don't sort output

[tedsharpe]

No description provided.

[codecov]

Codecov Report

Merging #8053 (93b072e) into master (19778c1) will decrease coverage by 0.003%.
The diff coverage is 33.333%.

Additional details and impacted files

@@               Coverage Diff               @@
##              master     #8053       +/-   ##
===============================================
- Coverage     86.643%   86.640%   -0.003%     
+ Complexity     38929     38928        -1     
===============================================
  Files           2334      2334               
  Lines         182616    182619        +3     
  Branches       20049     20049               
===============================================
- Hits          158224    158222        -2     
- Misses         17357     17361        +4     
- Partials        7035      7036        +1     

Impacted Files	Coverage Δ
...hellbender/tools/AnalyzeSaturationMutagenesis.java	45.757% <33.333%> (-0.120%)	⬇️
.../hellbender/utils/python/PythonUnitTestRunner.java	75.410% <0.000%> (-3.279%)	⬇️

[mwalker174]

Looks fine to me. Just curious what was the issue with sorted headers? I've had problems with it in the past where the sequences in the header aren't sorted the same as in the reference.

[tedsharpe]

These reads are all from a small ORF that has been systematically mutated. Their order is more or less irrelevant. I was accidentally buffering/sorting them, which just wastes (sometimes a lot) of memory.

The other issue is that the input BAMs typically claim to be queryname sorted, but aren't. (They're sorted as if by -V rather than lexically.) So I can't just pass them through without adjusting the header, because the BAMWriter notices that they aren't actually presorted.