Running Signac gene activity scoring through docker. This is built from the signac vignette. It installs one additional package which is necessary for the vignette to work. I had many difficulties installing the dependencies for building gene activity scores, using this command only requires docker (and sufficient memory).
To run this code use the following command:
docker run -it -v [PATH TO DATA]:/data/ -e RETICULATE_MINICONDA_ENABLED=FALSE atong01/signac-gene-activities:latest Rscript gene_activities.R
[PATH TO DATA] should be a writable path to a folder containing: fragments.tsv.gz, filtered_peak_bc_matrix.h5, and singlecell.csv.
setting the environment variable RETICULATE_MINICONDA_ENABLED=FALSE disables the prompt for install when loading Signac this auto accepts in a script which messes up package versioning. In particular Seurat. rstudio/reticulate#608
You may need to reconfigure the script depending on your specific use case. We provide another script 10X_gene_activities.R that worked for a 10X multiome datasets.
We output the gene activities matrix in a form similar to other Cell Ranger output, in a folder gene_activities_matrx which contains genes.tsv, barcodes.tsv and matrix.tsv, we do this to maintain compatability with other loading software especially for R / Python tools. This will appear in the [PATH TO DATA].
TODO: integrate running the script into the docker command to make this a single command script to get the gene activities matrix.
TODO: accommodate read only data, i.e. separate write folder
TODO: accommodate missing metadata in singlecell.csv
TODO: add quality control plots