What is inverse folding in the context of ESM?

Inverse folding is the capability to design a protein sequence that will fold into a specific target 3D structure, useful for structural biology and engineering.

Can I use this skill for local protein modeling?

Yes, the skill supports local usage of open-weight models like esm3-sm-open-v1 and esmc-300m, allowing for experimentation without external API calls.

What is the difference between ESM3 and ESM C?

ESM3 is a multimodal generative model designed for complex protein design across sequence, structure, and function, while ESM C is optimized for efficiency and generating high-quality protein embeddings.

Does this skill support the EvolutionaryScale Forge API?

Yes, it includes full support for the Forge API, enabling access to high-parameter models like esm3-large and scalable batch processing.

ESM Protein Design & Analysis

Name: ESM Protein Design & Analysis
Author: ricable

byricable

•

データサイエンスとML

Empowers protein engineering and biological research through state-of-the-art ESM3 generative models and ESM C embeddings.

概要

The ESM skill integrates Evolutionary Scale Modeling into Claude's workflow, enabling advanced protein design across sequence, structure, and function tracks. It provides a comprehensive toolkit for tasks ranging from novel protein generation and structure prediction to high-efficiency representation learning. By supporting both local inference with open weights and the scalable Forge API, this skill streamlines complex bioinformatics tasks like inverse folding, variant screening, and functional annotation, making it indispensable for researchers and developers working in drug discovery and synthetic biology.

主な機能

Function-conditioned generation for creating proteins with specific biological annotations
Multimodal protein generation using ESM3 across sequence, structure, and function tracks
Automated structure prediction from sequence and inverse folding for structure-based design
Support for both local model deployment and scalable cloud-based Forge API inference
High-performance protein embeddings with ESM C for downstream machine learning tasks
1 GitHub stars

ユースケース

Designing novel proteins with specific functional properties or structural characteristics
Predicting 3D protein structures and performing sequence-from-structure inverse folding
Generating high-quality embeddings for sequence similarity, classification, and function prediction

概要

主な機能

Function-conditioned generation for creating proteins with specific biological annotations
Multimodal protein generation using ESM3 across sequence, structure, and function tracks
Automated structure prediction from sequence and inverse folding for structure-based design
Support for both local model deployment and scalable cloud-based Forge API inference
High-performance protein embeddings with ESM C for downstream machine learning tasks
1 GitHub stars

ユースケース

Designing novel proteins with specific functional properties or structural characteristics
Predicting 3D protein structures and performing sequence-from-structure inverse folding
Generating high-quality embeddings for sequence similarity, classification, and function prediction