MedChem Radar is a commented watch on medicinal chemistry advances in drug discovery: one standout paper at a time, and what it changes for the working medicinal chemist.
About me
I’m Jérôme Meneyrol, PhD, a medicinal chemist and drug hunter with more than 23 years in small-molecule drug discovery, from hit identification to clinical candidate nomination. I have contributed to four clinical candidates, co-authored ten peer-reviewed papers, four of them in the Journal of Medicinal Chemistry, and I am co-inventor on seven patents.
My work sits where medicinal chemistry meets data and AI: structure- and ligand-based design, AI/ML property and activity models, and AI agents that help teams decide faster in the design–make–test–learn cycle. I have worked on most target classes and therapeutic areas, including small molecules targeting RNA, and molecular glues and degraders. I lead mixed teams of medicinal chemists, data scientists and computational chemists, and I train and mentor chemists in advanced medicinal chemistry.
Two ideas I stand for
D²MTL (Design–Decide–Make–Test–Learn). I created this design methodology, which adds an explicit, data-driven decision step to the classic design–make–test cycle. It is described in Real-world applications and experiences of AI/ML deployment for drug discovery (J. Med. Chem. 2025, 68, 851–859).
Every Compound a Candidate. With colleagues, I argued for experience-led risk-taking to reach drug candidates faster, treating every compound made as a potential candidate. Read it in Drug Discovery Today 2025, 30, 104354.
PhD in therapeutic chemistry (Université Paris V René Descartes) · Chemical engineer (ENSC Mulhouse).
Posts reflect my personal views only. They are based on public literature and contain no confidential or partner data.