[RivieraKid]

I would guess that there have been many attempts to use ML for protein folding. It's one of the most obvious ways to approach the problem.