Repository navigation
Store MD, MC, SA and XA in columns of their own - #112
Merged
Merged
Conversation
After #111 the tags left as text were mostly MD, MC, SA and XA, and on chr20 of NA12878 at 30x they still took 118 MB of the 577 MB file in one vector of strings per record. They are now string members of the record, with their place among the other tags kept in tag_order like the other typed tags. Second copies, and these names with another type such as MD:i or XA:A, stay text. The format version becomes 5. With the new columns `ramdump -h` took 15% more CPU on chr20, because writing a typed tag back took four appends: the tab, the name, the type and the value. RAMTagReader now builds the prefix, "\tMD:Z:" and the rest, once per file, and appends it in one go. On that chr20 the file goes from 576.8 to 563.1 MB; the tags take 133.1 MB, 0.8 MB of it text. `ramdump -h` gives the input SAM byte for byte and takes 24.9 to 25.7 s against 27.4 to 27.6 s before, in alternating runs. Region dumps, conversion from SAM on 4 threads and conversion from BAM take the same time within the noise of a loaded machine, and flagstat and idxstats are unchanged. Closes #110.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## develop #112 +/- ##
===========================================
+ Coverage 88.39% 88.47% +0.08%
===========================================
Files 25 25
Lines 2946 2968 +22
Branches 981 991 +10
===========================================
+ Hits 2604 2626 +22
Misses 185 185
Partials 157 157
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
|
clang-tidy review says "All clean, LGTM! 👍" |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
After #111 the tags left as text were mostly MD, MC, SA and XA, and on
chr20 of NA12878 at 30x they still took 118 MB of the 577 MB file in
one vector of strings per record. They are now string members of the
record, with their place among the other tags kept in tag_order like
the other typed tags. Second copies, and these names with another type
such as MD:i or XA:A, stay text. The format version becomes 5.
With the new columns
ramdump -htook 15% more CPU on chr20, becausewriting a typed tag back took four appends: the tab, the name, the
type and the value. RAMTagReader now builds the prefix, "\tMD:Z:" and
the rest, once per file, and appends it in one go.
On that chr20 the file goes from 576.8 to 563.1 MB; the tags take
133.1 MB, 0.8 MB of it text.
ramdump -hgives the input SAM byte forbyte and takes 24.9 to 25.7 s against 27.4 to 27.6 s before, in
alternating runs. Region dumps, conversion from SAM on 4 threads and
conversion from BAM take the same time within the noise of a loaded
machine, and flagstat and idxstats are unchanged.
Closes #110.