The information in the results is a bit cryptic

[jjuarez84]

It is not super clear why runs terminate. There is some info in the log file, but I think a more detailed output in the results will help gathering statistics to compare binders from non-binders. I don't have a super clear idea how that could work, happy to discuss at some point.

[ggutierrez-bio]

I agree, adding the stopping reason to the report could be interesting.
For comparison between binders and non-binders, i propose using:

• biasing power
• total number of nodes
• any other property related to how big the sampled space is
  We should think of other relevant metrics