Teaching generative models to hallucinate
There are currently two main approaches to computational protein binder design: optimization (exemplified by BindCraft) and generative models (e.g. BoltzGen). A design campaign using either method looks very similar: you first generate a large collection of designs and then rank and filter them down to a handful for wetlab
There are currently two main approaches to computational protein binder design: optimization (exemplified by BindCraft) and generative models (e.g. BoltzGen). A design campaign using either method looks very similar: you first generate a large collection of designs and then rank and filter them down to a handful for wetlab testing. In practice the main difference is that optimization (also known as hallucination[[1]]) has high average per-design in silico quality, but is very slow; while generative models are much, much faster but average design quality is lower[[2]]. We'll show that, by…
saved by
related reading
- One-shot design of functional protein binders with BindCraftnature.com
- Ludovico Comitoludocomito.dev
- ~180 lines of code to win the in silico portion of the Adaptyv Nipah binding competitionblog.escalante.bio
- Fast and Ultra-Capable Protein Design: Advancing the Frontier Through Atomistic SE(3)-Equivariance with Genie 3biorxiv.org
- prompts/prompts/multi_target_binder_design_prompt.md · Anthropic/claude-protein-binder-design at mainhuggingface.co
- Minibinder design is just not that hardblog.escalante.bio
- Protein World Model | Biohubbiohub.ai
- Anthropic has not solved the peptide-binder design problem, but maybe bi[o]hub hasblog.genesmindsmachines.com
- Anthrogenanthrogen.com
- AlphaProteo generates novel proteins for biology and health researchdeepmind.google
- Using Self-Correcting Search to Accelerate Materials Discoverygoodfire.ai
- Engineering Plastic-Degrading Enzymes and PCSK9 Binders with Protein AI Toolsblog.latch.bio