Predict protein, nucleic-acid, and small-molecule complex structures with Boltz-2.
npx skills add https://github.com/guanxiaol/open-science --skill boltz-guanxiaol
Please help me install this Agent Skill. Skill: boltz Source: https://github.com/guanxiaol/open-science/tree/main/runtime/skills/boltz Command: npx skills add https://github.com/guanxiaol/open-science --skill boltz-guanxiaol
This Skill addresses the challenge of predicting complex structures for proteins, nucleic acids, and small molecules, providing a powerful tool for validating designed binders, co-folding proteins, and offering an open-source alternative to AlphaFold3.
Run the boltz skill to predict the structure of a protein-ligand complex using the provided YAML file 'complex.yaml'.
High-intent search queries and answers about installing and using this skill.
To predict protein-ligand complex structures, you can use the Boltz-2 model to co-fold proteins with small molecules or nucleic acids. It accurately predicts these interactions to help you understand binding mechanisms.
Yes, Boltz-2 serves as an open-source alternative to AlphaFold3 for co-folding proteins with SMILES or CCD ligands. It also offers optional binding-affinity prediction for designed binders.
Validating designed binders requires predicting their complex structures with the target protein. Boltz-2 performs this prediction, allowing you to analyze how your designed binder interacts with the target.
Yes, predicting protein, nucleic-acid, and small-molecule complex structures with Boltz-2 requires a GPU. This hardware is necessary to run the structure prediction computations efficiently.
You can co-fold a protein with a SMILES or CCD ligand by running the boltz skill with a provided YAML file. This process uses Boltz-2 to accurately predict the resulting complex structure.
PKU-YuanGroupPredict protein, nucleic-acid, and small-molecule complex structures with Boltz-2.
PharMolixPredict protein complex binding affinity scores from PDB structures using Prodigy Prot.
PKU-YuanGroupDock small-molecule ligands into protein pockets with DiffDock-L and rank poses by confidence.
shushuznPredict protein–ligand binding poses with per-pose confidence scores.
