Clemens Isert

New review: Many geometric #deeplearning methods use the 3D structure of macromolecules to predict properties that are relevant for #drugdesign. Preprint: arxiv.org/abs/2210.11250 Work with @atzkenneth & Gisbert Schneider. #compchem #machinelearning #drugdiscovery (1/4)

📢 The VantAI team is thrilled to be at NeurIPS 2025! 💥 If you’re interested in meeting with a member of our team for a coffee chat, drop your details at this link and we’ll follow up to schedule forms.gle/5SD1GK4bS24cKK…

@json_yim For a normal prediction, diffusion only happens in latent space and we decode at the end. But in order to visualize what’s happening, we’ve decoded at every step here, showing the analogous coordinate space at each point along the trajectory

@clemensisert Is it latent diffusion then coordinate diffusion? Gif makes it seem like coordinate diffusion is happening.

Really cool to see Neo-1 released. Latent diffusion to combine structure prediction and molecular generation across modalities. #ai #drugdiscovery #genai

With the scarcity of high-quality, leakage-free benchmarking datasets in the chem-bio space, it's really exciting to see this released today!

RT @vant_ai: We are thrilled to announce a new milestone collaboration with Bristol Myers Squibb (@bmsnews), utilizing our geometric deep l…

@GabriCorso Roche’s PDE10A dataset might be useful link.springer.com/article/10.100…

Dear academic, biotech & drug discovery twitter colleagues, I need your help! I'm collecting a list of benchmarks & evaluation datasets for protein-small molecule affinity and virtual screening capacity (e.g. published hit discovery campaign results), which ones do you recommend?

Building on success in ongoing projects, we are thrilled to announce the increased scope and range of our partnership with $BPMC bridging the gap between scientific advances in #inducedproximity/targeted protein degradation #TPD and #generativeAI businesswire.com/news/home/2024…

Fantastic work led by @atzkenneth which I really enjoyed being a part of! Especially seeing low micromolar activities confirmed experimentally for molecules generated by deep learning models was a highlight! #compchem #pharma

@chupvl QMugs - 665k drug-like molecules, 3 conformers each. Disclaimer: I‘m one of the authors :) nature.com/articles/s4159…

Hey mighty #compchem and #cheminf folks! Is there a database of precalculated conformers for drug-like compounds? Want to test some ML algo tuned for 3D. Thanks!

@EligeCarter 😂😂😂

Check our preprint for more details! Preprint: doi.org/10.26434/chemr… 📜 Code: github.com/cisert/bcpaff 🐍

These bond-critical points describe non-covalent interactions. We feed them into a 3D-aware graph neural network and use this representation to predict binding affinities.

Excited to share the next manuscript from my PhD - “Exploring protein-ligand binding affinity prediction with electron density-based geometric deep learning”. With @atzkenneth, @rinikerlab & Gisbert Schneider @ETH @ETH_en #compchem #machinelearning #quantummechanics

Happy to share our latest preprint. We show how graph transformer neural networks can be applied to in silico reaction screening.  Preprint: lnkd.in/eVzBRTwm @SEC_ETH @Roche @LMU_Muenchen @DavidBKonrad @ETH_en #compchem #deeplearning #drugdiscovery

Great to see our review “Structure-based drug design with geometric deep learning” published in @COSB_CRSB. Paper: sciencedirect.com/science/articl… (w/ @atzkenneth & Gisbert Schneider) #compchem #deeplearning

@josejimlun Is this a code review? o_O

@clemensisert does that mean that you're officially a globglobglobber?

The world is divided between `glob` and `glob.glob` people. This is a fact.

@langer_han Individual features can transform (change) either equivariantly (predictably) with the input, or invariantly (not changing at all). So while you might have both types of features, you’d normally treat them separately.

@clemensisert Does it make sense that features encoded in NNs can encode both equivariance and invariance ? Would this make sense ?

Hi #MachineLearning can someone recommend me intuitive material on equivariance ?

@chaitjo @langer_han Perfect timing ;)

@clemensisert @langer_han Thanks for sharing! :D