RIKEN Center for Computational Science Medicinal Chemistry Applied AI Unit

Research Summary

At each stage of drug discovery, "hit to lead" (from initial hits to lead compounds showing in vivo efficacy) stage has become a bottleneck along with phase II clinical trials, and the stagnation of small molecule drug discovery in recent years. This unit aims to improve the time required and cost of the hit-to-lead and lead optimization process (so-called medicinal chemistry process) by AI. The cause of the increased time and cost of the medicinal chemistry process is the difficulty of optimizing potency and ADMET (pharmacokinetics and toxicity) at the same time. In collaboration with Kyoto University and LINC, we are building AI models for new structure generation, synthetic pathway prediction, and potency / ADMET prediction in order to speed up the medicinal chemistry process.

Main Research Fields

• Computer Science

Related Research Fields

• Chemistry
• Pharmacology & Toxicology

Keywords

• In silico screening
• Database
• Artificial intelligence
• Deep learning
• Middle molecule

Selected Publications

Papers with an asterisk(*) are based on research conducted outside of RIKEN.

• 1. Shimizu Y., Ohta M., Ishida S., Terayama K., Osawa M., Honma T., Ikeda K.:
  "AI-driven molecular generation of not-patented pharmaceutical compounds using world open patent data."
  J Cheminform. 2023, 15 (1), 120.
• 2. *Ikeda K., Maezawa Y., Yonezawa T., Shimizu Y., Tashiro T., Kanai S., Sugaya N., Masuda Y., Inoue N., Niimi T., Masuya K., Mizuguchi K., Furuya T., Osawa M.:
  "DLiP-PPI library: An integrated chemical database of small-to-medium-sized molecules targeting protein–protein interactions."
  Frontiers in Chemistry 2023 10, 1090643.
• 3. *Ikeda K., Kezuka Y., Nonaka T., Yonezawa T., Osawa M., Katoh E.:
  "Comprehensive Approach of 19F Nuclear Magnetic Resonance, Enzymatic, and In Silico Methods for Site-Specific Hit Selection and Validation of Fragment Molecules that Inhibit Methionine γ-Lyase Activity."
  J Med Chem. 2021, 64(19):14299-14310.

Related Links

Lab Members

Contact Information