🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Thrilled to release Boltz-1x! The physical quality of poses was a recurring piece of feedback we received. With Boltz-1x, we use inference-time steering to remove clashes, respect chirality and other physical properties. A few other goodies too. Work led by the amazing Noah Getz

More updates on the cheminfo side. Physical quality for these types of GenAI in cheminfo has been a big problem since then. Kudos the team 🫡

Great!

We are betting on the right horse I guess

🚀🚀🚀

AF3系モデルはGPUメモリーをかなり要求されるので、そのあたりが改善されているとかなりありがたい。

Boltz-1のver.3プレプリントに引用されてました　うれしいね biorxiv.org/content/10.110…

Boltz-1: Democratizing Biomolecular Interaction Modeling biorxiv.org/content/10.110… 🧬🖥️🧪

(1/3) Speaker Spotlight: Gabriele Corso We’re excited to welcome Gabriele Corso @ICLR2025 to MoML 2025 on June 18th — hosted at Mila - Institut québécois d'IA. Gabriele is a PhD student at Massachusetts Institute of Technology (MIT) MIT CSAIL developing ML methods for structural biology and drug discovery.

Hi– I'm defending my PhD thesis next week, Tue May 6, 2 PM ET, in the Stata Center at MIT (32-G449, Kiva). You're all welcome to drop by or join virtually if you're interested in hearing about my work in (causality ∪ LLMs) ∩ (molecular biology ∪ perturbations)!

Limits of deep-learning-based RNA prediction methods biorxiv.org/content/10.110… #biorxiv_bioinfo

Limits of deep-learning-based RNA prediction methods 1. This study presents a large-scale benchmark of recent deep learning models for RNA structure prediction, evaluating both single-chain RNAs and RNA complexes to identify their strengths and limitations. 2. Among the nine

Very interesting study on RNA and RNA-protein structure prediction! 🧬 Main conclusions: - Boltz-1 and AF3 are the state-of-the-art! - RNA remains a tough challenge and there is significant progress left to be done!

I failed to specify in my original post, but we are running Boltz-1x through Rowan—not Boltz-1. Boltz-1x improves on the speed and performance of Boltz-1, making an amazing technology even better. And now you can try it in seconds with a free Rowan account.

Co-folding, the Future of Docking – Prediction of Allosteric and Orthosteric Ligands １．This study benchmarks emerging protein-ligand co-folding methods for predicting both orthosteric and allosteric binding modes, revealing that current deep learning models still strongly

Nice study from AstraZeneca (EvaNittinger, Özge Yoluk, Alessandro Tibo, Gustav Olanders) on the prediction of allosteric vs orthosteric ligands using a curated set of 17 targets! I've always been curious about the performance of cofolding models on allosteric interactions and