Hail the chosen one

Yeah, I dont think it’s a good think to show. Generating crap faster is not a bonus 😅

We could make a lot more trajectories pass if we wanted to, but we on purpose give it roadblocks along the way to make sure only very best interfaces pass. In a sense, every single hallucination iteration is a filtering step, while with diffusion you only have one step to get it right

The main difference are the self consistency steps along the way: 5 multimer models randomly swapped during design, interface must be robust enough to survive MPNN redesign, final AF2 monomer model filtering (most important), etc. Pyrosetta doesnt do much tbh

Congratulations Kiarash, that’s awesome!

Vote for Mosaic too!
proteinbase.com/collections/...

proteinbase.com/collections/...
Also vote for Rita's method, which may be the first MD affinity ranking method I have seen so far that works.

Thanks Nick 🤣 hope so

Gratulujem Hanka!!!