dnaapler

Dnaapler is a simple tool that reorients complete circular microbial genomes.

Quick Start

# creates empty conda environment
conda create -n dnaapler_env

# activates conda environment
conda activate dnaapler_env

# installs dnaapler
conda install -c bioconda dnaapler

# runs dnaapler all 
dnaapler all -i input_mixed_contigs.fasta -o output_directory_path -p my_bacteria_name -t 8

# runs dnaapler all with a gfa file from e.g. Flye, Unicycler or Autocycler
dnaapler all -i assembly.gfa -o output_directory_path -p my_bacteria_name -t 8

If you have a MacOS machine with Apple Silicon (M1/M2/M3/M4) and are having installation issues, please try

conda create --platform osx-64 -n dnaapler_env dnaapler

conda activate dnaapler_env

dnaapler all -i input_mixed_contigs.fasta -o output_directory_path -p my_bacteria_name -t 8

Paper

Dnaapler has been published in JOSS here. If you use Dnaapler in your work, please cite it as follows:


George Bouras, Susanna R. Grigson, Bhavya Papudeshi, Vijini Mallawaarachchi, Michael J. Roach (2024). Dnaapler: A tool to reorient circular microbial genomes. Journal of Open Source Software, 9(93), 5968, https://doi.org/10.21105/joss.05968

Additionally, please consider citing the dependencies where relevant:

Altschul S.F., Gish W., Miller W., Myers E.W., Lipman D.J. Basic local alignment search tool. J Mol Biol. 1990 Oct 5;215(3):403-10. doi: 10.1016/S0022-2836(05)80360-2. PMID: 2231712.

Steinegger M, Söding J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol. 2017 Nov;35(11):1026-1028. doi: 10.1038/nbt.3988.

Larralde, M., (2022). Pyrodigal: Python bindings and interface to Prodigal, an efficient method for gene prediction in prokaryotes. Journal of Open Source Software, 7(72), 4296, https://doi.org/10.21105/joss.04296.

Hyatt, D., Chen, GL., LoCascio, P.F. et al. Prodigal: prokaryotic gene recognition and translation initiation site identification. BMC Bioinformatics 11, 119 (2010). https://doi.org/10.1186/1471-2105-11-119.

v1 and other recent changes

1.3.0

Thanks @mbhall88 for extending the functionality of --ignore
If your input FASTA or GFA is mixed (e.g. has chromosome and plasmids), you can also use dnaapler all, with the option to ignore some contigs with the --ignore parameter. The --ignore parameter accepts either:
1. A file path containing contig names to ignore (one per line)
2. A comma-separated list of contig names (e.g., chr1,chr2,chr3)
3. - to read contig names from stdin (one per line)

1.2.0

Thanks to the one and only @rrwick, Dnaapler now supports the GFA format as input. This was done to ensure support for Ryan’s new bacterial genome assembly tool Autocycler, the successor to Trycycler, but may also be useful if you have GFA files from e.g. Unicycler, Flye, Spades or other assemblers.
- If you run dnaapler with GFA input, you will get a GFA output as well.
- If you run dnaapler with GFA input, only circular contigs will be reoriented
Relaxes the MMSeqs2 dependency to >=13.45111

1.1.0

Adds support for reorienting contigs where the gene of interest spands the contig ends - fixes this issue. Thanks @marade @oschwengers.
- Specifically, this is done by rotating each contig in the input by half the genome length, then running MMseqs2 for both the original and rotated contigs. The MMseqs2 hit with the highest bitscore across the original and rotated contigs will be chosen as the top hit to rotate by, therefor enabling detection of partial hits (on the original contig) that span the contig ends.
This has only been implemented for dnaapler all (this should be the command used by 99% of users).

v1.0

BREAKING CHANGE - dnaapler now uses MMSeqs2 v13.45111 rather than BLAST. You will need to install MMSeqs2 if you upgrade (if you use conda, it should be handled for you). The CLI is identical.
There are 2 reasons for this:
1. Users reported problems installing BLAST on MacOS with Apple Silicon (see e.g. here). MMseqs2 works on all platforms and is dilligently maintained.
2. MMSeqs2 is much much faster than BLAST (what took BLAST a few minutes takes MMSeqs2 seconds). We probably should have written dnaapler with MMseqs2 to begin with. MMSeqs2 v13.45111 was chosen to ensure interoperability with pharokka
The alignment resuls may not be identicial to dnaapler v0.8.1 (i.e. they might find different top hits), but the actual reorientation is likely to be identical (at least in my tests). Please reach out or make an issue if you notice any discrepancies

For example - on my machine (Ubuntu 20.04, Intel i9 13th gen 13900 CPU with 32 threads), for a Staphylococcus aureus genome with 1 small plasmid, dnaapler -i staph.fasta -o staph_dnaapler -t 8 took ~129 seconds wallclock with v0.8.1 using BLAST, while it took ~3 seconds wallclock with v1.0.0 using MMseqs2.

Google Colab Notebooks

If you don’t want to install dnaapler locally, you can run dnaapler all without any code using the Google Colab notebook.

dnaapler
1.3.0
1.2.0
1.1.0
v1.0
Google Colab Notebooks
- Table of Contents
- Description
- Documentation
- Commands
- Installation
  - Conda
  - Pip
- Usage
- Example Usage
- Databases
- Motivation
- Contributing
- Acknowledgements

Description

Dnaapler Figure

dnaapler is a simple python program that takes a single nucleotide input sequence (in FASTA or GFA format), finds the desired start gene using MMseqs2 against an amino acid sequence database, checks that the start codon of this gene is found, and if so, then reorients the chromosome to begin with this gene on the forward strand.

It was originally designed to replicate the reorientation functionality of Unicycler with dnaA, but for for long-read first assembled chromosomes. We have extended it to work with plasmids (dnaapler plasmid) and phages (dnaapler phage), or for any input FASTA or GFA desired with dnaapler custom, dnaapler mystery or dnaapler nearest.

For bacterial chromosomes, dnaapler chromosome should ensure the chromosome breakpoint never interrupts genes or mobile genetic elements like prophages. It is intended to be used with good-quality completed bacterial genomes, generated with methods such as Autocycler, Dragonflye or my own pipeline hybracter.

Additionally, you can also reorient multiple bacterial chromosomes/plasmids/phages at once using the dnaapler bulk subcommand.

If your input FASTA or GFA is mixed (e.g. has chromosome and plasmids), you can also use dnaapler all, with the option to ignore some contigs with the --ignore parameter. The --ignore parameter accepts either:

A file path containing contig names to ignore (one per line)
A comma-separated list of contig names (e.g., chr1,chr2,chr3)
- to read contig names from stdin (one per line)

As of v1, in practice, dnaapler all is the only command you will likely need, as it contains all the functionality of bulk, chromosome, plasmid, phage but with much more flexibility and user-friendliness

When provided with a GFA file, dnaapler will process only circular sequences – those with a single circularising link and no additional links – while leaving all other sequences unchanged. The output format will match the input: FASTA input produces FASTA output, and GFA input produces GFA output.

Documentation

The full documentation for dnaapler can be found here.

Commands

dnaapler all: Reorients 1 or more contigs to begin with any of dnaA, terL, repA or COG1474.
- Practically, this should be the most useful command for most users.
dnaapler chromosome: Reorients your sequence to begin with the dnaA chromosomal replication initiator gene
dnaapler plasmid: Reorients your sequence to begin with the repA plasmid replication initiation gene
dnaapler phage: Reorients your sequence to begin with the terL large terminase subunit gene
dnaapler archaea: Reorients your sequence to begin with the COG1474 archaeal Orc1/cdc6 gene.
dnaapler custom: Reorients your sequence to begin with a custom amino acid FASTA format gene that you specify
dnaapler mystery: Reorients your sequence to begin with a random CDS
dnaapler largest: Reorients your sequence to begin with the largest CDS
dnaapler nearest: Reorients your sequence to begin with the first CDS (nearest to the start). Designed for fixing sequences where a CDS spans the breakpoint.
dnaapler bulk: Reorients multiple contigs to begin with the desired start gene - either dnaA, terL, repA or a custom gene.

Installation

dnaapler requires only MMseqs2 v13.45111 as an external dependency.

Installation from conda is highly recommended as this will install MMseqs2 automatically.

Conda

dnaapler is available on bioconda.

conda install -c bioconda dnaapler

Pip

You can also install dnaapler with pip.

pip install dnaapler

If you install dnaapler with pip, then you will then need to install MMseqs2 v13.45111 separately. It will need to be available in the $PATH or else dnaapler will not work.

Usage

Usage: dnaapler [OPTIONS] COMMAND [ARGS]...

Options:
  -h, --help     Show this message and exit.
  -V, --version  Show the version and exit.

Commands:
  all         Reorients contigs to begin with any of dnaA, repA...
  archaea     Reorients your genome to begin with the archaeal COG1474...
  bulk        Reorients multiple genomes to begin with the same gene
  chromosome  Reorients your genome to begin with the dnaA chromosomal...
  citation    Print the citation(s) for this tool
  custom      Reorients your genome with a custom database
  largest     Reorients your genome the begin with the largest CDS as...
  mystery     Reorients your genome with a random CDS
  nearest     Reorients your genome the begin with the first CDS as...
  phage       Reorients your genome to begin with the terL large...
  plasmid     Reorients your genome to begin with the repA replication...

Usage: dnaapler all [OPTIONS]

Reorients contigs to begin with any of dnaA, repA, terL or archaeal COG1474 Orc1/cdc6

Options:
-h, --help               Show this message and exit.
-V, --version            Show the version and exit.
-i, --input PATH         Path to input file in FASTA or GFA format
                         [required]
-o, --output PATH        Output directory   [default: output.dnaapler]
-t, --threads INTEGER    Number of threads to use with MMseqs2  [default: 1]
-p, --prefix TEXT        Prefix for output files  [default: dnaapler]
-f, --force              Force overwrites the output directory
-e, --evalue TEXT        e value for MMseqs2  [default: 1e-10]
--ignore TEXT            Text file listing contigs (one per row) that are to
                         be ignored OR comma separated list of contig names
                         to ignore OR '-' to read from stdin
-a, --autocomplete TEXT  Choose an option to autocomplete reorientation if
                         MMseqs2 based approach fails. Must be one of: none,
                         mystery, largest, or nearest [default: none]
--seed_value INTEGER     Random seed to ensure reproducibility.  [default:
                         13]

The reoriented output will be {prefix}_reoriented.fasta in the specified output directory. If the input file was in GFA format, then the output will be named {prefix}_reoriented.gfa.

Example Usage

For more detailed example usage, please see the examples section of the documentation.

dnaapler all -i input.fasta -o output_directory_path -p my_genome_name --ignore list_of_contigs_to_ignore.txt

dnaapler all -i input.fasta -o output_directory_path -p my_genome_name --ignore chr1,chr2,chr3

echo -e "chr1\nchr2\nchr3" | dnaapler all -i input.fasta -o output_directory_path -p my_genome_name --ignore -

dnaapler chromosome -i input.fasta -o output_directory_path -p my_bacteria_name -t 8

dnaapler phage -i input.fasta -o output_directory_path -p my_phage_name -t 8

dnaapler plasmid -i input.fasta -o output_directory_path -p my_plasmid_name -t 8

dnaapler archaea -i input.fasta -o output_directory_path -p my_archaea_name -t 8

dnaapler custom -i input.fasta -o output_directory_path -p my_genome_name -t 8 -c my_custom_database_file

dnaapler mystery -i input.fasta -o output_directory_path -p my_genome_name

dnaapler nearest -i input.fasta -o output_directory_path -p my_genome_name

dnaapler largest -i input.fasta -o output_directory_path -p my_genome_name

# to reorient multiple bacterial chromosomes
dnaapler bulk -i input_file_with_multiple_chromosomes.fasta -m chromosome -o output_directory_path -p my_genome_name

Databases

dnaapler chromosome uses 584 proteins downloaded from Swissprot with the query “Chromosomal replication initiator protein DnaA” on 24 May 2023 as its database for dnaA. All hits from the query were also filtered to ensure “GN=dnaA” was included in the header of the FASTA entry.

dnaapler plasmid uses the repA database curated by Ryan Wick in Unicycler.

dnaapler phage uses a terL database curated using PHROGs. All the AA sequences of the 55 phrogs annotated as ‘large terminase subunit’ were downloaded, combined and depduplicated using seqkit seqkit rmdup -s -o terL.faa phrog_terL.faa.

dnaapler archaea uses a database of 403 archaeal COG1474 Orc1/cdc6 genes curated from here.

dnaapler all uses all four databases combined into one.

dnaapler custom uses a custom amino acid FASTA format file that you specify using -c.

The matching is strict - it requires a strong MMseqs2 match (default e-value 1E-10), and the first amino acid of a MMseqs2 hit gene to be identified as Methionine, Valine or Leucine, the 3 most used start codons in bacteria/phages.

For the most commonly studied microbes (ESKAPE pathogens, etc), the dnaA database should suffice.

If you try dnaapler on a more novel or under-studied microbe with a dnaA gene that has little sequence similarity to the database, you may need to provide your own dnaA gene(s) in amino acid FASTA format using dnaapler custom.

After this issue, dnaapler mystery was added. It predicts all ORFs in the input using pyrodigal, then picks a random gene to re-orient your sequence with.

Motivation

I couldn’t get Circlator to work and it is no longer supported.
berokka doesn’t orient chromosomes to begin with dnaa.
After reading Ryan Wick’s masterful bacterial genome assembly tutorial, I realised that it is probably optimal to run 2 polishing steps, once before then once after rotating the chromosome, to ensure the breakpoint is polished. Further, for some “complete” long read bacterial assemblies that didn’t circularise properly, I figured that as long as you have a complete assembly (even if not “circular” as marked as in Flye), polishing after a re-orientation would be likely to circularise the chromosome. A bit like Ryan’s rotate_circular_gfa.py script, without the requirement of strict circularity.
While researching MGEs in S. aureus whole genome sequences, I repeatedly found instances where MGEs were interrupted by the chromosome breakpoint. So I thought I’d add a tool to automate it in my pipeline.
It’s probably good to have all your sequences start at the same location for synteny analyses.

Contributing

If you would like to help improve dnaapler you are very welcome!

For changes to be accepted, they must pass the CI checks.

Please see CONTRIBUTING.md for more details.

Acknowledgements

Thanks to Torsten Seemann, Ryan Wick and the Circlator team for their existing work in the space. Also to Michael Hall, whose repository tbpore we took and adapted a lot of scaffolding code from because he writes really nice code.