sbx_subsample

This is an extension to the Sunbeam pipeline to subsample reads after host decontamination and quality control.

Installing

To install:

sunbeam extend https://github.com/sunbeam-labs/sbx_subsample/

Paramaters for sbx_subsample are automatically added on sunbeam init. If you're installing an extension in a project where you already have a config file, run the following to add the options for your newly added extension to your config (the -i flag means in-place config file modification; remove the -i flag to see the new config in stdout):

sunbeam config update -i sunbeam_config.yml

Make sure to edit your sunbeam_config.yml file to include the number of reads you'd like to sample:

sbx_subsample:
  threads: 4
  num_reads: 100000

See legacy instructions for older Sunbeam versions below.

Running

As this extension replaces an intermediate step of the Sunbeam pipeline, you must run all_subsample before running any step that uses quality-controlled reads (for example: all_annotate, all_classify, all_mapping, etc.).

For example, if you want to annotate contigs built from subsampled reads, first run:

sunbeam run --configfile sunbeam_config.yml --use-conda all_subsample

The --use-conda flag is required to let Snakemake know that you want to use the conda environment(s) included with your extension.

This will leave you with quality-controlled reads, subsampled to the target number. The starting (non-subsampled), decontaminated files will be backed up in sunbeam_output/qc/decontam/full/. Then, run downstream steps as usual:

sunbeam run --configfile sunbeam_config.yml all_annotate

Contents

• requirements.txt specifies the extension's dependencies
• config.yml contains configuration options that can be specified by the user when running an extension
• sbx_template.rules contains the rules (logic/commands run) of the extension

---

Installing (legacy instructions for Sunbeam <3.0)

Clone the repo into your sunbeam extensions/ folder, installing requirements through Conda, and adding the new options to your existing configuration file. Make sure you've installed Sunbeam first!

source activate sunbeam
cd $SUNBEAM_DIR
git clone https://github.com/sunbeam-labs/sbx_subsample/ extensions/sbx_subsample
conda install --file extensions/sbx_subsample/requirements.txt

Add the options to your config file (replace "sunbeam_config.yml" with the name of your config file).

cat $SUNBEAM_DIR/extensions/sbx_subsample/config.yml >> sunbeam_config.yml

Make sure to edit the config file to include the number of reads you'd like to sample.