POB_snMultiome_pipeline

This snakemake pipeline was developed for processing snMultiome data and currently under active development. It has been developed solely on Biowulf.

1.Overview

The pipeline currently begins with tared cellranger output, completing per sample quality control, and multiple sample integration.
Overview of snMultiome analysis Pipeline

2.Setup Dependencies

• 1. The R and python used in the pipeline are R4.3.2 and Python 3.9
• 2. Please check if the R package in file workflow/snMultiome_split_v2/scripts/R/main/load_packages.R or python packages in file workflow/snMultiome_split_v2/scripts/python/check_py.py have been installed.
• 3. Please modify the path in workflow/snMultiome_split_v2/config/program.py file.
• 4. Please check if reticulate has been installed and modify the path in workflow/snMultiome_split_v2/scripts/R/main/1.Multiome_preprocess_rawGEX_wSoupX_perSample.R file. reticulate is the R interface to use Python script. Python based software scrublet was used in R script for double removal.

3.Usage

• 1. Preparing Input Manifest File in assets/input_manifest_cellranger.csv
     “uniqueID”, “prefix”, "masterID",“subfolder” are required in the Column 1, 2,3 and last column. All the other fields between masterID and subfolder will be added into seurat object metadata.
     example:
     
     subfolder: is the folder that contains the tared cellranger output. The full path which will be used in the wrapper for looking for the tar files are project_rawdata_dir/subfolder. Please modify the wrapper in order to generate softlink of the tar files.
• 2. Use the wrapper to generate softlink of the tar files, generate config files, copy the scripts to the analysis folder
     Please modify the source path in the wrapper!!!!!

  export PATH=$PATH:/path/to/snMultiomePipeline #this is the path which contain the run_snakemake_sc_v2.py  
  run_snakemake_sc_v2.py -p ccrccdi4 -a N -atacmin Y -mem regular -umapres 0.2 -parallel Y /data/CCRCCDI/rawdata/ccrccdi4 snmultiome hg38 #This step only prepare the analysis folder  
  run_snakemake_sc_v2.py --rerun #This step will start the pipeline   
  

  run_snakemake_sc_v2.py --help

• 3. Here is the detailed documentation

4. References

• This Pipeline was built by Ying Wu based on vigenette from Seurat and Signac
• Incorporate the snMultiome pipeline from POB staff scientist Xiyuan Zhang
• Incorporate several functions from CCBR SINCLAIR pipeline

5. Feedback

For comments/suggestions/advice please reach out to Ying Wu