AI Science Tracker

Tencent (China)

1 discovery from 2021

Discoveries

Tencent (China)

A Novel Scalarized Scaffold Hopping Algorithm with Graph-Based Variational Autoencoder for Discovery of JAK1 Inhibitors.

Source check: Source checked. Not independent scientific confirmation.

Finding: GraphGMVAE scaffold hopping from upadacitinib yielded seven synthesized JAK1 inhibitors from 25 prioritized candidates. In biochemical assays, separated Ten01/Ten02 had IC50 values of 5.0/16.8 nM; activity does not establish therapeutic efficacy.

Technology: GraphGMVAE graph variational autoencoder with message-passing encoders

Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.

AI contribution
GraphGMVAE, a graph variational autoencoder with Gaussian-mixture latent space, generated scaffold-hopped molecules from upadacitinib; docking and CNN affinity scoring helped prioritize compounds for synthesis.
Test and result
Seven synthesized compounds were tested in an ADP-Glo JAK1 kinase assay. Ten01 and Ten02 had IC50 values of 5.0 and 16.8 nM after chiral separation; the other isomeric mixtures ranged from 0.50 to 9.6 μM. The Ten01/Ten02 mixture had IC50 64 nM.
Limitation / still unknown
The assays establish biochemical JAK1 inhibition only. The campaign synthesized seven compounds from 25 prioritized molecules; some potency values followed chiral separation and other compounds were tested as isomeric mixtures.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://doi.org/10.1021/acsomega.1c03613

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 18 Nov 2026. Review scope: Full PMC paper, including synthesis/assay methods, results and discussion..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 3/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open the interactive profile for the timeline, or browse all organizations and model families.