TomHarrop/container-funannotate

container-funannotate

Docker image for funannotate.

Usage

You can download the container from GHCR and use it with Docker or Singularity (Apptainer):

apptainer exec \
    docker://ghcr.io/tomharrop/container-funannotate:1.8.15_cv1 \
    funannotate check --show-versions

By default, funannotate uses the FUNANNOTATE_DB environment variable to define the path to the funannotate database. You can provide the path to the database with the -d argument to funannotate, or set the environment variable when you run the container

Genemark is installed without a license. You need to get your own license, and bind it to ${HOME}/.gm_key when you run the container.

The following dependencies have issues in funannotate-conda_1.7.4:

• ete3 isn't installed (see here).
• RepeatMasker isn't installed. Use the Dfam-consortium/TETools Docker image to run RepeatMasker separately.
• signalp can't be installed because of licensing issues.
• pslCDnaFilter isn't installed. It only seems to run if blat is used in funannotate predict (see this code)
• emapper.py isn't working because the conda container is missing gcc. Run emapper.py with biocontainers/eggnog-mapper and provide the results to funannotate annotate.

Other containers

The version tags on Singularity.funannotate-base and Singularity.funannotate-deps refer to the version of funannotate they were built for rather than the software in the containers.

Singularity.tetools

Base for funannotate using Dfam's TE Tools container. My singularity recipe pulls the docker image from dfam/tetools, adds trf and makes the RepeatMasker library directory writeable.

Singularity.funannotate-base

Adds the funnanotate dependencies that can be installed from apt, pip and cpan to Singularity.tetools

Singularity.funannotate-deps

Manually installs the remaining dependencies into Singularity.funannotate-base

Singularity.funannotate

Adds funannotate into Singularity.funannotate-deps and sets up environment

Singularity.interproscan

A container recipe to run interproscan outside funannotate. Because the container is 9.3 GB it's not hosted on Singularity Hub. Build it locally, and download and expand the Panther 14.1 database. interproscan looks for Panther data in /interproscan/data/panther/14.1, so bind the path to your downloaded Panther database into the interproscan as follows:

singularity exec \
    --writable-tmpfs \
    -B /path/to/panther:/interproscan/data/panther \
    interproscan_5.44-79.0.sif \
    interproscan.sh \
    -i /interproscan/test_proteins.fasta \
    -f tsv -dp \
    --output-dir test \
    --tempdir temp

You can also use the interproscan biocontainer.