How to get two clear peaks? Heterozygous and Homozygous peaks

[hrpelg]

Hi,

I am ploting my ONT long reads against different assemblies using kat comp, using the k=27 by default and I get a plot like this. First peak is not totally clear. I have increased the k to 31 but still see the same. Any suggestions?

[gonzalogacc]

Hi hrpelg,
the shape of the distribution depends on the quantity/quality of data a lot more than on the k value (for this range of k values).
ONT data always produce this kind of "bad" spectras, if you want 2 clear peaks use a high base by base accuracy data type like illumina pe.
Best
Gonza.-