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

Explainable Graph Neural Networks (GNNs) have been developed and applied todrug-protein binding prediction to identify the key chemical structures in adrug that have active interactions with the target proteins. However, the keystructures identified by the current explainable GNN models are typicallychemically invalid. Furthermore, a threshold needs to be manually selected topinpoint the key structures from the rest. To overcome the limitations of thecurrent explainable GNN models, we propose our SLGNN, which stands for usingSparse Learning to Graph Neural Networks. Our SLGNN relies on using achemical-substructure-based graph (where nodes are chemical substructures) torepresent a drug molecule. Furthermore, SLGNN incorporates generalized fussedlasso with message-passing algorithms to identify connected subgraphs that arecritical for the drug-protein binding prediction. Due to the use of thechemical-substructure-based graph, it is guaranteed that any subgraphs in adrug identified by our SLGNN are chemically valid structures. These structurescan be further interpreted as the key chemical structures for the drug to bindto the target protein. We demonstrate the explanatory power of our SLGNN byfirst showing all the key structures identified by our SLGNN are chemicallyvalid. In addition, we illustrate that the key structures identified by ourSLGNN have more predictive power than the key structures identified by thecompeting methods. At last, we use known drug-protein binding data to show thekey structures identified by our SLGNN contain most of the binding sites.

Building explainable graph neural network by sparse learning for the drug-protein binding prediction