constraint to specify NUPACK/ViennaRNA non-complementary domain binding affinities

[dave-doty]

One can specify domain binding affinities to their perfect complements according to the nearest neighbor duplex model.

But sometimes we want to specify non-zero binding affinities for domains against other domains that are not their perfect complements, e.g.,

https://drops.dagstuhl.de/opus/volltexte/2023/18787/pdf/LIPIcs-DNA-29-4.pdf
https://www.nature.com/articles/s41557-022-01111-y

In general one would specify a binding affinity table stating the desired binding energy between every pair of domains, or more usefully, lower and upper bounds (one or both optional). Perhaps specify it as a dict with pairs of domains as keys, and a key missing means we don't care what is the binding affinity of that pair.