Large-scale experimental validation of phenotype-guided generative AI for de novo drug discovery
Fabjan, J.; Wenda, J. M.; Pecoraro-Mercier, C.; Andrea Marin, P.; Wichard, J.; Clevert, D.-A.; Mendez-Lucio, O.; Rouquie, D.
Show abstract
Generative AI is increasingly used to design drugs with desired physicochemical and biological properties. We have previously used the cGAN algorithm trained on cell painting data to generate molecules targeting 10 molecular cancer targets. Here, we provide experimental evidence of the cGAN performance in designing chemical compounds with desired biological activity. We chose a subset of the AI-generated structures for chemical synthesis and tested their activity in a cell painting and a transcriptomic assay. In summary, 88% of the compounds had a pronounced impact on overall cellular morphology, and 37% of the compounds significantly affected transcriptomic pathways associated with the intended target. In a cellular activity assay for an exemplary target, TP53, 6 out of 9 compounds showed potential to modulate TP53 activity. Overall, we show the value of conditioning a generative model on phenotypic readouts for hit discovery of new small molecule drug candidates and pave the way toward AI-guided chemical safety by design.
Matching journals
The top 10 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Tales of 1,008 Small Molecules: Phenomic Profiling through Live-cell Imaging in a Panel of Reporter Cell Lines 96%
- HTRF-based identification of small molecules targeting SARS-CoV-2 E protein interaction with ZO-1 PDZ2 95%
- PKIS Deep Dive Yields a Chemical Starting Point for Dark Kinases and a Cell Active BRSK2 Inhibitor 94%
Similar papers in this journal
Similar papers in this journal
- A repurposed drug screen identifies compounds that inhibit the binding of the COVID-19 spike protein to ACE2 93%
- Machine learning enables accurate and rapid prediction of active molecules against breast cancer cells 91%
- Several FDA-approved drugs effectively inhibit SARS-CoV-2 infection in vitro 91%
Similar papers in this journal
- Repurposing of the RIPK1 selective benzooxazepin-4-one scaffold for the development of a type-III LIMK1/2 inhibitor 95%
- DRUG-seq Provides Unbiased Biological Activity Readouts for Drug Discovery 94%
- Maramycin, a cytotoxic isoquinolinequinone terpenoid produced through heterologous expression of a bifunctional indole prenyltransferase /tryptophan indole-lyase in S. albidoflavus 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.