Generate protein sequences, structures, and embeddings using ESM3 and ESM C models.
npx skills add https://github.com/ovachiever/droid-tings --skill esm-ovachiever
Please help me install this Agent Skill. Skill: esm Source: https://github.com/ovachiever/droid-tings/tree/main/skills/esm Command: npx skills add https://github.com/ovachiever/droid-tings --skill esm-ovachiever
This Skill provides a comprehensive toolkit for protein language models, enabling work with ESM3 for multimodal protein design (sequence, structure, function) and ESM C for efficient embeddings. It helps design novel proteins, generate embeddings, and run both local models and cloud-based Forge API inference to scale experimentation.
Example: Load a small esm3 model locally or via Forge, generate a sequence with masked regions, and obtain embeddings for downstream tasks.
High-intent search queries and answers about installing and using this skill.
ESM3 enables multimodal protein design by generating sequences, structures, and functional annotations. Load the model locally or via Forge API, configure GenerationConfig parameters (num_steps, temperature, top_p), and use ESMProtein to design novel proteins or complete partial sequences end-to-end.
Yes. ESM and ESM C models derive protein embeddings for downstream ML applications like clustering and similarity search. Generate embeddings locally or at scale via Forge API, then export them for analysis in your ML pipeline.
Local execution uses ESMProtein for direct model inference on your hardware; Forge API enables cloud-based scalable inference and batch processing. Choose local for development and experimentation, Forge for production-scale workloads and distributed inference.
Yes. ESM3 predicts structures directly from sequences as part of its multimodal generation pipeline. Configure the model with sequence input and structure prediction parameters to obtain 3D structure predictions without additional tools.
Yes. ESM3 supports inverse folding (sequence design from structures) and coordinates conditioning through GenerationConfig parameters. Provide structure or coordinate constraints to guide sequence generation for targeted protein design.
Load ESM3 or ESM C models and initialize ESMProtein client with your environment. No external dependencies are required; configure GenerationConfig with design parameters (num_steps, temperature, top_p) and begin generating sequences, structures, or embeddings immediately.
AlterLab-IEUDesign and analyze protein sequences and structures with multimodal AI.
silversteinGenerate protein sequences, predict structures, and compute embeddings with ESM3 and ESM C.
jackspaceDesign or optimize protein sequences, structures, and functions using ESM3 and ESM C models.
jimmc414Generate protein sequences and extract embeddings using ESM3 and ESM C.
