ggCallaroo: a snakemake worflow of ggCaller, Panaroo and Bakta

A Snakemake pipeline combining graph-based gene prediction with [ggCaller](https://github.com/bacpop/ggCaller), pangenome gene clustering with [Panaroo](https://github.com/gtonkinhill/panaroo), and functional annotation of representative proteins with [Bakta](https://github.com/oschwengers/bakta).

Pipeline overview

Input assemblies
      │
      ▼
 ggCaller  ── gene prediction (GFF + protein FASTA per sample)
      │
      ▼
  Panaroo   ── pangenome clustering → pan_genome_reference.fa
      │
      ▼
 Translation ── nucleotide → amino-acid FASTA (Biopython)
      │
      ▼
bakta_proteins ── functional annotation of representative proteins
      │
      ▼
Annotate Panaroo ── update Panaroo outputs with functional annotation

1. ggCaller – graph-based gene prediction on bacterial genome assemblies.
2. Panaroo – pangenome gene clustering; produces pan_genome_reference.fa (one representative nucleotide sequence per gene cluster).
3. Translation – converts representative nucleotide sequences to amino acids.
4. Bakta – annotates representative proteins with bakta_proteins.
5. Annotate Panaroo - uses Bakta representatives to update annotations in Panaroo outputs

From docker

Installation

Install docker, then run:

docker pull samhorsfield96/ggcallaroo:main

Usage

To run within the container, navigate to the directory containing the FASTA files (e.g. work_dir) and run the below code to generate the input files:

cd work_dir && ls -d -1 $PWD/*.fa > refs.txt && sed "s|^.*/|/data/|" refs.txt > refs_docker.txt

Then run inside the docker container, changing other parameters as required:

docker run -v $(pwd):/output -v $(pwd):/data samhorsfield96/ggcallaroo:main snakemake --cores 4 --use-conda --conda-frontend mamba --config refs=/data/refs_docker.txt ggcaller_cli_args="--save" panaroo_cli_args="--clean-mode moderate" bakta_db=bakta_db/db-light output_dir=/output/results

The results will be generated in results

From source

Installation

Install the required packages using conda/mamba:

git clone https://github.com/samhorsfield96/ggCallaroo.git
cd ggCallaroo
mamba env create -n ggcallaroo "snakemake>=9.19.0" "bakta>=1.12.0"
mamba activate ggcallaroo

Usage

1. Download bakta database

Install Bakta and download the latest bakta database:

bakta_db download --output <output-path> --type [light|full]

Add the bakta_db filepath to the config.yaml file.

2. Prepare a samples file

Create a plain-text file (e.g. input.txt) with one genome assembly FASTA path per line:

/path/to/sample1.fasta
/path/to/sample2.fasta
/path/to/sample3.fasta

You can do this using the command:

ls -d -1 $PWD/*.fasta > input.txt

3. Edit config/config.yaml

Set samples to your samples file, output_dir to the desired results directory, and supply any extra tool arguments via the cli_args fields. Set bakta.db to your local Bakta database path.

4. Run the pipeline

snakemake --cores <N> --use-conda

If using mamba, run instead:

snakemake --cores <N> --use-conda --conda-frontend mamba

Configuration

All tool-specific parameters are controlled via config/config.yaml.

Key	Description
refs or reads	Path to reference genome files or reads FASTA files (one per line)
output_dir	Base output directory
ggcaller_cli_args	CLI arguments passed verbatim to ggcaller
panaroo_cli_args	CLI arguments passed verbatim to panaroo
bakta_db	Path to Bakta database directory

Output

Path	Description
{output_dir}/ggcaller/	ggCaller output (GFF files, protein FASTAs, etc.)
{output_dir}/panaroo/	Panaroo output (pangenome files)
{output_dir}/bakta_proteins/	Bakta annotation output
{output_dir}/annotated/	Directory of annotated Panaroo outputs
{output_dir}/logs/	Log files for each pipeline step