The way I see it, the

igenomes.config

file is meant to be modified so that it match what you actually need in your pipeline

#CE8SSJV3N had a perfect usage for it, because that's why we created it at the beginning

But since #CELEMAGC9 was never updated, #CE8SSJV3N is not using it

#CGFUX04HZ is a good example of what I think people should do with the

igenomes.config

file

meaning, modify it

I know that @Sateesh_Peri has started recently to use it in #CFP55NZ5G

I'm pushing #C025UH66DA8 to use it

but more or less, most of the genomic pipelines mainly use params to directly access references

My main issue with this

igenomes.config

is that it's complex to update

Because no one (with minor exceptions) has an idea about what is actually in #CELEMAGC9

so the plan is 3 fold. 1 - update igenomes with #C073ZDLNC92 the pipeline (currently missing a few tools, but almost ready to go) That way we can already start updating

igenomes.config

if we want to. (but let's be realistic, I do think only a small amount of people will do this). 2 - provide an alternative to the

igenomes.config

that is yaml based, stored in a single separate repo (à la #CGRTMASKY in the idea) and contains all references stored in igenomes (plus additional values, similar to the

igenomes.config

in #CGFUX04HZ), with a schema, that can be versioned and can evolve depending on what has been updated on igenomes. The idea is to keep using the

--genome

key, with a slight modification (

GRCh38

->

Ensembl/GRCh38

or

GATK/GRCh38

). That way it's easier to get access to what is available for any genome. 3 - keep using this system for the reboot That way the transition is easy