Conversation
This comment was marked as off-topic.
This comment was marked as off-topic.
|
Latest version is a tool release. Template needs to be manually updated to satisfy nf-core-bot
tlitfin
left a comment
There was a problem hiding this comment.
I have requested minor changes.
I am also a bit conflicted about whether it is best to:
a) Only support mode-specific flags (so there is no confusion if using the global flag for a non-supported mode). We do expose max-template-date as a mode specific flag only which is a similar precedent.
b) Only support global flag (to avoid creating many additional pipeline params since users will never want to set different mode-level seeds).
c) Current behaviour with global flag (for convenience) + mode-specific flags. Do the mode specific flags add value here?
My gut feel is a slight preference for b > a > c. Not sure if others have a strong opinion here.
We should probably review the way we define these shared parameters to make sure they are consistent (eg max-template-date - although this can probably be deprecated as a pipeline parameter and hard-coded to future date in ext.args).
This pull request adds support for explicit random seed control across the
alphafold2,boltz, andcolabfoldmodules, improving reproducibility for stochastic inference. Users can now set a global--random_seedparameter, or override it with mode-specific parameters. Documentation and schema files have been updated to reflect these new options.boltz_random_seed, andcolabfold_random_seedparameters in bothnextflow.configandnextflow_schema.json. These override the global seed when specified.random_seedparameter tonextflow.configandnextflow_schema.json, allowing users to set a global seed for all supported modules. Each module now uses its mode-specific seed if provided, otherwise falls back to the global seed (random_seed).PR checklist
nf-core pipelines lint).nextflow run . -profile test,docker --outdir <OUTDIR>).docs/usage.mdis updated.CHANGELOG.mdis updated.