BoltzGen
Toward universal binder design across modalities
What's new
A generative design counterpart to Boltz covering proteins, peptides, antibodies, nanobodies and small-molecule binders.
BoltzGen aims to be a universal binder-design model. From the authors of Boltz, it supports several design protocols out of the box.
Protocols
- protein-anything — design proteins to bind arbitrary targets.
- protein-small_molecule — design proteins to bind small molecules.
- antibody / nanobody — design CDRs for antibody and nanobody formats.
- redesign — optimise or redesign existing proteins.
MIT licensed, with high reported hit rates at the cost of more inference compute than fast backbone models.
Related tools
Open source
AI Models
Boltz-2
Open-source structure and binding-affinity prediction, MIT licensed
Python libraryCommand lineModel weights
Open source
AI Models
BindCraft
One-shot de novo binder design with AlphaFold2 backpropagation
Python libraryCommand line
Open source
AI Models
RFdiffusion
Generative diffusion for de novo protein backbone design
Python libraryCommand lineModel weights