Single Cell Analysis Pipeline

1. Download the PPI file and the motif file from GRAND.
2. Download the Waelchli dataset from GEO (GSE256493).
3. Install the SCORPION and NetZoo R packages.
4. Run add_symbols_to_motif.R to add gene symbols to the motif data, changing the following file paths:
   • motifFile: The path to the file with the motif data.
   • motifOutFile: The path to the file where you wish to save the motif data with the symbols.
5. Modify the run_scorpion.R file to include the following file paths:
   • motifFile: The path to the file with the motif data.
   • ppiFile: The path to the file with the PPI data.
6. Run run_scorpion.sh to run all SCORPION models. This will call the run_scorpion.R script for each data file. Change the following file paths as needed:
   • EXPRESSION_DIR: The directory containing the downloaded GEO data.
   • OUTPUT_DIR: The directory where you wish to save your output.
   • LOGGING_DIR: The directory where you wish to save your logs
7. Generate a null PANDA distribution using netZooR::GenerateNullPANDADistribution, setting ppiFile to your PPI file and motifFile to your motif file.
8. Run run_BLOBFISH.R, changing the following variables as needed:
   • sourceDirectory: The directory containing the SCORPION output.
   • resultDirectory: The directory where you wish to save your results.
   • allGenes: The list of target genes of interest.
   • nullFile: The file containing the null PANDA distribution generated in the previous step.
9. Run figures_sox4_new_gene_set_HSPG2_new_immunosuppressive_set.R to generate all plots, changing the following variables:
   • networkDir: The directory where the BLOBFISH results are stored.
   • combMatDir: The directory where the combinatorial matrices used in generating the UpSet plots should be stored.