Megan Shand
- Total activity 54
- Last activity
- Member since
- Following 0 users
- Followed by 1 user
- Votes 0
- Subscriptions 15
Activity overview
Latest activity by Megan Shand-
Megan Shand commented,
I don't think it will make a substantial difference for the mitochondria workflow between those two interval lists, but we currently use the wgs_coverage_regions.hg38.interval_list with CollectWgsM...
-
Megan Shand commented,
Hi Rich Roberts, SplitNCigarReads should only be used for variant calling. HaplotypeCaller should be looking at supplementary reads so the read depth should be unaffected. You will want to run Hapl...
-
Megan Shand commented,
Unfortunately a reference sample is required. Ideally we would move away from this requirement, but the work is not prioritized at the moment since the thought was that a reference sample could be ...
-
Megan Shand commented,
Following up here that the docker is now public here: us.gcr.io/broad-dsde-methods/broad-gatk-snapshots:UG_vc_0.8.1_399932 This is a newer version that includes a bug fix and also note that the ou...
-
Megan Shand commented,
Great that you were able to get through some portions of the pipeline! That docker being private is definitely unintentional. I'll make a ticket in the WARP repo and look into that.
-
Megan Shand commented,
When you run ReblockGVCFs you should also include these arguments: "--annotations-to-keep TREE_SCORE --annotations-to-keep ASSEMBLED_HAPS --annotations-to-keep FILTERED_HAPS" in order to keep those...
-
Megan Shand commented,
Perfect, that should fix it. I now see what made ReblockGVCFs so slow, so I'm going to add an error message when the input doesn't contain TREE_SCORE but that tree-score cutoff argument is used. Th...
-
Megan Shand commented,
Ah, so I can confirm that we do run with the tree_score_cutoff argument. But, since you mentioned this, I did notice that you don't have the TREE_SCORE annotation in the gVCFs that you sent earlier...
-
Megan Shand commented,
Hmm, that is definitely not expected. When we run on typical gVCFs with ~90M records it takes about 30 minutes. Did you notice any other tasks that seemed particularly slow? What kind of environmen...
-
Megan Shand commented,
I'm not sure why chrX would be excluded, but I think fixing the sequencing dictionaries to match is the first step. If the tool runs to completion without an error and you're still not getting outp...