SpatialGE normalization

This repository contains a WebMeV-compatible tool for running normalization on spatial transcriptomics using the spatialGE R package (https://github.com/FridleyLab/spatialGE).

To run this tool, pull the Docker container (see https://github.com/web-mev/mev-spatialge-normalize/pkgs/container/mev-spatialge-normalize). Change to the directory where your data lives, then run:

docker run -d -v $PWD:/work ghcr.io/web-mev/mev-spatialge-normalize:<tag> \
    Rscript /usr/local/bin/stnormalize.R \
        -f /work/<COUNTS FILE> \
        -c /work/<COORDS FILE> \
        -s <SAMPLE_NAME> \
        -n <METHOD> \
        -o <OUTPUT_PREFIX>

If successful, this should create a normalized expression file on your current working directory.

Note:

• <tag>: This is the fully-qualified reference to the Docker image, which is tagged to correspond to the Git commit hash. This ensures each Docker image is clearly linked with a specific state of the repository.
• /work/<COUNTS FILE>: This is a path to your tab-delimited count file, with genes in rows and barcodes/spots in columns. Since we assume you have changed to the directory containins your raw data and mounted it in the Docker container at /work, we write this as /work/<COUNTS FILE>.
• /work/<COORDS FILE>: This is a path to your tab-delimited coordinates file which has three columns (in order):
  • The barcodes. These barcodes must match those in the count matrix (the column headers there).
  • The x/horizontal position of the spot.
  • The y/vertical position of the spot. As with the counts file, we assume you have changed to the directory containins your raw data and mounted it in the Docker container at /work, we write this as /work/<COORDS FILE>.
• <SAMPLE_NAME> is the name of the sample. We incorporate this name into the name/path of the output normalized expression.
• <METHOD> is either log or SCTransform, verbatim. Case does not matter.
• <OUTPUT_PREFIX> is an optional prefix which will be added to the front of the output file name.