Expose maximum-training-variants VQSR parameter [VS-634] #8029
Conversation
…k in PopulateFilterSetInfo
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## ah_var_store #8029 +/- ##
================================================
Coverage ? 86.219%
Complexity ? 35201
================================================
Files ? 2173
Lines ? 165004
Branches ? 17791
================================================
Hits ? 142265
Misses ? 16405
Partials ? 6334 |
| @@ -332,6 +336,7 @@ task ExtractFilterTask { | |||
| disks: "local-disk 10 HDD" | |||
| bootDiskSizeGb: 15 | |||
| preemptible: 3 | |||
| maxRetries: 3 | |||
There was a problem hiding this comment.
OOC is this related to these changes or just a drive-by improvement?
There was a problem hiding this comment.
I tripped over this on one of the runs testing out different values of sample_every_nth_variant and maximum_training_variants, so not related but it seemed weird to make it its own PR.
| @@ -24,6 +24,7 @@ task SNPsVariantRecalibratorCreateModel { | |||
| Boolean use_allele_specific_annotations | |||
| Int max_gaussians = 6 | |||
| Int sample_every_nth_variant = 1 | |||
| Int maximum_training_variants = 2500000 | |||
There was a problem hiding this comment.
Can there / should there be a comment as to the significance of this value?
There was a problem hiding this comment.
the ticket that's linked to this PR (VS-634) has the context; not sure I want to start the precedent of explaining each parameter inline
There was a problem hiding this comment.
I think it's out of the scope of this pr then, but I do think we want the precedent of explaining each parameter BUT NOT inline. I feel like we have some gestural stuff in the beta user docs and we'll ultimately want it for everything
| @@ -160,6 +162,7 @@ task IndelsVariantRecalibrator { | |||
| File dbsnp_resource_vcf_index | |||
| Boolean use_allele_specific_annotations | |||
| Int max_gaussians = 4 | |||
| Int maximum_training_variants = 2500000 | |||
There was a problem hiding this comment.
same comment here as per SNP analog
Expose the maximum-training-variants parameter for both INDEL and SNP model creation in the
GvsCreateFilterSetWDL.Closes https://broadworkbench.atlassian.net/browse/VS-634