Deep learning enables rapid identification of potent DDR1 kinase inhibitors

  1. Photo Alex Zhavoronkov Alex Zhavoronkov
  2. Photo Yan Ivanenkov Yan Ivanenkov
  3. Photo Alex Aliper Alex Aliper
  4. Photo Mark Veselov Mark Veselov
  5. Photo Vladimir Aladinskiy Vladimir Aladinskiy
  6. Photo Anastasiya Aladinskaya Anastasiya Aladinskaya
  7. Photo Victor Terentiev Victor Terentiev
  8. Photo Daniil Polykovskiy Daniil Polykovskiy
  9. Photo Maxim Kuznetsov Maxim Kuznetsov
  10. Photo Arip Asadulaev Arip Asadulaev
  11. Photo Yury Volkov Yury Volkov
  12. Photo Artem Zholus Artem Zholus
  13. Photo Rim Shayakhmetov Rim Shayakhmetov
  14. Photo Alexander Zhebrak Alexander Zhebrak
  15. Photo Lidiya Minaeva Lidiya Minaeva
  16. Photo Bogdan Zagribelnyy Bogdan Zagribelnyy
  17. Photo Lennart H. Lee Lennart H. Lee
  18. Photo Richard Soll Richard Soll
  19. Photo David Madge David Madge
  20. Photo Li Xing Li Xing
  21. Photo Tao Guo Tao Guo
  22. Photo Alán Aspuru-Guzik Alán Aspuru-Guzik

We have developed a deep generative model, generative tensorial reinforcement learning (GENTRL), for de novo small-molecule design. GENTRL optimizes synthetic feasibility, novelty, and biological activity. We used GENTRL to discover potent inhibitors of discoidin domain receptor 1 (DDR1), a kinase target implicated in fibrosis and other diseases, in 21 days. Four compounds were active in biochemical assays, and two were validated in cell-based assays. One lead candidate was tested and demonstrated favorable pharmacokinetics in mice.

papercode