skills.rest
LiorZ
LiorZCOMMUNITY
Explore all skills by this author →
𝕏@LiorZmanGitHub

fair-esm

Run FAIR's ESM protein models for embeddings, ESMFold structure prediction, inverse folding, and variant analysis.

2|Updated May 12, 2026
Search Tags:#bioinformatics#machine-learning#esm#inverse-folding#structure-prediction#variant-prediction#protein-models
One-click install
npx skills add https://github.com/LiorZ/protein-design-skills --skill fair-esm
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: fair-esm
Source: https://github.com/LiorZ/protein-design-skills/tree/main/skills/fair-esm
Command: npx skills add https://github.com/LiorZ/protein-design-skills --skill fair-esm

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Provides authoritative references and tooling for running FAIR's ESM family models (ESM-2, ESMFold, ESM-IF1, ESM-1v, MSA Transformer) to accelerate protein-design research and analysis.

Core Features & Use Cases

  • Reference implementations and docs for running ESM-2, ESMFold, ESM-IF1, ESM-1v, and MSA Transformer in research pipelines.
  • Supports obtaining embeddings, end-to-end structure predictions, inverse folding, and variant scoring across single- and multi-chain setups.
  • Useful in educational contexts, prototyping research pipelines, and reproducible experiments across protein design and analysis tasks.

Quick Start

Install the fair-esm package and run a basic workflow to verify model loading and a sample inference.

Frequently Asked Questions about fair-esm

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

FAQPage Schema
How do I run ESMFold for protein structure prediction from a single sequence?▼

ESMFold provides end-to-end structure prediction from single protein sequences. You need Python, PyTorch, and OpenFold installed, plus model weights cached locally to execute the inference workflow.

What is inverse folding and how does ESM-IF1 generate protein sequences from structures?▼

Inverse folding predicts amino acid sequences from given protein structures. ESM-IF1 executes this task across single- and multi-chain contexts to facilitate protein design research and analysis.

Do I need PyTorch and OpenFold to use ESM-2 for protein embeddings?▼

You need Python and PyTorch to run ESM-2 for embeddings. OpenFold is an optional extra required specifically for running ESMFold structure prediction, not for standard embeddings.

Can I use ESM-1v for variant-effect prediction on multi-chain protein complexes?▼

ESM-1v supports variant-effect analysis and scoring. The tooling accommodates both single-sequence and multi-chain contexts, allowing variant prediction workflows across complex protein setups.

What are the limitations of using MSA Transformer for sequence analysis workflows?▼

MSA Transformer requires multiple sequence alignment inputs rather than single sequences. It is limited to research, prototyping, and educational contexts, not optimized for scaled production environments.

Related Skills

Explore More Education & Research Skills→
swaruplabswaruplab
94

esm

Design and analyze proteins with ESM3/ESM C models via local or Forge inference.

Community
Advanced
junior1pjunior1p
11

ipsae

Rank protein binder designs using ipSAE_min, LIS, and pDockQ scores.

Community
Intermediate
jimmc414jimmc414
557

esm

Generate protein sequences and extract embeddings using ESM3 and ESM C.

Community
Advanced
hxk622hxk622
8

esm

Generate protein sequences and structures with ESM3 and embeddings with ESM C.

Community
Advanced