BindCraft2 protein design suite launches with full open-source code, free for academic and industry use
mmbronstein · x · 2026-09-21
PacesaLab released BindCraft2 ahead of its paper, open-sourcing the full code on GitHub the same day, free for both academic and industry use, explicitly to let researchers apply it to the ongoing Adaptyv protein design competition.
- Unified workflow: combines de novo miniproteins, scaffolded binders, cyclic peptides and multistate design; users describe a target, pick a binder modality and specify properties, with named presets supplying design settings and acceptance filters.
- Pipeline: sequence optimization via AlphaFold 2 plus ProteinMPNN redesign, candidate evaluation with separate AlphaFold models and structural filters, returning sequences, predicted complexes and ranked results with interface, fold and molecular property metrics.
- Caveat: authors note these are computational designs — binding, selectivity and requested properties still require experimental validation.
More from Research
- Tencent open-sources T-Mem, a memory architecture that anticipates instead of archiving — blaizedsouza · 2026-09-21
- Radial Duality paper turns constrained optimization into unconstrained Lipschitz problems — prof_grimmer · 2026-09-21
- LoRA beats full fine-tuning for physics inference: same posterior, 1/5 the memory — bravo_abad · 2026-09-21
- Aletheia's Quest competition crowns winners: 478 entries from 19 teams vied for $50k to build LLM lie detectors — gsarti_ · 2026-09-21
- IntBMoE Decouples MoE Participation, Execution and Memory, Deployed in AMap RecSys — Ran Cheng · 2026-09-21
- Quantizing Cellpose-SAM for stem cell imaging: W4/W8 hits 6.76x compression with zero failures — capicu-ai · 2026-09-21