artemsokolovdh/pancanstem repository overview

⁠PanCanStem: Reproducing mRNAsi from PMID: 29625051

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 container
• artemsokolovdh/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