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

Computer-aided molecular design (CAMD) studies quantitativestructure-property relationships and discovers desired molecules usingoptimization algorithms. With the emergence of machine learning models, CAMDscore functions may be replaced by various surrogates to automatically learnthe structure-property relationships. Due to their outstanding performance ongraph domains, graph neural networks (GNNs) have recently appeared frequentlyin CAMD. But using GNNs introduces new optimization challenges. This paperformulates GNNs using mixed-integer programming and then integrates this GNNformulation into the optimization and machine learning toolkit OMLT. Tocharacterize and formulate molecules, we inherit the well-establishedmixed-integer optimization formulation for CAMD and propose symmetry-breakingconstraints to remove symmetric solutions caused by graph isomorphism. In twocase studies, we investigate fragment-based odorant molecular design with morepractical requirements to test the compatibility and performance of ourapproaches.

Augmenting optimization-based molecular design with graph neural networks