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

Molecular design and synthesis planning are two critical steps in the processof molecular discovery that we propose to formulate as a single shared task ofconditional synthetic pathway generation. We report an amortized approach togenerate synthetic pathways as a Markov decision process conditioned on atarget molecular embedding. This approach allows us to conduct synthesisplanning in a bottom-up manner and design synthesizable molecules by decodingfrom optimized conditional codes, demonstrating the potential to solve bothproblems of design and synthesis simultaneously. The approach leverages neuralnetworks to probabilistically model the synthetic trees, one reaction step at atime, according to reactivity rules encoded in a discrete action space ofreaction templates. We train these networks on hundreds of thousands ofartificial pathways generated from a pool of purchasable compounds and a listof expert-curated templates. We validate our method with (a) the recovery ofmolecules using conditional generation, (b) the identification of synthesizablestructural analogs, and (c) the optimization of molecular structures givenoracle functions relevant to drug discovery.

Amortized Tree Generation for Bottom-up Synthesis Planning and Synthesizable Molecular Design