What models are supported by the ESM skill?

The skill supports the ESM3 family for generative multimodal design and the ESM C family for high-efficiency protein representation and embeddings.

Can I use this skill for structure prediction?

Yes, it includes capabilities for predicting 3D structures from sequences and performing inverse folding to design sequences for target structures.

Does it require a paid API to function?

No, it supports local usage of open-weight models like esm3-sm-open-v1, though the Forge API is available for higher-scale production tasks.

How do I generate protein embeddings for ML?

You can use the ESM C models (e.g., esmc-300m) included in the skill to encode protein sequences into high-quality tensors for downstream analysis.

ESM Protein Design & Modeling

Name: ESM Protein Design & Modeling
Author: Zehong-Wang

byZehong-Wang

0•

データサイエンスとML

Accelerates protein engineering and research using state-of-the-art ESM3 and ESM C language models for sequence generation and structure prediction.

ESM (Evolutionary Scale Modeling) is a comprehensive toolkit that integrates powerful protein language models into your development environment. This skill enables advanced tasks such as multimodal protein design across sequence, structure, and function tracks, high-fidelity structure prediction, and inverse folding. By supporting both local model execution and the scalable cloud-based Forge API, it provides a flexible framework for generating protein embeddings, performing functional annotations, and conducting complex protein engineering workflows. Whether you are a bioinformatician building ML models or a researcher designing novel enzymes, this skill streamlines the interaction with cutting-edge biological AI.

主な機能

01Seamless integration with Forge API for scalable cloud inference

02Chain-of-thought refinement for iterative protein design tasks

03Multimodal generative protein design (sequence, structure, and function)

04Efficient protein embedding generation using ESM C models

050 GitHub stars

06High-accuracy structure prediction and inverse folding capabilities

ユースケース

01Automating protein variant generation and screening for engineering projects

02Generating high-quality protein representations for downstream machine learning tasks

03Designing novel proteins with specific functional properties or structural constraints

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add zehong-wang/kosmos esm

For use in Claude.ai and ChatGPT

Download Skill