sample-specific information in each sample's BAM file

[aryarm]

new_bam.py should remove sample-specific information from each sample's BAM file and probably replace it with something universal across all of the samples

What do we need to change?

1. The RG tags (both in the header and in each read)
2. The PG tags (both in the header and in each read)
3. The CO tags (from the header only)
4. The read IDs?
5. The header, in its entirety Like, we need to somehow prepare the headers for merging

[aryarm]

My strategy

In new_bam.py

• [x] change the RG tags from each BAM to the same value
• [x] remove the PG tags from the header
  • [x] but make sure that doing this is ok (ie doesn't break the sam format). otherwise, we can just set it to the same value like we do for the RG tags
  • [x] and make sure they also don't appear in each read
• [x] remove the CO tags from the header

  When calling samtools merge

• [x] Pass the -c option to merge RG tags correctly
• [x] Pass the -f option to overwrite previously created files
• [x] Pass the -h option with the first file as an argument so that we just use the first file's header

[zrcjessica]

After making all of these changes, can you still make a reference table mapping the old sample barcodes to these new ones? I think this would make it easier to evaluate the results of demuxlet.

[aryarm]

@zrcjessica isn't that information in the barcodes.tsv files that get generated from the script you're writing? none of that information gets changed by new_bam.py, the script I'm writing

[zrcjessica]

Okay, you're right, that sounds good!