English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Inverse molecular design with diffusion models holds great potential foradvancements in material and drug discovery. Despite success in unconditionalmolecule generation, integrating multiple properties such as synthetic scoreand gas permeability as condition constraints into diffusion models remainsunexplored. We introduce multi-conditional diffusion guidance. The proposedTransformer-based denoising model has a condition encoder that learns therepresentations of numerical and categorical conditions. The denoising model,consisting of a structure encoder-decoder, is trained for denoising under therepresentation of conditions. The diffusion process becomes graph-dependent toaccurately estimate graph-related noise in molecules, unlike the previousmodels that focus solely on the marginal distributions of atoms or bonds. Weextensively validate our model for multi-conditional polymer and small moleculegeneration. Results demonstrate our superiority across metrics fromdistribution learning to condition control for molecular properties. An inversepolymer design task for gas separation with feedback from domain expertsfurther demonstrates its practical utility.

Inverse Molecular Design with Multi-Conditional Diffusion Guidance