
, containing downloaded data. From here, the stemness signature and indices can be generated by running
docker run --rm -v "$PWD":/data artemsokolovdh/pancanstem /app/main.R