NextFlow pipeline for 10X snRNA-seq data

Dependencies

Singularity (v. 3) and NextFlow (>= v. 20.10.0). Containers with the software for each step are pulled from the Sylabs cloud library (https://cloud.sylabs.io/library).

Configuration

Paths to reference files must be included in the nextflow.config file -- check that file and change paths accordingly. These include:

1. STAR indices (compatible with STAR v. 2.7.9a)
2. GTF files

When launching the pipeline, as shown in the nextflow command below, you'll also need to set the following:

1. The location of the results directory (e.g., --results /path/to/results)
2. The 10X Chromium chemistry version ('V2', 'V3', 'GEX5', or 'multiome'; e.g., `--chemistry multiome``). Note for 5' GEX ('GEX5'), read 1 is currently assumed to include cDNA in addition to the CB, UMI, and adapter; i.e. it's length should be > 39 bp.
3. The location of the barcode whitelist (e.g., --barcode-whitelist /path/to/737K-arc-v1.txt). The whitelists are within this directory but must be unzipped before use. For Chromium 3' GEX v3, this is the 3M-february-2018.txt.gz file; for 3' GEX v2 or 5' GEX v1 and v2, this is the 737K-august-2016.txt.gz file; for multiome, this is 737K-arc-v1.txt.gz)

Lastly, you'll need to include information about each RNA-seq library, including the genome to which it should be mapped, and the paths to the fastq files for each readgroup. Organize this information in a JSON file, as in library-config.json. For each readgroup, the '1' fastq file corresponds to the sequencing read including the UMI and the nucleus index (and, for 5' GEX, some cDNA); the '2' fastq file refers to the sequencing read representing only cDNA. Also, note that the 'genome' attribute is given as a list (because I will be adding the ability to map to multiple genomes, in the case that nuclei from multiple species are mixed together).

Running

Once you have all of the above information, you can run the pipeline as follows (in this case, indicating the path to the results on the command line):

nextflow run -resume -params-file library-config.json --barcode-whitelist /path/to/barcode-whitelist.txt --chemistry multiome --results /path/to/results /path/to/main.nf

Output

• cellbender/*: Cellbender results
• multiqc/fastq/*: multiqc summaries of fastqc results
• multiqc/star/*: multiqc summaries of STAR logs
• prune/*: filtered bam files (duplicates NOT removed)
• qc/*: Per-barcode QC metrics and QC metric plots
• starsolo/*: starsolo output. Count matrices derived using a variety of counting methods (see STAR manual) are in starsolo/{library}/{library}.Solo.out/*