Unmapped Region in Y Chromosome Assembly Using Verkko

[LeoHongboWANG]

Hi,

I am assembling the Y chromosome of my study species using Verkko with the following data:

90x ultra-long ONT reads
100x HiFi reads
200x Hi-C reads
Assembly comparison:
Hifiasm: 5 contigs, total length 43 Mb.
Verkko: Single contig, 47 Mb, with telomeres at both ends.
I am very excited, however, when mapping HiFi and ONT reads back to the Verkko assembly, I found a 10 kb gap in the Y chromosome contig where no reads align. (minimap2 -ax map-ont -t 32 -a -k 19 -O 5,56 -E 4,1 -B 5 -z 400,50 -r 2k --eqx --secondary=no hap1.fa pass.ul.fq.gz >ont.sam)

Questions:
Why do no reads align in this 10 kb region?
Do you know if this additional sequence in Verkko's assembly is accurate?
Thanks for your help!

Best regards,
Hongbo

[skoren]

Could be several reasons including mapping parameters or genome repetitiveness or incorrect/higher error sequence. Since the region is relatively short I suspect verkko would have resolved it correctly since it's also shorter than most HiFi reads. I'd suggest mapping as in the T2T polish pipeline as we've found those alignments to be more reliable and see whether the region is covered then: https://github.com/arangrhie/T2T-Polish.

[LeoHongboWANG]

Hi @skoren,

Thank you for your suggestion! Using the T2T-Polish pipeline, I found some ultralong reads spanning the 10 kb region, but coverage remains sparse. Would this limited support be enough to confirm the region’s accuracy, or could this still be due to this region's repetitiveness or potential sequencing errors?

Thanks again for your help!

Best regards,
Hongbo

[skoren]

The support doesn't look great so I suspect it's an assembly error, the regions probably do go near each other but the sequence in between is not accurate. The T2T polish will flag regions with bad clipping reads/etc and I suspect this will be flagged as such an issue region. I'd recommend flagging this as an issue in the assembly (or updating it have Ns).