With the release of Boltz 2, is there already a review comparing the efficiency of different (new) docking tools? (I’ve already tried Pocket Vina, which is actually quite good for high-throughput)
A direct comparison with Boltz-2 hasn’t been done yet, but it would be interesting to see one between co-folding and the classical/hybrid docking benchmarks!
Illustration with overlaid text: The image shows a cancer cell in the bloodstream. The overlaid text appears in white, all-caps sans-serif font inside a dark blue rectangular box at the top left. It reads: “USING DEEP LEARNING FOR PRECISION CANCER THERAPY.” A small credit at the bottom right reads: “© Annie Cavanagh / Wellcome Collection.”
Nearly 50 new cancer drugs are approved each year – but which one fits which patient?
At the #mdcBerlin, @al2na.bsky.social’s team built Flexynesis, a deep learning toolkit to guide precision cancer care.
Learn more:
👉 www.mdc-berlin.de/news/press/u... 👈
is this how small molecules bind?? 😼
🚀 GPU-accelerated docking to P2Rank-predicted pockets
All results, code (MIT License), and data are open and available:
📄 Paper: arxiv.org/abs/2506.20043
📦 Data: zenodo.org/records/1573...
💻 Code: github.com/BIMSBbioinfo...
Huge thanks to co-authors @al2na.bsky.social, @borauyar.bsky.social, and Vedran Franke!
We benchmarked PocketVina across four widely used datasets (PDBbind, PoseBusters, Astex, DockGen), and introduce TargetDock-AI — a large-scale benchmark of >500K protein–ligand pairs with activity labels from PubChem.
(5/n)
• Achieves state-of-the-art success rates on physically valid pose prediction
• Works across ligand flexibility levels and diverse, unseen protein targets
(4/n)
PocketVina offers a robust alternative:
• Identifies multiple pocket centers using P2Rank
• Performs GPU-accelerated docking with QuickVina 2-GPU 2.1
• Completes docking + binding affinity prediction in under 1.5 seconds, with no model training
(3/n)
...physically realistic ligand poses — and are not always as efficient or accurate as often claimed. (2/n)
I'm excited to share our new preprint: PocketVina — a fast, scalable, and accurate multi-pocket molecular docking method.
Docking remains essential in early-stage drug discovery, but recent deep learning–based approaches still face limitations in generating...
Thread - (1/n)
Artistic rendering of a biochemical model: a small molecule ligand, shown as a ball-and-stick model colored by element, is bound in a pocket in a protein surface, shown as a space filling model colored off-white.
We're changing the field of #compchem by creating free and open-source software for performing alchemical free energy calculations. Our flagship protocol calculates relative binding free energies of protein-ligand systems. Try it out in your browser: colab.research.google.com/github/OpenF...
I remember when I first started learning ML—Andrew Ng offered a Coursera course that uses Octave and covers neural networks for image classification with MNIST. You might find it helpful! :)
Join us to connect with the vibrant #singlecell community.
📢Register for the #ISCO'25 Conference "Innovations in #SingleCell #OMICS" in Berlin!
🗓️ 12-13 May 2025
🎤 Fantastic Keynote and Invited Speakers
🫵🏿 Many slots for talks: submit your abstract
🔗http://isco-conference.eu
Please spread the word!
1/4 🧵 Preprint alert: In "Metrics Matter: Why We Need to Stop Using Silhouette in #SingleCell #Benchmarking," we reveal critical flaws in common #Evaluation metrics for #Integration and propose robust alternatives. @uweohler.bsky.social
www.biorxiv.org/content/10.1...
