VantAI

We're thrilled to partner with Halda Therapeutics to expand the reach of their groundbreaking RIPTAC platform. Together, we're working to identify new therapeutic target-effector combinations that will help enable this innovative proximity-based approach to unlock new treatment possibilities across cancer and immune disorders. Read more: http://bit.ly/4monuwV

Neo just got much faster! 💥 Working closely with NVIDIA Healthcare, VantAI adopted #cuEquivariance, leading to massive speedups. Today's announcement is a big advancement for the field! Read more: nvda.ws/4jHJFM6 Try now: https://lnkd.in/ddXqW9dE

🧬 Excited to announce our second expansion in three years with Blueprint Medicines. 🧪 Together, we will prioritize novel target-effector pairs for molecular glue development, aiming at previously undruggable targets, using our proximity modulator discovery platform powered by our Neo-1 foundation model and NeoLink XLMS technology. The collaboration includes up to $1.67B in potential milestone payments. See press release for more details: https://lnkd.in/eF3fB-4K

We are excited to announce the first alpha release of 💥 peppr - package for evaluation of predicted poses 💥 an open source framework for structural model evaluation Find out more at peppr.vant.ai/ built on www.biotite-python.org, peppr dramatically simplifies evaluation for structure prediction and de-novo generation tasks. We built peppr because existing evaluation tools are heavy native binaries with limited cross-platform support and hard to extend. Peppr is built on familiar python and numpy, making it highly extensible without sacrificing speed. Built on biotite primitives, it also makes interfacing with RDKit and Biotite-based workflows much simpler.

Drug discovery turned out to be too hard, so VantAI has made the important decision to pivot from molecular glues to regular glues, and today we are thrilled to announce a landmark $800B biobucks deal with Elmer’s. CTO Luca Naef explains, “When we first prompted Neo-1, our foundation model, we accidentally forgot to specify 'molecular' glues. Instead, it generated formulations for regular glues. Upon testing, we had a eureka moment—the adhesives were outstanding, showcasing the unexpected versatility and power of our AI platform.” Inspired by our Chief Scientist Michael Bronstein love of horses, all VantAI products will be vegan, leveraging geometric deep learning to revolutionize a space untouched by innovation for decades. “Molecular glues were exciting,” says Bronstein, “but the real opportunity is macro-scale. Our original business plans have, as they say, Склеить ласты—they’ve "had their fins glued together", but luckily everyone’s got something they want stuck together - by accurately predicting polymer interactions and curing dynamics, we’re ushering in a new era of computationally engineered adhesives that outperform conventional products in every measurable way.” Traditional bonding agents rely on trial-and-error experimentation, leading to unpredictable adhesion strength, durability, and curing times. VantAI's cutting-edge geometric deep learning platform models adhesive interactions at the molecular level, enabling rational, de novo design of glues optimized for specific substrates, environmental conditions, and application methods. This translates to dramatically improved bond reliability, reduced curing times, and tailored mechanical properties. For inquiries about our innovative adhesive solutions or to explore customized bonding opportunities tailored to your specific needs, please reach out at stickybusiness@vant.ai. Our dedicated team is ready to help you achieve exceptional adhesion outcomes.

Introducing Neo-1: We're thrilled to announce Neo-1—the world's most advanced atomistic foundation model, unifying all-atom structure prediction and de novo generation to decode and design life's molecular structures. Neo-1 employs the latest latent diffusion techniques to reason simultaneously about molecules and their structures, replacing multiple specialist models with one versatile framework. Design: Neo-1 generates diverse, drug-like small molecules and natural proteins from sequences, properties, or partial inputs. This enables completely novel applications, such as rational molecular glue design directly from sequences. Decode: Neo-1 achieves state-of-the-art accuracy in all-atom structure prediction, excelling particularly in complex ternary systems. NeoLink: NeoLink is VantAI's experimental platform using cross-linking mass spectrometry (XLMS) to capture structural constraints across the entire proteome, which Neo-1 integrates into precise atomic models. Programmability: Highly programmable, Neo-1 supports molecular generation, protein design, structure prediction, and molecular inpainting. Neo-1 can be guided using partial structures, sequences, or desired molecular properties. Drug Discovery & Optimization: Beyond hit discovery, Neo-1 leverages experimental data, accelerating iterative optimization in drug discovery. Applications already include molecular glues, PROTACs, and antibodies, significantly streamlining the path from computational design to experimental validation. Learn more: https://www.vant.ai/neo-1 Interested in pioneering the next frontier in biology and AI research? We're hiring—contact recruiting@vant.ai. https://lnkd.in/ekF8nfPK

Excited to share our talk with José Jiménez Luna and Yu Xie on using deep learning to emulate protein equilibrium ensembles - now live on YouTube! Watch here: https://lnkd.in/eFJ4Bkza

We had the pleasure to have Jeremy Wohlwend and Gabriele Corso give a sneak peak of these on our https://lnkd.in/e4kHMimm lecture a week ago. Recording is now live, link in comments!

🚀 Don't miss this week's talk with Jeremy Wohlwend and Gabriele Corso, where they will talk about their new model for 3D structure prediction of molecular complexes: Boltz-1. 🗓️ Friday - 4pm CET / 10am ET

🔥 Join us for a talk with Jeremy Wohlwend & Gabriele Corso on their recent paper Boltz-1 - a new open source SOTA for 3D structure prediction of biomolecular complexes. As always with Michael Bronstein and Bruno Correia 🗓️ When: Fri, Dec 06 - 4p CET/10a ET 🗒️ Sign-up: https://lnkd.in/e42FCCT9