Data as of Sep 29, 2026 · Based on 3,017 AI responses · See how Parse measures this
BindCraft is an open-source pipeline that designs protein binders by integrating AlphaFold2 backpropagation, MPNN, and PyRosetta to generate binder designs against a specified target and stop when enough designs are ready to order. It provides an installation script and requires a CUDA-enabled Nvidia GPU and a PyRosetta license for commercial use; while a Google Colab notebook exists for testing, the project recommends running locally with substantial GPU memory (at least 32 GB). The workflow is configured by a target.json in settings_target, with outputs saved under a design_path and binder_name prefixes, and the repository includes example files and instructions for cloning and running the pipeline.
Hosted on GitHub
0%No change
of AI answers about BindCraft and its rivals. Week of Sep 21
“Bindcraft: Uses an iterative structure-prediction and hallucination approach to optimize de novo protein binders”
“BindCraft: An increasingly popular automated pipeline built on top of RFdiffusion and AlphaFold”
AI mentioned BindCraft in 0% of answers about BindCraft and its rivals in the week of Sep 21.
Excerpts where BindCraft appeared in the AI's answer
BindCraft: An increasingly popular automated pipeline built on top of RFdiffusion and AlphaFold
Excerpts where BindCraft appeared in the AI's answer
Bindcraft: Uses an iterative structure-prediction and hallucination approach to optimize de novo protein binders
