Mikrokondo is a tidy workflow for performing routine bioinformatic tasks like, read pre-processing, assessing contamination, assembly and quality assessment of assemblies. It is easily configurable, provides dynamic dispatch of species specific workflows and produces common outputs.
Mikrokondo takes in either, Illumina, Nanopore or Pacbio data (Pacbio data only partially tested). You can also use mikrokondo for hybrid assemblies or even pass it pre-assembled genomes to handle annotation for you. Additionally, mikrokondo required minimal upfront knowledge of your sample.
Nextflow is required to run mikrokondo, but fortunately it is not too hard to install (Linux is required). The instructions for installing Nextflow can be found at either resource: Nextflow Home or Nextflow Documentation
Nextflow and Mikrokondo only supports running the pipeline using containers such as: Docker, Singularity (now apptainer), podman, gitpod, shifter and charliecloud. Currently only usage with Singularity has been fully tested, (Docker and Apptainer have only been partially tested) but support for each of the container services exists. Note: Singularity was adopted by the Linux Foundation and is now called Apptainer. Singularity still exists, but it is likely newer installs will use Apptainer.
Docker or Singularity (Apptainer) Docker requires root privileges which can can make it a hassle to install on computing clusters (there are work arounds). Apptainer/Singularity does not, so running the pipeline using Apptainer/Singularity is the recommended method for running the pipeline.
Containers are not perfect, below is a list of some issues you may face using containers in mikrokondo, fixes for each issue will be detailed here as they are identified.
- Exit code 137, likely means your docker container used to much memory.
Besides the Nextflow run time (requires Java), and container engine the dependencies required by mikrokondo are fairly minimal requiring only Python 3.10 (more recent Python versions will work as well) to run. Currently mikrokondo has been tested with fully with Singularity (partially with Apptainer, containers all work not all workflow paths tested) and partially tested with Docker (not all workflow paths tested). Dependencies can be installed with Conda (e.g. Nextflow and Python). To download the pipeline run:
git clone https://github.com/phac-nml/mikrokondo.git
- Python (3.10>=)
- Nextflow (22.10.1>=)
- Container service (Docker, Singularity, Apptainer have been tested)
- The source code:
git clone https://github.com/phac-nml/mikrokondo.git
Please check out the documentation for usage instructions here: docs
Under the usage section you can find example commands, instructions for configuration and a reference to a utility script to reduce command line bloat!
- GTDB Mash Sketch: required for speciation and determination if sample is metagenomic
- Decontamination Index: Required for decontamination of reads (it is simply a minimap2 index)
- Kraken2 nt database: Required for binning of metagenommic data and is an alternative to using Mash for speciation
- Bakta database: Running Bakta is optional and there is a light database option, however the full one is recommended. You will have to unzip and un-tar the database for usage. You can skip running Bakta however making the requirement of downloading this database optional.
- StarAMR database: Running StarAMR is optional and requires downloading the StarAMR databases. Also if you wish to avoid downloading the database, the container for StarAMR has a database included which mikrokondo will default to using if one is not specified making this requirement optional.
The above downloadable resources must be updated in the following places in your nextflow.config. The spots to update in the params section of the nextflow.config are listed below:
// Bakta db path, note the quotation marks
bakta {
db = "/PATH/TO/BAKTA/DB"
}
// Decontamination minimap2 index, note the quotation marks
r_contaminants {
mega_mm2_idx = "/PATH/TO/DECONTAMINATION/INDEX"
}
// kraken db path, not the quotation marks
kraken {
db = "/PATH/TO/KRAKEN/DATABASE/"
}
// GTDB Mash sketch, note the quotation marks
mash {
mash_sketch = "/PATH/TO/MASH/SKETCH/"
}
// STARAMR database path, note the quotation marks
// Passing in a StarAMR database is optional if one is not specified the database in the container will be used. You can just leave the db option as null if you do not wish to pass one
staramr {
db = "/PATH/TO/STARMAR/DB"
}
Three test profile with example data are provided and can be run like so:
- Assembly test profile:
nextflow run main.nf -profile test_assembly,<docker/singularity> --outdir <OUTDIR> - Illumina test profile:
nextflow run main.nf -profile test_illumina,<docker/singularity> --outdir <OUTDIR> - Nanopore test profile:
nextflow run main.nf -profile test_nanopore,<docker/singularity> --outdir <OUTDIR> - Pacbio test profile:
nextflow run main.nf -profile test_pacbio,<docker/singularity> --outdir <OUTDIR>- The pacbio workflow has only been partially tested as it fails at Flye due to not enough reads being present.
- Add Shigella detection to GTDB sketch
- allow autodetect of pointfinder db
- Add export of parameters after pipeline runs
- Provide a script for database downloads
- Update details in documentation
- Wait for allele caller
- Upload test data
- Swap Seqtk with Rasusa for down sampling
- Integrate nf-validate
- Trim pilon prefix added to contigs by pilon
An extensive list of references for the tools used by the pipeline can be found in the CITATIONS.md file.
This pipeline uses code and infrastructure developed and maintained by the nf-core community, reused here under the MIT license.
The nf-core framework for community-curated bioinformatics pipelines.
Philip Ewels, Alexander Peltzer, Sven Fillinger, Harshil Patel, Johannes Alneberg, Andreas Wilm, Maxime Ulysse Garcia, Paolo Di Tommaso & Sven Nahnsen.
Nat Biotechnol. 2020 Feb 13. doi: 10.1038/s41587-020-0439-x.