You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I made a branch using "conda:" directives to isolate kaiju's environment, and it runs, but I realized that one of the basic setup steps is tricky to handle if you're doing it from scratch-- you can't easily run kaiju's makeDB.sh unless you have a good grasp of the .snakemake/conda/$hash directory structures and can get the environment created first so you can activate and use it. Alternatively you could always manually create/have a different environment containing kaiju for the initial setup, but that seems like extra work since logic is already in the extension. I'll keep thinking on this... but any other ideas, anybody?
Will address this issue as part of sunbeam-labs/sunbeam#260. I think (as long as it's well-documented) we could actually add a rule to do the database building, if we add an option for which database the user wants to search to the config. Since the kaiju db name differs based on the reference sequence it might require a little wizardry but I think would be easier for users in the long run.
We should rewrite the rules to use their own conda environments, as the dependencies here may conflict with other extensions.
https://snakemake.readthedocs.io/en/stable/snakefiles/deployment.html#integrated-package-management
The text was updated successfully, but these errors were encountered: