It was a pleasure to welcome Paula Torren Peraire in Leuven, where she gave a talk about her work on the effect of single-step retrosynthesis models on multi-step planning performance. AIDD

Time flies! Today marks the last day of Paula Torren Peraire and my #secondment AstraZeneca. We are extremely grateful for the support and guidance our colleagues AstraZeneca provided during our secondment. Special thanks to Samuel Genheden! Next: the AIDD school in #Leuven 💪🏼🎓

Paula Torren Peraire AIDD 3/4 🧵 Found synthesis routes differ between SSM, but some models produce more chemically valid routes. SotA models find a synthesis route for nearly all tested molecules but struggle to find the gold-standard lab route (route acc.) or skip vital reactions (building block acc.)

Our paper is now published in Digital Discovery! 🥳👇

doi.org/10.1039/D3DD00…

Shout-out to my AIDD colleagues Paula Torren Peraire, Samuel Genheden, Jonas Verhoeven, Djork-Arné Clevert, Mike Preuss & Igor Tetko for the great collaboration!

Paula Torren Peraire
presents her work 'Models Matter: The Impact of Single-Step Retrosynthesis on Synthesis Planning' at the ACS meeting in San Francisco American Chemical Society

Paula Torren Paula Torren Peraire introduces differences and challenges of singe and multistep reaction prediction during seminar at the Institute of structural Biology Sattler Lab of Helmholtz Munich | @HelmholtzMunich

Great resource if you're interested in ML methods and their application to drug discovery ai-dd.eu/lectures

Our new preprint is now available! Check it out at arxiv.org/abs/2308.05522. I'm at #ACSFall2023 this week so reach out if you're interested in discussing all things synthesis prediction.

If you're interested in AI and drug discovery, make sure to check it out! Great opportunity!

Excited to be at #NeurIPS2022 this week and to present the work from myself and Alan Kai Hassen at the #AI4Science workshop on Friday! Come by our poster if you're interested in single-step retrosynthesis approaches in a multi-step environment! openreview.net/forum?id=LjdtY…

Mind the Retrosynthesis Gap, today at the AI for Science workshop at #NeurIPS2022 (openreview.net/forum?id=LjdtY…) Come see our poster Room 388 at 12:05pm and 17:10pm! Alan Kai Hassen AIDD

Thrilled to have attended the Helmholtz AI Conference these past few days! Inspiring to hear about novel applications of machine learning in diverse fields and a great opportunity to engage with colleagues in the applied ML space. #HelmholtzAICon AIDD

Paula Torren Peraire AIDD 4/4 🧵 A huge shoutout to our incredible collaborators from AIDD : Samuel Genheden, Jonas Verhoeven, @OkkoClev , Mike Preuss , Igor Tetko. For any questions or deeper insights, catch Paula Torren Peraire , currently present at #ACSFall . Gratitude for the collective effort! 🙌🔬

Nice to see published results for the presentation delivered yesterday by Paula Torren Peraire at #ACS meeting in section 'Past, Present and Future of AI and Predictive Analytics for Chemical Reactions'.

Alan Kai Hassen Digital Discovery Paula Torren Peraire Mike Preuss Congratulations!! A great collaborative effort