Gabriele Corso
Gabriele Corso
Hi @Jameel9, this is unfortunate. We are actively working on improving the quality of the model on protein + RNA complexes!
Yes, it's definitely possible that they made some extra optimizations. We are working on improving these, but we definitely welcome contributions from the community!
Could you add a raise in the data processing code where that message is printed to see the stack trace?
Hi Dhiray, Unfortunately, we there is no module trained to predict protein-protein affinity currently in the repo, but we are working on it! Gabriele
Hi, we will soon publish code to process structures and train/finetune the model (most likely before the end of next week)
Hi @holgerroth we recently published the code and instructions to process structures and train the model. Apologies for the delay!
Hi @jadolfbr we are currently working through a couple of such fine-tuning pilot projects, that we will use to draw clear guidance on how to do fine-tuning best. Right now...
Hi @Vincentx15, are you sure that Boltz didn't use a preprocessed config? Can you run the prediction in a fresh folder or make sure that you delete the preprocessed files?
Hi @nguyenbinhchem, thank you for pointing this out. Yes, there is a limit above which the code currently crashes, we'll work on fixing this. But in general the model should...
Unfortunately, Boltz-2 has not really been trained or tested to predict affinity in this setting so I cannot really comment on its expected performance. However, we are definitely working on...