Have you thought of the many implications of #LLMs in #Education 🤔? This is one big concern of the community since ChatGPT came out... But there are also many good things, so in our new perspective, we discuss potential opportunities!👇👇 chemrxiv.org/engage/chemrxi…

The final version of 3DReact (née EquiReact) is out! Here: pubs.acs.org/doi/10.1021/ac…. Ksenia Briling Charlotte Bunne @_vsomnath Rubén Laplaza Andreas Krause clemence corminboeuf NCCR Catalysis ! vs the preprint, discussion on InReact vs EquiReact: do we need equivariance for current rxn datasets?

Molecular generative models can *directly* optimize for synthesizability using retrosynthesis models! Check out initial results which can be an alternative to synthesizability-constrained generation Pre-print: arxiv.org/abs/2407.12186… Code: github.com/schwallergroup… (1/2)

Another excellent molecular design study by Jeff (Jeff Guo)! In this work, he shows it's possible to directly optimize for synthesizability in generative molecular design using retrosynthesis models. His approach efficiently generates drug-like molecules optimized for

Our ICML paper: ML + quantum computing + optimal transport = 🚀! Learn distributions conditionally with quantum. Including a 24-qubit hardware run on the assignment problem (predicting doubly-stochastic matrices) IBM Research 📖 ibm.biz/qontot-icml 📹 ibm.biz/qontot-yt

Thrilled to present our new preprint - AdsMT: A multi-modal transformer for predicting global minimum adsorption energy (GMAE). Thanks to my supervisors Philippe Schwaller (he/him) and LIAC group LIAC at EPFL ! preprint: go.shr.lc/4fBntT1 code: github.com/schwallergroup…

Excited to share our preprint on a scalable ML optimisation framework applied to automated HTE reaction optimisation in highly parallel batches! preprint: chemrxiv.org/engage/chemrxi… Thanks to my PI Philippe Schwaller (he/him) (LIAC at EPFL) and Roche for this exciting collaboration! (1/3)

First PhD preprint from Joshua (Joshua W. Sin) together with Roche - excellent work! We are exploring reaction optimization for large batches, such as those typically used in high-throughput experimentation. With this study, we released a dataset of 672 high-quality Ni-catalyzed

Would it not be nice to be able to enforce building blocks in the synthesis routes of generated de novo molecules? To overcome the challenge of a sparse reward landscape, Jeff Guo introduced a new reward function, TANGO (TANimoto Group Overlap). Using TANGO, we demonstrate

Excited to share Atinary's latest collaborative research in #ML on Multi-Fidelity Bayesian Optimization (MFBO)! Our study addresses a key R&D challenge: 🤔 how to balance accuracy vs. cost in R&D experiments 🎯⚖️💲 👉 Article: arxiv.org/html/2410.0054… #selfdrivinglabs #sdlabs

We'll be hosting the "AI & the Molecular World" track at Applied Machine Learning Days. Join us on February 13th to learn more about ML for biology & chemistry. The call for posters is now open! Bruno Correia Philippe Schwaller (he/him) Arne Schneuing Damiano Sgarbossa Jeff Guo Julius Wenckstern 2025.appliedmldays.org/media-34-tracks

FSscore has now been published in Chemistry-Methods (Chemistry Europe)! 🚀 Check out the updated paper on personalized synthesizability scoring here: …mistry-europe.onlinelibrary.wiley.com/doi/10.1002/cm…

We are hiring (resharing appreciated)! Given a few recent successful grant applications (I got my SNSF Starting Grant 🚀), we are extending the LIAC (LIAC at EPFL) team and have multiple openings (PhD/postdoc) for 2025. Are you interested in #AI4Science and machine learning

Join the team ⬇️ #chemtwitter #mljobs #chemjobs

Here's a first: I'm hiring a PhD student at Wageningen U&R. If you have (or know someone who has) a strong background and interest in computational (bio)chemistry or computer science (especially 3D computer vision), let's work together 😊 wur.nl/nl/vacature/ph…

Finally our XpertAI paper is published in Nature Comm Chem!🧪 This work shows how explainable AI can be used for hypothesis generation. This project is extra special because this was my first project at LIAC at EPFL with Philippe Schwaller (he/him) ☺️ nature.com/articles/s4200…

🎉Our BoLudo paper is in J. Am. Chem. Soc.! Bayesian Optimization for nanocrystaL strUcture Design Optimization (jk, it's BOjana & LUDO😂) We show you can optimize ANYTHING when put on the right scale - even nanocrystals with surface energy! Also we discovered a new Cu shape! See how👇

Collaborations with experimental groups are not always straightforward. But this NCCR Catalysis one with Raffaella Buonsanti is exciting and has already yielded excellent results! By introducing a surface energy scale in our Bayesian optimization loop, we discovered a previously

Check out the updated published version of our pre-print in Chemical Science! (1) *General-purpose* generative model + retrosynthesis model = design molecules with optimized properties with an *explicit* predicted synthesis pathway Paper: pubs.rsc.org/en/content/art… (1/2)

Making molecules is hard. How can we simplify the predicted synthesis routes of generated, property‑optimized small molecules? Our last work presents a framework that lets you "mix-and-match" multiple reaction constraints in the synthesis of your generated molecules.