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

Publicly available collections of drug-like molecules have grown to comprise10s of billions of possibilities in recent history due to advances in chemicalsynthesis. Traditional methods for identifying ``hit'' molecules from a largecollection of potential drug-like candidates have relied on biophysical theoryto compute approximations to the Gibbs free energy of the binding interactionbetween the drug to its protein target. A major drawback of the approaches isthat they require exceptional computing capabilities to consider for evenrelatively small collections of molecules.  Hyperdimensional Computing (HDC) is a recently proposed learning paradigmthat is able to leverage low-precision binary vector arithmetic to buildefficient representations of the data that can be obtained without the need forgradient-based optimization approaches that are required in many conventionalmachine learning and deep learning approaches. This algorithmic simplicityallows for acceleration in hardware that has been previously demonstrated for arange of application areas. We consider existing HDC approaches for molecularproperty classification and introduce two novel encoding algorithms thatleverage the extended connectivity fingerprint (ECFP) algorithm.  We show that HDC-based inference methods are as much as 90 times moreefficient than more complex representative machine learning methods and achievean acceleration of nearly 9 orders of magnitude as compared to inference withmolecular docking. We demonstrate multiple approaches for the encoding ofmolecular data for HDC and examine their relative performance on a range ofchallenging molecular property prediction and drug-protein bindingclassification tasks. Our work thus motivates further investigation intomolecular representation learning to develop ultra-efficient pre-screeningtools.

HD-Bind: Encoding of Molecular Structure with Low Precision, Hyperdimensional Binary Representations