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

Comprehensive and unambiguous identification of small molecules in complexsamples will revolutionize our understanding of the role of metabolites inbiological systems. Existing and emerging technologies have enabled measurementof chemical properties of molecules in complex mixtures and, in concert, aresensitive enough to resolve even stereoisomers. Despite these experimentaladvances, small molecule identification is inhibited by (i) chemical referencelibraries representing <1% of known molecules, limiting the number of possibleidentifications, and (ii) the lack of a method to generate candidate matchesdirectly from experimental features (i.e. without a library). To this end, wedeveloped a variational autoencoder (VAE) to learn a continuous numerical, orlatent, representation of molecular structure to expand reference libraries forsmall molecule identification. We extended the VAE to include a chemicalproperty decoder, trained as a multitask network, in order to shape the latentrepresentation such that it assembles according to desired chemical properties.The approach is unique in its application to small molecule identification,with its focus on m/z and CCS, paired with its training paradigm, whichinvolved a cascade of transfer learning iterations. This allows the network tolearn as much as possible at each stage, enabling success with progressivelysmaller datasets without overfitting. Once trained, the network can rapidlypredict chemical properties directly from structure, as well as generatecandidate structures with desired chemical properties. Additionally, theability to generate novel molecules along manifolds, defined by chemicalproperty analogues, positions DarkChem as highly useful in a number ofapplication areas, including metabolomics and small molecule identification,drug discovery and design, chemical forensics, and beyond.

Deep learning to generate in silico chemical property libraries and candidate molecules for small molecule identification in complex samples