Hello. Reviving another close-for-a-bit BioStars thread to share that I made a DNA/RNA folding library that has a function for something like this. The library, seqfold, is based on the same algo as mfold/unafold, etc, and returns comparable minimum free energies (see: examples). seqfold.dg_cache is the function that generates a cache that could be used for window querying.

Documentation

https://github.com/Lattice-Automation/seqfold

Examples

https://github.com/Lattice-Automation/seqfold/tree/master/examples

Python

The example below includes dg_cache which is the function for generating a 2D matrix for querying windows, as you asked about in your question.

from seqfold import dg, dg_cache

# just returns minimum free energy
dg("GGGAGGTCGTTACATCTGGGTAACACCGGTACTGATCCGGTGACCTCCC", temp = 37.0)  # -12.94

# `dg_cache` returns a 2D array where each (i,j) combination returns the MFE from i to j inclusive
cache: List[List[float]] = dg_cache("GGGAGGTCGTTACATCTGGGTAACACCGGTACTGATCCGGTGACCTCCC")

CLI

seqfold has a CLI but I can't think of an intuitive way to expose this 2D matrix other than to write it to a CSV externally. If anyone wants a windows/cache function exposed through a CLI let me know and I'll add something like CSV export (so this is available through the CLI in addition to the Python interface).