MD tags with Nanopore data

[leonard-creator]

Hi Hongyu Ding,
I was succesgfully working with MAMnet on small training data. I moved on to use the GIAB OxfordNanopore data. You stated in your README that u tested with that data and the minimap2 aligner. Unfotunately, I ran inbto the following Problem:

I calculated the missing MD tags for the nanopore data with samtools calmd but the program was not able to calculate the MD tags for secondary alignments (flag 256 in Bam-file), because it was missing the SEQ, QUAL columns for those reads. Hence, MAMnet was throwing an error and the program aborted.

Since you were stating you where able to use OxfordNanopore data, what data did you use and how did you use it? Is there a workaround? Can MAMnet handle those errors without crashing?
I tried remapping with minimap2 and the -secondary-seq flag which was not giving the promised results unfortunately.

Bam file data from GIAB

[micahvista]

Dear leonard-creator:
I just updated the code. Now MAMnet allows running on alignments without MD tag.
Hongyu Ding

[leonard-creator]

Thank you very much. I am still interested how you used the nanopore and did not encounter that error. What if I want to include the secondary alignment reads into the analysis?

[micahvista]

Dear leonard-creator:
I updated the code to skip secondary alignment. Recently, I have been studying long read mapping. For my experience, the secondary alignment often led to wrong result. But if you are interested in more challenging regions like repetitive regions, secondary alignment may be the correct one. However, MAMnet is not designed for these challenging regions, and may not benefit from the additional information within secondary alignments.
Hongyu, Ding