skills.rest
robotlearning123
robotlearning123COMMUNITY
Explore all skills by this author →
𝕏@wan73184085GitHub

esm

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

Updated Dec 17, 2025
Search Tags:#sequence analysis#structure prediction#protein design#protein language models#function annotation
One-click install
npx skills add https://github.com/robotlearning123/claude-scientific-skills --skill esm-robotlearning123
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: esm
Source: https://github.com/robotlearning123/claude-scientific-skills/tree/main/scientific-skills/esm
Command: npx skills add https://github.com/robotlearning123/claude-scientific-skills --skill esm-robotlearning123

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires esm, torch, torchvision, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill provides advanced protein language models to generate sequences, structures, and functions, enabling users to design proteins, predict their properties, and analyze their functions.

Core Features & Use Cases

  • Protein Sequence Generation: Generate novel protein sequences with desired properties.
  • Structure Prediction and Inverse Folding: Predict protein structures from sequences or design sequences for given structures.
  • Protein Embeddings: Generate embeddings for protein sequences for downstream tasks like classification or similarity analysis.
  • Function Conditioning and Annotation: Generate proteins with specific functional annotations or predict function from sequence.
  • Chain-of-Thought Generation: Iteratively refine protein designs using a chain-of-thought approach.
  • Batch Processing: Process multiple proteins efficiently using the Forge API.
  • Model Selection Guide: Choose the appropriate model based on the task requirements and performance needs.
  • Installation and Usage: Detailed instructions for installing and using the models locally or via the Forge API.
  • Common Workflows: Examples of common workflows like novel GFP design, protein variant generation, and function prediction pipelines.
  • References: Comprehensive documentation and API references for the models and their usage.
  • Best Practices: Guidelines for responsible use and optimization of the models.
  • Resources and Documentation: Links to the GitHub repository, Forge platform, scientific papers, and community resources.
  • Responsible Use: Follow the Responsible Biodesign Framework when designing novel proteins.

Quick Start

Use the esm skill to generate a novel protein sequence with specific properties.

Frequently Asked Questions about esm

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

FAQPage Schema
How do I predict protein structures from amino acid sequences?▼

Protein sequence analysis with this skill generates structural predictions by processing amino acid chains through ESM3 and ESM C models. It maps sequence data directly to structural representations using protein language models.

Can I generate novel protein sequences with specific functional annotations?▼

Yes, you can generate novel protein sequences with specific functional annotations. The skill uses function conditioning to design proteins that meet desired functional criteria and applies annotations during the sequence generation process.

How do I extract protein embeddings for downstream classification tasks?▼

Extract protein embeddings for classification by passing protein sequences through the ESM models. The skill generates numerical representations of sequences that can be used for similarity analysis and downstream machine learning tasks.

Does protein design with ESM require local GPU resources or can it run via API?▼

Protein design with ESM supports both local model usage requiring torch and torchvision, and cloud-based Forge API inference for scalable batch processing. Choose local execution for individual designs and the Forge API for processing multiple proteins efficiently.

What is the best way to iteratively refine novel protein designs?▼

The best way to iteratively refine novel protein designs is using the chain-of-thought generation approach. This method allows you to progressively improve protein sequences and structures by applying constraints across multiple generation cycles.

Are there guidelines for responsible use when generating novel proteins?▼

Yes, responsible use guidelines apply when generating novel proteins. The skill includes a Responsible Biodesign Framework that provides best practices and precautions to follow when designing novel protein sequences and structures.

Related Skills

Explore More Software Engineering Skills→
JosephWoodallJosephWoodall
1

esm

Generate protein sequences and predict structures with ESM3 and ESM C models.

Community
Advanced
junior1pjunior1p
11

uniprot

Retrieve protein sequences and annotations from UniProt via its REST API.

Community
Intermediate
ovachieverovachiever
52

esm

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

Community
Advanced
SologaSologa
1

esm

Generate protein sequences and embeddings with ESM3 and ESM C.

Community
Advanced