MCView maintains a docker image with all the R and python dependencies. By default, the image starts an app with the project mounted at /project in the docker fs, but you can also use it to create projects or import datasets.
Pull the image
Pull the MCView docker image using:
docker pull tanaylab/mcview:latest
Create a project
mkdir PBMC
docker run -v $(pwd)/PBMC:/project tanaylab/mcview R -e 'MCView::create_project("/project/PBMC")'
Import a dataset
Download the PBMC example dataset:
wget http://www.wisdom.weizmann.ac.il/~atanay/metac_data/PBMC_processed.tar.gz
tar xvzf PBMC_processed.tar.gz
Import the dataset:
docker run -v $(pwd):/workdir tanaylab/mcview R -e 'MCView::import_dataset("/workdir/PBMC", dataset = "PBMC163k", anndata_file = "/workdir/metacells.h5ad", cell_type_colors_file = "/workdir/cluster-colors.csv")'
Run the app
docker run -v $(pwd)/PBMC:/project -p 3838:3838 tanaylab/mcview
We can now go to port 3838 on our machine to view the running app.