Design

Innovation 1 · Beta

De novo generation of novel, drug-like molecules. Set target property ranges (MW, logP, HBD/HBA) + an optional seed, and Physon.Design proposes 20 new candidates distant from the seed in fingerprint space. Every candidate is RDKit-validated and Lipinski-checked before returning.

Get an API key

Configure constraints on the left, then click Generate designs. Results appear here — one row per candidate with property + novelty metrics.

How it works

SMILES-diffusion model (D3PM) trained on ChEMBL drug-like subset. Reverse-diffuses from an all-mask sequence to a valid molecule, classifier-free-guided by your target property vector.

Novelty scoring

Morgan-fingerprint Tanimoto distance to your seed. Score = 1 - max similarity. 0 = same as seed; 1 = completely novel scaffold.

Guarantees

Every returned SMILES is RDKit-validated (parseable + sanitizable), passes your property filter, and — when Ro5 mode is on — satisfies Lipinski Rule of 5.