GenomeScope Output from PacBio Hifi ccs Reads is Confounding about genomescope HOT 3 OPEN

Now that I've been looking at this a bit, I realize I made a mistake with the jellyfish count run. My value for -s was way off. Although now I'm having a difficult time determining how to calculate a value for this parameter. The documentation says:

The size parameter (given with -s) is an indication of the number k-mers that will be stored in the hash. For sequencing reads, this size should be the size of the genome plus the k-mers generated by sequencing errors. For example, if the error rate is e (e.g.Illumina reads, usually e~1%), with an estimated genome size of G and a coverage of c, the number of expected k-mers is G+Gcek.

So if the estimated genome size is 2.2 Gb and the coverage (for my PacBio Hifi sequencing?) was ~80X, the error rate was 1%, and the k-mer size I'm using is 31, should the value for -s be 2200000000 + 2200000000 * 80 * 1 * 31 or 5458200000000? That can't be right. That seems completely insane...

from genomescope.

Continuing on with this, I reran jellyfish count with -s 50G and got the same results.

Attempting to run jellyfish count with -s 5000G and -s 500G failed due to memory issues, so I'm attempting it with -s 100G. If that doesn't produce something that makes more sense, I'm not entirely sure how to resolve this.

from genomescope.

I'm still getting the same issue. I'd love to know why that might be.

from genomescope.