skills.rest
K-Dense-AI
K-Dense-AIOFFICIAL
Explore all skills by this author →
Website𝕏@k_dense_aiGitHub

esm

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

33.0k|3.2k|Updated Oct 19, 2025
Search Tags:#embeddings#bioinformatics#protein-design#structure-prediction#esm3#forge-api#esm-c
One-click install
npx skills add https://github.com/K-Dense-AI/scientific-agent-skills --skill esm-k-dense-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: esm
Source: https://github.com/K-Dense-AI/scientific-agent-skills/tree/main/scientific-skills/esm
Command: npx skills add https://github.com/K-Dense-AI/scientific-agent-skills --skill esm-k-dense-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Protein researchers need a unified toolkit to design, analyze, and leverage protein language models (ESM3 and ESM C) both locally and via Forge for scalable inference.

Core Features & Use Cases

  • Sequence design, generative design, and inverse folding across protein sequences
  • Structure prediction, embedding generation, and function-conditioned generation
  • Batch processing and cloud-based design via Forge for large-scale workflows

Quick Start

Load a model (e.g., esm3-sm-open-v1) and begin a simple design task for a target protein.

Frequently Asked Questions about esm

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

FAQPage Schema
How do I use protein language models for sequence design and structure prediction?▼

Protein language models like ESM3 and ESM C enable sequence design, structure prediction, and function-conditioned generation. You can load models locally or use the Forge API to apply generative design and inverse folding across protein sequences.

What is the difference between local ESM3 models and the Forge API for protein embeddings?▼

Local ESM3 models like esm3-sm-open-v1 run directly for individual protein analysis, while the Forge API provides cloud-based batch processing and scalable inference for large-scale protein design workflows.

Can I generate protein embeddings and perform inverse folding with ESM C?▼

Yes, ESM C supports embedding generation and inverse folding tasks. You can load esmc variants to generate sequence embeddings and apply inverse folding to design protein sequences targeting specific structural folds.

How do I condition protein generation on specific functions using ESM3?▼

ESM3 supports function-conditioned generation through its GenerationConfig and ESMProtein components. You can specify desired functional constraints to guide the model in generating protein sequences with targeted biological functions.

Which ESM3 model variants are available for local protein design tasks?▼

Available local ESM3 model variants include esm3-sm-open-v1, esm3-medium-2024-08, and esm3-large-2024-03. These models support varying scales of protein sequence design, structure prediction, and embedding-based analyses.

Does ESM3 support batch processing for large-scale protein design workflows?▼

Yes, ESM3 supports batch processing and cloud-based design via the Forge client. This allows researchers to execute large-scale protein design and embedding workflows efficiently across biological research projects.

Related Skills

Explore More Education & Research Skills→
Zeyuyang-0420Zeyuyang-0420

esm

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

Community
Advanced
Lord1EgyptLord1Egypt
2

protein-structure-prediction

Predict 3D protein conformations from amino acid sequences using AlphaFold2 and ESMFold.

Community
Advanced
PKU-YuanGroupPKU-YuanGroup
288

solublempnn

Inverse-fold proteins with the SolubleMPNN model to optimize solubility.

Official
Intermediate
adaptyvbioadaptyvbio
151

chai

Predict protein structures from FASTA sequences using Chai-1.

Official
Advanced