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

Elucidating and accurately predicting the druggability and bioactivities ofmolecules plays a pivotal role in drug design and discovery and remains an openchallenge. Recently, graph neural networks (GNN) have made remarkableadvancements in graph-based molecular property prediction. However, currentgraph-based deep learning methods neglect the hierarchical information ofmolecules and the relationships between feature channels. In this study, wepropose a well-designed hierarchical informative graph neural networksframework (termed HiGNN) for predicting molecular property by utilizing aco-representation learning of molecular graphs and chemically synthesizableBRICS fragments. Furthermore, a plug-and-play feature-wise attention block isfirst designed in HiGNN architecture to adaptively recalibrate atomic featuresafter the message passing phase. Extensive experiments demonstrate that HiGNNachieves state-of-the-art predictive performance on many challenging drugdiscovery-associated benchmark datasets. In addition, we devise amolecule-fragment similarity mechanism to comprehensively investigate theinterpretability of HiGNN model at the subgraph level, indicating that HiGNN asa powerful deep learning tool can help chemists and pharmacists identify thekey components of molecules for designing better molecules with desiredproperties or functions. The source code is publicly available athttps://github.com/idruglab/hignn.

HiGNN: Hierarchical Informative Graph Neural Networks for Molecular Property Prediction Equipped with Feature-Wise Attention