gaps

[Rhia15]

Hi,

I hope you are well. I have ran YaHs for scaffolding my genomes after a Hifiasm run with no HiC. I read in a previous issue that YaHs adds 200bp gaps as placeholders when Hi-C data suggests that two contigs are next to each other with no real overlaps, however after running YaHs with contigs from hifiasm no HiC, I get 100bp gaps. Are these also arbitrary placeholders? Traditional gap closing tools are also not working on these, do you have any suggestions on getting rid of these? Or would you usually leave these?

Thank you for any advice!

[c-zhou]

Hi @Rhia15,

In genome assembly, gaps filled with 'N's are typically added between contigs to form scaffolds. This indicates that the contigs are adjacent, but the sequence between them is unknown. When scaffolding with YaHS using Hi-C data, it is not possible to infer the sequences between contigs, so gaps are added. You can use gap-filling tools afterward to attempt filling these sequences, but success is not guaranteed in many cases. Leaving the gaps is acceptable unless you aim to produce a T2T (telomere-to-telomere) genome assembly.

There is no strict requirement for gap size. According to the AGP specification, a size of 100 is recommended for gaps of unknown length. For this reason, the default gap size in the newer versions of YaHS is set to 100. If needed, you can customise the gap size using the --gap-size option, which is detailed in the extended help message (yahs -?).

Best,
Chenxi