slide_snake

[UNDER CONSTRUCTION]

Flexible processing for spatial RNA-sequencing data

slide_snake is a snakemake workflow for preprocessing, alignment, and quantification of spatial RNA-sequencing data, designed to work for any commercial or custom platform.

Getting started

• Please see documentation/*.md for detailed info on custom recipes, pipeline details, etc.

Installation:

1. Clone the repository:

git clone https://github.com/mckellardw/slide_snake.git

2. Build base environment w/ mamba:

mamba env create --name slsn --file=envs/slsn.yml
mamba activate slsn

3. Run on the provided test data for Visium, SlideSeq/Seeker, and Patho-DBIT with the default sample sheet (see below for runtime options).

4. Build your own sample sheet (see `docs/1_sam)

How to set up a run:

1. Install and ensure that the executable paths are functioning
2. Build a sample sheet, containing details on each sample you would like to analyze. Be sure to add the path to the SAMPLE_SHEET_PATH variable in configs/config.yaml.
   • Fill out the file paths for the input .fastqs (can pass multiple sequencing runs, illumina and/or ONT data), reference genomes, etc.
   • Add the recipe(s) for each sample, separated by spaces.
3. Comment out any unwanted output files in the target rule in Snakefile
4. Run the snakemake pipeline!

Runtime details

Example local run:

snakemake -k -p --use-conda --conda-frontend mamba -j 56

Example run w/ slurm:

Make sure the slurm plugin is installed first! (it is included in envs/slsn.yml)

snakemake -k -p --nt --use-conda --conda-frontend mamba --executor slurm --workflow-profile profiles/slurm -j 24

Or, my preferred way to run it in the background:

nohup snakemake -k -p --use-conda --conda-frontend mamba -j 36 > snake.log 2>&1 &

Helpful links:

• Barcode download from Curio
• Extract DNB barcode whitelist for StereoSeq with ST_BarcodeMap
  • Use the "mask format change" code mentioned in the README; example:

  ST_BarcodeMap-0.0.1 --in B01807A3.barcodeToPos.h5 --out B01807A3.barcodeToPos.txt --action 3 --thread 24
  

• "Documentation" on kallisto-lr link