Really happy that our work on conditioning diffusion models is published - have a read if you fancy hearing how we tried to encourage the models to generate more physically plausible compounds by training conditionally with a synthetic, distorted molecules!

Digital Discovery@digital-discovery.rsc.org · last yr.

In this week's first featured article Deane et al. introduce an approach for improving generative molecular design. Distorted molecules with quality labels are added to training data, allowing the model to distinguish unfavourable conformations. Read more: doi.org/10.1039/D4DD...