Given everything I learned since then (and also given that the omnideconv benchmark will be a lot more computationally demanding) I would strongly recommend to use a workflow manager (i.e. Nextflow or Snakemake). They make the pipeline portable and trivial to parallelize.
@kathireinisch I don't know if you already have previous experience with either nextflow or snakemake and would leave it up to you. But it is definitely a very useful skill to learn ;).
I'm on team nextflow, so I cuold better help you with that.
For the immunedeconv benchmark I used R bookdown for building a reproducible pipeline.
Given everything I learned since then (and also given that the omnideconv benchmark will be a lot more computationally demanding) I would strongly recommend to use a workflow manager (i.e. Nextflow or Snakemake). They make the pipeline portable and trivial to parallelize.
@kathireinisch I don't know if you already have previous experience with either nextflow or snakemake and would leave it up to you. But it is definitely a very useful skill to learn ;).
I'm on team nextflow, so I cuold better help you with that.