AIMNet2(Score) ranks protein–ligand binding from one structure and no affinity labels
olexandr · x · 2026-07-24
AIMNet2(Score) scores protein–ligand binding without affinity labels
A new ChemRxiv preprint introduces AIMNet2(Score), a label-free method for ranking protein–ligand complexes from a single bound structure.
- The score is composed of three physics-based terms: protein–ligand interaction energy, ligand desolvation, and conformational strain.
- It relies on QM-level physics from AIMNet2 machine-learned interatomic potentials rather than experimental affinity data.
- The authors frame it as a direct alternative to label-dependent scoring methods for protein–ligand binding prediction.
Related event: AIMNet2(Score) Enables Label-Free Protein-Ligand Binding Scoring(3 posts)→
More from Research
- ICML 2026 oral paper replication scores stay middling after a stricter re-scoring — profjamesevans · 2026-07-27
- Long-running agents will need immutable event logs, this thread argues — sebpaquet · 2026-07-27
- Seed IQ navigates Doom II, prompting questions about benchmarks beyond ARC-AGI — Fit_Transition8824 · 2026-07-27
- Agentic Data Science in Practice: Agents Write Code but Answer Wrong Questions — hugobowne · 2026-07-27
- A concise canon of foundational papers in ML, systems, NLP, speech, and audio — deliprao · 2026-07-27
- TechCrunch says brain-wave signals could be the next unlock for physical AI training — TechCrunch AI · 2026-07-27