plot_position() prints wrong positions about nanocompore HOT 5 CLOSED

I like the last option, I think it make perfect sense.

from nanocompore.

I have looked at this more carefully, and it seems that plot_signal() and plot_position() report the same x-axis labels. The only difference between the two is that plot_position() reports only the first base of each kmer rather than the kmer sequence. This shoulb be fixed.
However, it also seems that the positions reported by both function are off by one relative to the actual sequence...

from nanocompore.

It looks like nanopolish and nanopolishComp are 0 based. @a-slide what do you agree?

from nanocompore.

Yes True. I knew there were a few off by one errors sprinkled in the code... :S
I was about to suggest to report the kmer instead of the base for plot position, but I saw that you implemented the change already.

from nanocompore.

Yes, now all plotting function report the kmer and the 0-based position.
If think it's fine to stick to 0-based coordinates for everything, we just have to make it clear in the docs.
The only thing that remains to decide is what to do when reporting the significant sites in BED format.
At the moment the function annotates the p-value on a single nucleotide (the first one of the kmer, i.e. 'pos'). This is a bit misleading though, because it would make more sense to assign the p-value to the central nt of the kmer...
An alternative (which is my favorite) is to report BED records of kmer size rather than single nucleotide.. what do you think?

from nanocompore.