ref-match-db: Create mash sketch databases for referenceseeker

Background

Identifying an appropriate reference genome for use in pathogen genomics is an important first step in many analyses. One approach to quickly identify a potentially suitable reference is to use minhash sketches of possible genomes and compare these to minhash sketches of one or more of the samples of interest. However, creating and maintaining a database of minhash sketches for all possible references is both time consumming and space intensive. The size of the database also makes it hard to use in workflows run on the cloud or in other situations where the database must be downloaded on the fly for use during runtime. This tool automates the process of creating a mash sketch databases that are targetted for the organism of interest. By making the databases targetted, they are much smaller and can be easily created on the fly, thus always being up to date.

Here, we combine the tools ncbi-genome-download (Blin 2016) and referenceseeker (Schwengers et al. 2020) to download relevant genomes from NCBI and create a mash sketch database for use with referenceseeker. The workflow was built using the principles, modules, and tools from the nf-core community (Ewels et al. 2020).

Requirements

• Nextflow
• Conda or Docker

Usage

Running a quick test

The test will create a database for Salmonella enterica subsp. enterica serovar Dublin. This shows that the databases can be quite specific.

Run a test with Conda:

nextflow run lgi-onehealth/ref-match-db -profile test,conda

Run a test with Docker:

nextflow run lgi-onehealth/ref-match-db -profile test,docker

Running the pipeline

The primary parameter that most users will be concerned with is the --organisms parameter. This parameter is used to specify the organism of interest. The --organisms parameter is passed directly to ncbi-genome-download --genera parameter and so can be anything that is acceptable for that tool.

For example, to create a database of Neisseria genomes, run:

nextflow run lgi-onehealth/ref-match-db --organisms "Neisseria" -profile docker

The pipeline will create a directory called databases in the current working directory, and in it will create a directory for each organism specified. In each organism directory, there will be referenceseeker database.

Multiple organisms can be specified by separating them with a comma:

nextflow run lgi-onehealth/ref-match-db --organisms "Listeria monocytogenes,Neisseria" -profile docker

This will run the pipeline in parallel for each organism, creating individual databases for each organism. The output will be in the databases directory with the following structure:

databases
├── listeria_monocytogenes
│   ├── db.msh
│   └── db.tsv
└── neisseria
    ├── db.msh
    └── db.tsv

To create databases for specific Salomonella serovars, use the --organisms parameter with the following format:

nextflow run lgi-onehealth/ref-match-db --organisms "Salmonella enterica subsp. enterica serovar Dublin" -profile docker

Workflow parameters

The following parameters are available for the pipeline:

Parameter	Description	Default
--organisms	Organism(s) to create databases for.	Salmonella enterica subsp. enterica serovar Dublin
--outdir	Output directory for the databases.	databases
--assembly-level	Assembly level to download.	complete
--group	Group to download	bacteria

The --assembly-level and --group parameters are passed directly to ncbi-genome-download and so can be anything that is acceptable for that tool. Possible values can be found in the ncbi-genome-download documentation.

References

Blin, K. (2016). ncbi-genome-download (version 0.3.1) [Source code]. https://github.com/kblin/ncbi-genome-download

Ewels, P. A., Peltzer, A., Fillinger, S., Patel, H., Alneberg, J., Wilm, A., Garcia, M. U., Di Tommaso, P., & Nahnsen, S. (2020). The nf-core framework for community-curated bioinformatics pipelines. Nature Biotechnology, 38(3), 276–278. https://doi.org/10.1038/s41587-020-0439-x

Schwengers, O., Hain, T., Chakraborty, T., & Goesmann, A. (2020). ReferenceSeeker: rapid determination of appropriate reference genomes. Journal of Open Source Software, 5(46), 1994. https://doi.org/10.21105/joss.01994