design-round-modal

Name: design-round-modal
Availability: InStock
Author: Alex

Automate protein binder design rounds with BoltzGen or RFdiffusion3 on Modal GPUs.

Updated Feb 27, 2026

Search Tags:#evaluation #modal #protein design #binder design #boltzgen #rfdiffusion3

Ready to run

npx skills add https://github.com/alex-hh/in-silico-pgdh --skill design-round-modal

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: design-round-modal
Source: https://github.com/alex-hh/in-silico-pgdh/tree/main/.claude/skills/design-round-modal
Command: npx skills add https://github.com/alex-hh/in-silico-pgdh --skill design-round-modal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill automates the entire lifecycle of protein binder design and evaluation using Modal's GPU infrastructure, providing an alternative when Lyceum is unavailable or unreliable.

Core Features & Use Cases

Full Lifecycle Automation: Orchestrates design, synchronization, evaluation, and publishing for protein binder design rounds.
Flexible Design Tools: Supports both BoltzGen and RFdiffusion3 for various design strategies (active site, dimer interface, surface, inpainting).
Use Case: Run a complete round of binder design using RFdiffusion3 for the active site strategy, evaluate the generated designs, and publish the results, all through a single, streamlined process.

Quick Start

Execute the design-round-modal skill to run a complete binder design and evaluation round using Modal GPUs.

Frequently Asked Questions about design-round-modal

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

FAQPage Schema

How do I automate protein binder design and evaluation using Modal GPUs?▼

You can automate the complete protein binder design lifecycle using Modal GPUs by submitting designs, synchronizing data, running fast and slow evaluations, scoring, and publishing the results through a single streamlined process.

Can I use RFdiffusion3 and BoltzGen for different protein design strategies?▼

Yes, you can use RFdiffusion3 and BoltzGen to execute multiple design strategies including active site, dimer interface, surface, and inpainting approaches for your protein binder design rounds.

What do I need to run a complete protein binder design round end-to-end?▼

To run a complete protein binder design round end-to-end, you need Modal authentication for GPU access and specific Python scripts to execute the design submission, evaluation, and publishing workflows.

How do I evaluate and score generated protein binder designs?▼

You can evaluate and score generated protein binder designs by running fast and slow evaluation phases on Modal GPUs, which handle the scoring metrics and publish the final results automatically.

Is there an alternative to Lyceum for running protein binder design workflows?▼

Yes, you can use Modal GPU infrastructure to run your protein binder design workflows, providing a reliable alternative when Lyceum is unavailable or unreliable for your evaluation pipelines.