Lower coverage cutoff higher than upper coverage cutoff about smudgeplot HOT 1 CLOSED

The cutoff program is not trying to be in any way smart. You can take a look how it guesses L and U respectively.

L is 1.25x of the coverage of the first local minimum (we expect that there will be a detectable coverage peak associated with the genome sequencing)
U is 99.8% of the data.

So what likely happens is that you don't have any detectable coverage peak - which means that the first local minimum is actually among really high coverage, higher than 99.8% of your data.

That's the thing. If you have a genome skim (less than 5x of coverage), I am afraid you won't be able to use smudgeplot, it's not designed for such data. You can find more about input here.

k-mers in general however can be useful. You can check what you can do with genome skims on this workshop webpage. Hope this helps.

from smudgeplot.