FlowFeatureMapper speed improvements#8982
FlowFeatureMapper speed improvements#8982dror27 wants to merge 14 commits intobroadinstitute:masterfrom
Conversation
…easing cache size
ilyasoifer
left a comment
ilyasoifer
left a comment
There was a problem hiding this comment.
Looks good! Some minor comments
...adinstitute/hellbender/tools/walkers/featuremapping/FlowFeatureMapperArgumentCollection.java
Show resolved
Hide resolved
| /** | ||
| * do not emit edit distance at all | ||
| **/ | ||
| @Argument(fullName = "no-edit-distance", doc = "do not emit edit distance at all ?", optional = true) |
There was a problem hiding this comment.
Do we need this option?
There was a problem hiding this comment.
I imagine that these changes are going to raise some questions. Can you please add description in the PR?
| } | ||
| } | ||
|
|
||
| private FlowBasedRead buildTrimmedFlowBasedRead(final GATKRead read, final TrimInfo trimInfo, final FlowBasedReadUtils.ReadGroupInfo rgInfo) { |
There was a problem hiding this comment.
I think that this code might be duplicated.
HaplotypeCaller routinely trims reads. Could you check if there is a function like that?
There was a problem hiding this comment.
I have refactored this code to be closer in naming convention to its purpose
There was a problem hiding this comment.
Really sorry for being a pain, can you explain why functions from ReadClipper (e.g. SoftClipByReadCoordinates) etc. can't be used for this task?
|
@ilyasoifer I have addressed your comments. Thank you |
|
@ilyasoifer could you please take a look at the two failing CI/tests and see if there is something to be done - or it is a usual/environmental thing? Thanx |
@dror27 - looks like an environment thing (because I see that other PRs also fail), but let's wait for a bit with merging |
2 participants
Additional changes include: