Can I use this skill without a local GPU?

Yes. While local model execution benefits significantly from a GPU, this skill supports the Forge API which allows you to perform high-performance inference in the cloud using an API token.

Does this skill support structure prediction from sequence?

Yes, it utilizes ESM3's structure track to predict 3D coordinates from protein sequences. It also supports inverse folding, which is the process of designing a sequence to fit a known structure.

What is the difference between ESM3 and ESM C in this skill?

ESM3 is a generative multimodal model designed for creating protein sequences, structures, and functions, while ESM C is specifically optimized for creating efficient, high-quality embeddings and representations for analytical tasks.

How do I authenticate for cloud-based protein generation?

You can obtain a Forge API token from the EvolutionaryScale platform and use it with the ESM3ForgeInferenceClient provided within the skill's SDK.

ESM Protein Modeling

Name: ESM Protein Modeling
Author: BbgnsurfTech

byBbgnsurfTech

•

데이터 과학 및 ML

Enables advanced protein sequence generation, structure prediction, and representation learning using Evolutionary Scale Modeling.

소개

This skill integrates the ESM (Evolutionary Scale Modeling) toolkit into Claude, providing a powerful interface for state-of-the-art protein engineering and biological analysis. It supports the ESM3 family for generative design across sequence, structure, and function, alongside ESM C for high-efficiency protein embeddings. Whether you are performing inverse folding, predicting 3D structures, or generating novel functional proteins through chain-of-thought refinement, this skill streamlines the transition between local model execution and scalable cloud-based inference via the Forge API.

주요 기능

3 GitHub stars
Multimodal protein design across sequence, structure, and function tracks
Support for both local inference and scalable cloud-based Forge API
Chain-of-thought protein generation for iterative design refinement
Inverse folding for designing sequences from target 3D structures
High-quality protein representation learning with ESM C embeddings

사용 사례

Extracting high-dimensional embeddings for downstream machine learning tasks in biology
Performing structure prediction and variant screening for protein engineering
Designing novel proteins with specific functional annotations or structural properties

소개

주요 기능

3 GitHub stars
Multimodal protein design across sequence, structure, and function tracks
Support for both local inference and scalable cloud-based Forge API
Chain-of-thought protein generation for iterative design refinement
Inverse folding for designing sequences from target 3D structures
High-quality protein representation learning with ESM C embeddings

사용 사례

Extracting high-dimensional embeddings for downstream machine learning tasks in biology
Performing structure prediction and variant screening for protein engineering
Designing novel proteins with specific functional annotations or structural properties