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/

Zendy Plus

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

Zendy Tools

Zendy Open

Graph Neural Networks (GNNs) have been widely adopted for drug discovery withmolecular graphs. Nevertheless, current GNNs mainly excel in leveragingshort-range interactions (SRI) but struggle to capture long-range interactions(LRI), both of which are crucial for determining molecular properties. Totackle this issue, we propose a method to abstract the collective informationof atomic groups into a few $\textit{Neural Atoms}$ by implicitly projectingthe atoms of a molecular. Specifically, we explicitly exchange the informationamong neural atoms and project them back to the atoms' representations as anenhancement. With this mechanism, neural atoms establish the communicationchannels among distant nodes, effectively reducing the interaction scope ofarbitrary node pairs into a single hop. To provide an inspection of our methodfrom a physical perspective, we reveal its connection to the traditional LRIcalculation method, Ewald Summation. The Neural Atom can enhance GNNs tocapture LRI by approximating the potential LRI of the molecular. We conductextensive experiments on four long-range graph benchmarks, covering graph-leveland link-level tasks on molecular graphs. We achieve up to a 27.32% and 38.27%improvement in the 2D and 3D scenarios, respectively. Empirically, our methodcan be equipped with an arbitrary GNN to help capture LRI. Code and datasetsare publicly available in https://github.com/tmlr-group/NeuralAtom.

Neural Atoms: Propagating Long-range Interaction in Molecular Graphs through Efficient Communication Channel