I want to start by saying that I’m new to Docker, and some basic concepts could not be obvious to me.

I’m trying to run a pipeline on Nextflow and use a docker image to use all necessary dependencies to perform the task.
Here is the GitHub link page: wf-artic

I pulled the docker image from Docker Hub, here is the link: docker image

I pulled that image with the following line:

docker pull staphb/artic-ncov2019-epi2me

I checked that the image was pulled correctly with:

docker images
docker inspect "repository_name"

Then I followed the instructions to download and check the integrity of the pipeline from the first link.

When I run the command:

nextflow run epi2me-labs/wf-artic -with-docker staphb/artic-ncov2019-epi2me  
--fastq test_data/reads.fastq.gz 
-profile standard

I got the following error (which is self-explanatory):

--fastq: command not found

So I decided to check if that dependency is present on the docker image with the following commands:

docker run -it --rm staphb/artic-ncov2019-epi2me /bin/bash
which fastq
exit

So far, that dependency is not present on the docker image. So I believe that I need to install it. However, before going further into my understanding the idea of docker is that, has everything (dependencies) on one image so you do not need to use conda for example.

Please correct me if I’m wrong. So my main question of this post is related to the combination of conda environment`` with docker image“`.
Is that recommended?

I want to avoid modifying the docker image, for that reason, I’m thinking of using conda.
Any comments is welcome