BASC: barcoded single-cell data processing pipeline.

BASC is a snakemake workflow for processing single-cell chromatin data containing multiple barcodes per sequenced fragment. Most single-cell chromatin datasets contain a single barcode (the cell barcode), and can be handled by other processing pipelines such as cellranger-atac, or simpler methods. This workflow is designed for experiments where there are 1 or 2 additional barcodes sequenced for each DNA fragment, which may encode different information such as the data modality measured by the DNA fragment, droplet combinatorial indexing well of origin, or both.

Installation

The workflow is implemented using Snakemake, and contains several dependencies. These dependencies can be easily installed using conda (or mamba).

To get started, clone the git repository:

git clone [email protected]:timoast/basc.git

To install dependencies in a new conda environment:

# using conda
conda env create -f env.yaml

# using mamba
mamba env create -f env.yaml

Configuring the BASC pipeline

To configure BASC, create a config file containing the barcode combinations for each sample. See config.example for the list of required parameters in the config file. Alternatively, individual configuration parameters can be passed to snakemake on the command line, for example:

snakemake --config reference=/path/to/reference name=NTT

Create a tab-delimited file describing the barcode sequences that were used for each sample or assay. Example samples.tsv file:

Running the workflow

To execute the snakemake workflow first activate the conda environment containing all of the dependencies: conda activate basc. Next, run the snakemake workflow specifying the configfile parameter.

Setting the -n option will perform a dry-run, and allow you to see what steps will be executed in the workflow without computing anyting. Setting the -j parameter will start the workflow with the desired number of cores.

snakemake --configfile /path/to/config -j 24

See the snakemake documentation for more information about running snakemake and all the available options.

Workflow steps

The BASC pipeline executes the following steps:

1. Barcode FASTA files are generated from the config file
2. Combinatorial barcode demultiplexing using a custom python script
3. Map reads to the genome using bwa-mem2
4. BAM file sorted and indexed using samtools
5. Fragment file created using sinto