Albert CardonaFrom the lab of Roger Guimerà and Marta Sales Pardo:
"CoCoGraph, a collaborative and constrained graph diffusion model capable of generating molecules that are guaranteed to be chemically valid."
"A collaborative constrained graph diffusion model for the generation of realistic synthetic molecules", Ruiz-Botella et al. 2026
https://www.nature.com/articles/s42256-026-01229-5.epdf
Preprint: https://arxiv.org/abs/2505.16365
A collaborative constrained graph diffusion model for the generation of realistic synthetic molecules | Nature Machine Intelligence
Developing new molecular compounds is crucial to address pressing challenges, from health to environmental sustainability. However, exploring the molecular space to discover new molecules is difficult owing to the vastness of the space. Here we introduce CoCoGraph, a collaborative and constrained graph diffusion model capable of generating molecules that are guaranteed to be chemically valid. Thanks to the constraints built into the model and to the collaborative mechanism, CoCoGraph outperforms state-of-the-art approaches on standard benchmarks while being more efficient. Analysis of 36 chemical properties also demonstrates that CoCoGraph generates molecules with distributions more closely matching real molecules than current models. To illustrate the potential of the model, we created a database of 8.2 million synthetically generated molecules, show how this database and CoCoGraph could be used for molecular discovery and conduct a Turing-like test with organic chemistry experts to further assess the plausibility of the generated molecules, and the potential biases and limitations of CoCoGraph. The collaborative constrained graph diffusion model CoCoGraph generates novel molecules that are guaranteed to be valid and more realistic than state-of-the-art outputs, while achieving faster performance with up to an order of magnitude fewer parameters.
Paul Balduf
Charo del Genio
David
N-gated Hacker News
Hacker News
jobRxiv
Prof. R. I. Sujith
NotARobot