esm

Design, analyze, and derive protein embeddings using ESM3 and ESM C models.

18|1|Updated Dec 27, 2025
One-click install
npx skills add https://github.com/LogauaEngstrom/claude-scientific-skills --skill esm-logauaengstrom
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: esm
Source: https://github.com/LogauaEngstrom/claude-scientific-skills/tree/main/scientific-skills/esm
Command: npx skills add https://github.com/LogauaEngstrom/claude-scientific-skills --skill esm-logauaengstrom

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Comprehensive toolkit for protein language models including ESM3 (generative multimodal protein design across sequence, structure, and function) and ESM C (efficient protein embeddings and representations). Use this skill when working with protein sequences, structures, or function prediction; designing novel proteins; generating protein embeddings; performing inverse folding; or conducting protein engineering tasks. Supports both local model usage and cloud-based Forge API for scalable inference.

Core Features & Use Cases

  • Multimodal generation across sequence, structure, and function with ESM3.
  • Efficient embeddings and representations with ESM C.
  • End-to-end workflows: sequence completion, inverse folding, structure conditioning, function conditioning.
  • Batch processing and Forge API access for scalable inference.
  • Use cases include protein design, variant generation, structure-based sequence optimization, and embedding-driven clustering and retrieval.

Quick Start

Load an ESM3 model and start a simple sequence design task for a short protein.

Frequently Asked Questions about esm

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

FAQPage Schema
How do I generate protein embeddings for sequence clustering and retrieval?

To generate protein embeddings for clustering and retrieval, you can use the ESM C model via this skill. It provides efficient representations and supports batch processing to compute embeddings across large protein datasets.

Can I design novel proteins by conditioning on structure and function?

Yes, you can design novel proteins by conditioning on structure and function using the ESM3 model. This skill enables multimodal generation across protein sequence, structure, and function for targeted protein engineering.

Do I need a local GPU to run ESM3, or can I use a cloud API?

You do not need a local GPU; you can access ESM3 and ESM C through the cloud-based Forge API for scalable inference. The skill also supports local model deployment if you have your own hardware.

What's the best way to perform inverse folding for structure-based sequence optimization?

The best way to perform inverse folding for structure-based sequence optimization is using the ESM3 model. This skill handles end-to-end workflows, allowing you to condition structure prediction directly to generate optimized sequences.

Does this skill support batch processing for generating multiple protein variants?

Yes, this skill supports batch processing for generating multiple protein variants. It includes optional scripts that enable batch workflows, allowing you to scale variant generation and multimodal generation tasks efficiently.