Possible haplotigs? Unscaffolded contigs at end?

[Miranda Sherlock]

Hello everyone, 

I am assembling my first genome of a teleost fish, using HiFi and Hi-C reads which I assembled using HiFiasm's integrated assembly. JellyFish indicates genomes was ~900Mb. I scaffolded with juicer and used the merged_nodups.txt for my run_ASM_pipeline.sh

Now I have my scaffolded assembly, and I notice a couple of things that stand out as a bit strange. 

1) There is this 'white cross' that disappears with normalisation by coverage. As this doesn't span the whole chromosome, I thought it wouldn't be a sex chromosome? And as it is not central I guessed not a centromere either. Perhaps this is a haplotig? I did not run purge_haplotigs or purge_dups at any stage.

2) Seems more self explanatory, the (large, 200Mb) 'tail' of low contact contigs that are not included in a chromosome. Should these be retained or skimmed out somehow?

I appreciate these are probably quite basic questions, but I appreciate any pointers here.

Many Thanks,

Miranda

normalised for covg:

Without normalisation:

Thanks,

Miranda