skills.rest
silverstein
silversteinCOMMUNITY
Explore all skills by this author →
Website𝕏@matsilversteinGitHub

esm

Generate protein sequences, predict structures, and compute embeddings with ESM3 and ESM C.

21|2|Updated Dec 8, 2025
Search Tags:#structure prediction#sequence modeling#esm3#protein embeddings#protein generation#function annotation#forge api
One-click install
npx skills add https://github.com/silverstein/claude-scientific-skills-desktop --skill esm-silverstein
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: esm
Source: https://github.com/silverstein/claude-scientific-skills-desktop/tree/main/corpus/esm
Command: npx skills add https://github.com/silverstein/claude-scientific-skills-desktop --skill esm-silverstein

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

ESM helps you design, predict, and analyze proteins without manually stitching together fragile, one-off modeling steps, so you can move from a sequence or structure to actionable outputs faster.

Core Features & Use Cases

  • Protein generation across modalities: Use ESM3 for sequence generation, structure prediction, inverse folding (sequence from structure), and function-conditioned design.
  • Protein embeddings for downstream ML: Use ESM C to compute efficient representations for similarity search, clustering, classification features, and transfer learning.
  • Local or Forge-powered inference: Run open-weight ESM3/ESM C locally or scale ESM3 via the Forge API when larger models are needed.

Quick Start

Ask for protein sequence completion or structure prediction for a specific amino-acid sequence you provide, and request embeddings for similarity or clustering if you need downstream analysis.

Frequently Asked Questions about esm

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

FAQPage Schema
How do I predict a protein structure from an amino acid sequence?▼

You can predict protein structures by processing an amino acid sequence through ESM3. The Skill handles structure prediction as one of its core track-based generation tasks, returning structural outputs directly from the provided sequence.

How do I compute protein embeddings for clustering and classification?▼

You can compute protein embeddings for clustering and classification using ESM C. The Skill derives efficient representations suitable for downstream ML workflows, including similarity search and transfer learning.

Can I generate a protein sequence based on specific functional annotations?▼

Yes, ESM3 supports function-conditioned protein generation. You can provide incomplete biological inputs and use GenerationConfig tracks to guide the generation of sequences matching your desired functional annotations.

How does inverse folding work for finding a sequence from a known protein structure?▼

Inverse folding uses ESM3 to predict an amino acid sequence from a given protein structure. The Skill applies this protein language modeling task to help you design sequences that fold into specific structural conformations.

Do I need the Forge API to run ESM3 for protein generation?▼

No, you do not need the Forge API to run ESM3. The Skill supports local execution with open-weight models, but you can scale up larger ESM3 models via the Forge API when local resources are insufficient.

What is the best way to complete a masked protein sequence?▼

The best way to complete a masked protein sequence is using ESM3's masked sequence completion. The Skill processes incomplete biological inputs and fills in the missing amino acids using protein language modeling.

Related Skills

Explore More Education & Research Skills→
K-Dense-AIK-Dense-AI
33.0k

esm

Design, predict, and embed protein sequences using ESM3 and ESM C models.

Official
Advanced
zechenzhangAGIzechenzhangAGI
11.5k

faiss

Perform billion-scale k-NN similarity search with FAISS index types and GPU acceleration.

Community
Intermediate
Zeyuyang-0420Zeyuyang-0420

esm

Model protein sequences, structures, and functions using ESM3 and ESM C.

Community
Advanced
ovachieverovachiever
52

esm

Generate protein sequences, structures, and embeddings using ESM3 and ESM C models.

Community
Advanced