esm

Generate protein sequences, predict structures, and create embeddings with ESM models.

8|Updated Nov 19, 2025
One-click install
npx skills add https://github.com/sanand0/scientific-research --skill esm-sanand0
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: esm
Source: https://github.com/sanand0/scientific-research/tree/main/.claude/skills/esm
Command: npx skills add https://github.com/sanand0/scientific-research --skill esm-sanand0

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill empowers researchers and developers to design novel proteins, predict their structures, and generate embeddings for downstream analysis, streamlining complex bioinformatics workflows.

Core Features & Use Cases

  • Generative Protein Design: Create new protein sequences and structures with desired functions using ESM3.
  • Protein Embeddings: Generate high-quality representations for classification, similarity search, and clustering using ESM C.
  • Structure Prediction & Inverse Folding: Predict 3D protein structures from sequences or design sequences for given structures.
  • Use Case: A synthetic biologist needs to design a new enzyme with enhanced catalytic activity. They can use this Skill to generate candidate sequences, predict their structures, and obtain embeddings for functional analysis.

Quick Start

Use the esm skill to generate a protein sequence for the provided structure file 'target.pdb'.

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 machine learning classification?

Generate high-quality protein embeddings for classification and similarity search using the ESM C model. You can run the model locally or use the Forge API to produce representations suitable for downstream machine learning tasks.

Can I design new protein sequences for a specific 3D structure file?

Yes, you can perform inverse folding to design protein sequences for a given 3D structure. The Skill uses state-of-the-art generative models to create candidate sequences that fold into the target structure.

What is the best way to predict protein 3D structure from an amino acid sequence?

Predict 3D protein structures directly from amino acid sequences using the ESM generative models. The Skill facilitates structure prediction through local execution or scalable cloud-based inference via the Forge API.

How do I design a novel enzyme with enhanced catalytic activity?

Design novel enzymes by using the ESM3 generative model to create candidate protein sequences with desired functions. You can then predict their structures and generate embeddings for functional analysis.

Can I use local execution for protein sequence generation or do I need cloud inference?

You can use either local model execution or cloud-based inference via the Forge API for protein sequence generation. Local execution supports direct analysis while the Forge API provides scalable solutions for larger tasks.

Are there limitations when using generative models for protein engineering?

Generative protein design requires validating candidate sequences for actual catalytic activity and structural stability. The Skill generates predictions and embeddings, but experimental verification of the designed proteins remains necessary.