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 activity prediction is critical in drug design. Machine learningtechniques such as kernel methods and random forests have been successful forthis task. These models require fixed-size feature vectors as input while themolecules are variable in size and structure. As a result, fixed-sizefingerprint representation is poor in handling substructures for largemolecules. In addition, molecular activity tests, or a so-called BioAssays, arerelatively small in the number of tested molecules due to its complexity. Herewe approach the problem through deep neural networks as they are flexible inmodeling structured data such as grids, sequences and graphs. We train multipleBioAssays using a multi-task learning framework, which combines informationfrom multiple sources to improve the performance of prediction, especially onsmall datasets. We propose Graph Memory Network (GraphMem), a memory-augmentedneural network to model the graph structure in molecules. GraphMem consists ofa recurrent controller coupled with an external memory whose cells dynamicallyinteract and change through a multi-hop reasoning process. Applied to themolecules, the dynamic interactions enable an iterative refinement of therepresentation of molecular graphs with multiple bond types. GraphMem iscapable of jointly training on multiple datasets by using a specific-task queryfed to the controller as an input. We demonstrate the effectiveness of theproposed model for separately and jointly training on more than 100Kmeasurements, spanning across 9 BioAssay activity tests.

Graph Memory Networks for Molecular Activity Prediction