Revert gCNV WDLs to tar calls from all samples. #5225
Conversation
Codecov Report
@@ Coverage Diff @@
## master #5225 +/- ##
=========================================
Coverage 86.8% 86.8%
Complexity 29713 29713
=========================================
Files 1820 1820
Lines 137406 137406
Branches 15160 15160
=========================================
Hits 119268 119268
Misses 12634 12634
Partials 5504 5504 |
There was a problem hiding this comment.
Thanks for fixing this @slee. Just one a few lines that need need to be deleted, as they are not necessary anymore
scripts/cnv_wdl/cnv_common_tasks.wdl
Outdated
| mkdir -p CALLS_$index/SAMPLE_${sample_index} | ||
| tar xzf $gcnv_calls_tar -C CALLS_$index/SAMPLE_${sample_index} | ||
| mkdir CALLS_$index | ||
| tar xzf $gcnv_calls_tar -C CALLS_$index | ||
| cp ${dollar}{calling_configs_array[$index]} CALLS_$index/ |
There was a problem hiding this comment.
You don't need to pass as input or untar any of these anymore:
Array[File] calling_configs
Array[File] denoising_configs
Array[File] gcnvkernel_version
Array[File] sharded_interval_lists
Look at the diff for cnv_common_tasks.wdl f0dddbc#diff-f3a83ecdd107a7c42aee06dc41842fd2
There was a problem hiding this comment.
Thanks for catching this. I removed them from the PostprocessGermlineCNVCalls task but left them as outputs of the GermlineCNVCaller tasks. Might be useful for debugging and we will need it for when we do address #4397.
There was a problem hiding this comment.
I also added the counts as outputs of the scattered case workflow. Perhaps at some point we can flatten the outputs there. This is trivial for all outputs except the ploidy calls, which need to be uncompressed, concatenated, and compressed again.
samuelklee
force-pushed
the
sl_revert_glob
branch
from
dc57692 to
668d763
Compare
|
Thanks, back to you @asmirnov239. EDIT: Nevermind, I see that you approved. Merging! |
Closes #5217.
@vruano, I've verified that cohort and scattered-case mode WDLs run correctly on FC (to be precise, I tested the WDLs from my sl_filter branch rebased on this branch, but this branch alone should be fine as well).
However, #4397 should still be addressed at some point in the future. This can wait until we have v1 of the FC evaluation in place.