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

Recently, artificial intelligence for drug discovery has raised increasinginterest in both machine learning and chemistry domains. The fundamentalbuilding block for drug discovery is molecule geometry and thus, the molecule'sgeometrical representation is the main bottleneck to better utilize machinelearning techniques for drug discovery. In this work, we propose a pretrainingmethod for molecule joint auto-encoding (MoleculeJAE). MoleculeJAE can learnboth the 2D bond (topology) and 3D conformation (geometry) information, and adiffusion process model is applied to mimic the augmented trajectories of suchtwo modalities, based on which, MoleculeJAE will learn the inherent chemicalstructure in a self-supervised manner. Thus, the pretrained geometricalrepresentation in MoleculeJAE is expected to benefit downstreamgeometry-related tasks. Empirically, MoleculeJAE proves its effectiveness byreaching state-of-the-art performance on 15 out of 20 tasks by comparing itwith 12 competitive baselines.

Molecule Joint Auto-Encoding: Trajectory Pretraining with 2D and 3D Diffusion