boltzgen

Design all-atom protein structures with side-chain packing around target binding sites.

25|5|Updated Mar 22, 2026
One-click install
npx skills add https://github.com/zongtingwei/Bioclaw_Skills_Hub --skill boltzgen-zongtingwei
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: boltzgen
Source: https://github.com/zongtingwei/Bioclaw_Skills_Hub/tree/main/skills/protein-design/skills/boltzgen
Command: npx skills add https://github.com/zongtingwei/Bioclaw_Skills_Hub --skill boltzgen-zongtingwei

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

BoltzGen addresses the challenge of designing proteins with precise all-atom detail, enabling researchers to generate backbone plus side-chain optimized designs around defined targets and ligands.

Core Features & Use Cases

  • All-atom design: simultaneous backbone, side-chain packing, and interface optimization.
  • YAML-based configuration: end-to-end design with configurable targets, protocols, and constraints.
  • Use Case: design a binder for a target protein or small-molecule, then export CIF-style designs ready for validation.

Quick Start

Provide a target structure and binding goals in YAML and run BoltzGen to generate all-atom designs.

Frequently Asked Questions about boltzgen

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design all-atom protein structures with accurate side-chain packing?

BoltzGen handles all-atom protein design by simultaneously optimizing backbones and side-chains around specified targets using a YAML configuration to generate CIF-like structures.

Can I design protein binders around small molecules using a YAML configuration?

Yes, you can design protein binders around small molecules by defining targets and binding goals in a YAML configuration, enabling simultaneous backbone and side-chain optimization.

What is the best way to generate backbone and side-chain optimized protein designs around ligands?

The best way to generate optimized designs around ligands is to use an end-to-end all-atom diffusion workflow that supports simultaneous backbone generation and side-chain packing around defined targets.

Does all-atom protein design support exporting CIF-style structures and metrics?

Yes, all-atom protein design supports exporting CIF-style designs and metrics, providing structures ready for downstream validation after completing the end-to-end design pipeline.

What do I need to provide to run an end-to-end all-atom protein design workflow?

You need to provide a target structure, such as a ligand or protein, and specify your binding goals within a YAML configuration to initialize the design and structure handling process.