Integrating new Long-Read Mode into HLA-LA

[mayalevy]

I am working at Ultima Genomics, and we've been investigating if HLA-LA genotyping works on our datasets. So far, the results look promising, but we found that HLA-LA does not work on the Ultima data out of the box. Specifically, we produce single-ended short-ish (300 base pair) reads and our error modality is more towards indels than to the mismatches (although at a much lower rate than in ONT or PacBio data).

We were able to run HLA-LA code in the ont2d mode with reducing the length cutoff of the alignment of unpaired reads (minAlignmentLength_unpaired) to 50, but probably would be good to add a specific run mode for using Ultima data.

If you are open to that, we would love to contribute these enhancements back to the main project. Could you advise on the best way to approach merging these changes?

Looking forward to your guidance!

Best regards,
Maya
Ultima Genomics