1. Create a new Julia environment and add the required packages.

mkdir ArgParse_CSV_DataFrames_FileIO
cd ArgParse_CSV_DataFrames_FileIO && julia -e 'using Pkg; Pkg.activate("."); Pkg.add(["ArgParse", "CSV", "DataFrames", "FileIO"]);'

2. Create the Dockerfile like this:

FROM julia:latest
WORKDIR /env
COPY . .
RUN julia -e 'using Pkg; Pkg.activate("."); Pkg.instantiate(); Pkg.precompile()'
COPY entrypoint.sh /entrypoint.sh
RUN chmod +x /entrypoint.sh
ENTRYPOINT ["/entrypoint.sh"]

and modify the entrypoint to reflect the required packages:

#!/bin/bash

# Set JULIA_PROJECT environment variable
export JULIA_PROJECT="/env"

# If no arguments are provided, start an interactive Julia session with the specified packages loaded
if [ "$#" -eq 0 ]; then
    julia -i --eval "using Pkg; Pkg.activate(\"/env\"); using ArgParse; using CSV; using DataFrames; using FileIO;" --banner=no
else
    # Execute the provided command with arguments
    exec "$@"
fi

3. build and push

to build (tag should be the installed packages):

docker build -t plaquette/julia_helper:ArgParse_CSV_DataFrames_FileIO .

to push (tag should be the installed packages)

docker push plaquette/julia_helper:ArgParse_CSV_DataFrames_FileIO  

4. run it

interactiv:

docker run -it plaquette/julia_helper:ArgParse_CSV_DataFrames_FileIO

or with a mounted script:

docker run -it -v "$(pwd):/scripts" plaquette/julia_helper:ArgParse_CSV_DataFrames_FileIO julia /scripts/wrapper.jl --i=/scripts/ref_coli.fa --k=10 --w=100 

add-on, clean up:

run clean-up like this:

 # Remove all stopped containers
docker container prune -f

# Remove <none> images
docker images | awk '/<none>/ {print $3}' | xargs -r docker rmi

add-on, nextflow usage:

in case you want to use this container with nextflow, add it to the config as follows:

singularity.enabled = true
singularity.cacheDir = "$baseDir/src/sif_images"

process {
        conda = 'env_file/kmers.yml'

        withName: 'mash_p' {
                container = 'plaquette/julia_helper:ArgParse_CSV_DataFrames_FileIO'
            }
        }

and activate the environment in the process as follows:

process julia_p {
        
    input:
    tuple path('in.fa'), val(k)

    script:
    """
    
    julia -e 'using Pkg; Pkg.activate(\"/env\"); Pkg.status()'
    ...
    ...
    ...
    
    """
}