Passing vcf indices correctly in M2 wdls #4381
Conversation
Codecov Report
@@ Coverage Diff @@
## master #4381 +/- ##
==============================================
- Coverage 79.093% 79.09% -0.003%
Complexity 16664 16664
==============================================
Files 1050 1050
Lines 59698 59698
Branches 9756 9756
==============================================
- Hits 47217 47215 -2
- Misses 8691 8696 +5
+ Partials 3790 3787 -3
|
|
@davidbenjamin I think this affects more than the PoN. |
|
Are you sure that @eitanbanks was specifying |
There was a problem hiding this comment.
Minor change. Also, please see my questions. Back to you @davidbenjamin ... If you address my comments, feel free to merge.
| gatk_override = gatk_override, | ||
| gatk_docker = gatk_docker, | ||
| preemptible_attempts = preemptible_attempts, | ||
| disk_space = m2_per_scatter_size, | ||
| auth = auth | ||
| } | ||
|
|
||
| Float sub_vcf_size = size(M2.output_vcf, "GB") | ||
| Float sub_vcf_size = size(M2.vcf, "GB") |
There was a problem hiding this comment.
I prefer output_vcf or something more descriptive than vcf... raw_vcf?
@LeeTL1220 False is and was the default for all final outputs, but the intermediate scatter vcfs were hard-coded to be compressed. |
This is the bug fix for @eitanbanks's pon. @LeeTL1220 can you review if tests pass? If they fail I will fix ASAP.