How can #MachineLearning & #automation accelerate the understanding and design of complex molecular systems? Have a look at our latest preprint, an interdisciplinary collaboration between 3 teams van Hest Lab Molecular Machine Learning & Brunsveld’s lab💪🏻 Read Andrea Gardin post to know more👇

ELLIS ML4Science workshop co-organised by our PI Francesca Grisoni & https://bsky.app/profile/aidd.bsky.social PI Günter Klambauer was a great event to learn advances in ML & promote https://bsky.app/profile/aidd.bsky.social results: ai-dd.eu/lectures. https://bsky.app/profile/aichemist.bsky.social fellows Dina Khasanova Marcel Hiltscher attend poster session. Berlin eve.

Thanks for the visit and the inspiring talk on #AI for antibiotic discovery, César de la Fuente ! Fascinating research at the intersection of biology, chemistry and machine learning 💪🏻 Huge thanks to Antoni Forner-Cuenca for co-organizing the event and to TU/e Artificial Intelligence Systems Institute for its support!

Over the past 18 months, we spent A LOT of time planning, discussing, writing, fine-tuning, and WAITING. I had such a blast though and couldn't have wished for a more creative, critical (!), and dedicated team! Thrilled beyond belief that all our efforts paid off!

Great news to conclude 2024! I couldn’t think of a better team than the ones and only narf42 and Robert Pollice to work on this exciting project. 🚀 Cool times ahead for the collaboration between TU Eindhoven and University of Groningen!

A Christmas read from Emanuele, both for MD and ML people in drug discovery! As he was leading this journey, I was happy to be on the next seat (both figuratively and literally, because we are deskmates 😛)

The surprising ineffectiveness of molecular dynamics coordinates for predicting bioactivity with machine learning 1. The study challenges the assumption that molecular dynamics (MD)-derived coordinates are superior for machine learning-based bioactivity predictions, revealing

Happy that our “Hitchhiker’s guide” to the universe of chemical language processing is out Digital Discovery. 🚀 👽 We provide guidelines on how to use deep learning to learn the “chemical language” of bioactivity. 🧪 Spearheaded by the one and only Rıza Özçelik 💪🏻 Molecular Machine Learning

If you are doing peptide informatics, check out our 'tidy and tiny' python package! Should be useful especially for new researchers to start building machine learning pipelines. Code, preprint, and documentation are available 🙂

Inconsistency of LLMs in Molecular Representations Neat study on whether LLMs know when IUPAC and SMILES code for the same molecule (this is called "consistency"). Unsuprisingly: NO, LLMs have different representations for the same molecule! P: doi.org/10.26434/chemr…

Are you using generative #DeepLearning for de novo molecule design?🧪 🖥️ Then check out Rıza Özçelik ‘s latest work, where we perform a (super) large scale analysis (~1 B designs!) & find ‘traps’, ‘treasures’ and ‘ways out’ in the jungle of generative drug discovery. 🌴 🐒 👇

"The Jungle of Generative Drug Discovery: Traps, Treasures, and Ways Out" by Rıza Özçelik , Francesca Grisoni Paper: arxiv.org/abs/2501.05457 #machinelearning

Proud of Helena Brinkmann’s first PhD paper on rethinking how SMILES augmentation is performed for generative #DeepLearning! 🚀 Check it out! 👇 📄 chemrxiv.org/engage/chemrxi… 🖥️ github.com/molML/fantasti… Molecular Machine Learning TU Eindhoven

📢📢📢The ML-GUIDE team (Francesca Grisoni, Robert Pollice and myself) are looking for 3 PhD students, passionate about combining the directed evolution of diverse biomolecules with deep learning approaches to develop better (bio)catalysts and drugs! Pls retweet tinyurl.com/3jz97ter

🚨Open PhD positions TU Eindhoven & University of Groningen! Are you intrigued by the potential of ML+directed evolution in drug discovery (Francesca Grisoni), peptide catalysis (Robert Pollice) or biocatalysis (narf42)? Do you like interdisciplinary, cutting-edge research? Then apply!👇

We just preprinted a fresh study on molecular machine learning on OOD molecules 🧠 Using joint modeling, we could detect distribution shifts, estimate prediction reliability, and capture meaningful molecular patterns! doi.org/10.26434/chemr… #AI #chemistry Molecular Machine Learning