This is a custom Dockerfile for building an image of EggNogMapper.
I have no affiliation with EggNogMapper or the development team. I am just a user.
This image builds in ~30min and is 93.1GB large.
It contains only the base database required to run emapper with Diamond2. i.e. the Dockerfile runs download_eggnog_data.py -y
These are the database versions:
- emapper-2.1.13
- eggNOG DB version: 5.0.2
- diamond version: 2.1.13
- MMseqs2 version: 18.8cc5c
- Compatible novel families DB version: 1.0.1
The databases are placed in
/data/eggnog-mapper-datawhich is available in the environment variableEGGNOG_DATA_DIR. I.e. no need to specify--data_dir
If you want to use hmmr or other advanced features, you may have to build the database yourself and supply it as a bind mount accessible via --data_dir.
Download the pre-built image from Dockerhub. Additional tags may be added sporadically.
Run with
docker run --rm marklyng/emapper:2.1.13 emapper.py [ARGS] like stated in https://docs.antismash.secondarymetabolites.org/command_line/
Use --rm to stop and remove the container once it has finished running https://docs.docker.com/reference/cli/docker/container/rm/
#E.g.
docker run --rm marklyng/emapper:2.1.13 emapper.py -h
You can bind mount directories as seen fit. Again, databases are not necessary - they are included in the container. https://docs.docker.com/engine/storage/bind-mounts/
#E.g.
docker run --rm \
-v /path/to/input:/data \
-v /path/to/output:/output \
marklyng/emapper:2.1.13 \
emapper.py \
-i /data/input.faa \
--cpu 8 \
--sensmode sensitive \
--excel \
--output-dir /output
See https://github.com/eggnogdb/eggnog-mapper/ for actual information about the software.