Installation
The pipeline is available as a Docker image, a conda package, or can be installed from source after resolving dependencies via conda.
Via Docker
Images are published to quay.io/artic/fieldbioinformatics for both amd64 and aarch64 in two variants:
| Tag | Clair3 models | Use when |
|---|---|---|
latest / vX.Y.Z |
Not included | You will mount a pre-downloaded model directory |
latest-models-included / vX.Y.Z-models-included |
Bundled | You want a fully self-contained image |
With models pre-bundled
docker pull quay.io/artic/fieldbioinformatics:latest-models-included
docker run --rm \
-v $(pwd):/data \
-w /data \
quay.io/artic/fieldbioinformatics:latest-models-included \
artic minion \
--scheme-name artic-inrb-mpox \
--scheme-version v1.0.0 \
--scheme-length 2500 \
--read-file my_sample.fastq \
my_sample
Without models (bring your own)
Download models first (see Clair3 Models), then mount the directory at run time:
docker pull quay.io/artic/fieldbioinformatics
docker run --rm \
-v $(pwd):/data \
-v /path/to/models:/models \
-w /data \
quay.io/artic/fieldbioinformatics:latest \
artic minion \
--model-dir /models \
--scheme-name artic-inrb-mpox \
--scheme-version v1.0.0 \
--scheme-length 2500 \
--read-file my_sample.fastq \
my_sample
Via conda
conda create -n artic -c bioconda artic
conda activate artic
After installation, download the Clair3 models:
artic_get_models
Via source
1. Install dependencies
The pipeline has several software dependencies. Use the provided conda environment file to resolve them. We strongly recommend conda >= 23.10.0 (which uses the libmamba solver by default):
git clone https://github.com/artic-network/fieldbioinformatics
cd fieldbioinformatics
conda env create -f environment.yml
2. Install the pipeline
conda activate artic
pip install .
3. Download Clair3 models
artic_get_models
4. Verify the installation
Check the pipeline can be called:
artic -v
To verify all dependencies are present, run the pipeline integration tests:
./test-runner.sh clair3
For further tests see Tests.