Generate sequences, structures, functions, and embeddings with ESM3 and ESM C via Forge API or locally.
npx skills add https://github.com/OwnLabAI/ownlab --skill esm-ownlabai
Please help me install this Agent Skill. Skill: esm Source: https://github.com/OwnLabAI/ownlab/tree/main/mart/skills/scientific-skills/esm Command: npx skills add https://github.com/OwnLabAI/ownlab --skill esm-ownlabai
Protein researchers need a unified toolkit to design, embed, and optimize proteins across sequence, structure, and function, with options for local or cloud-based inference.
Provide a partial or full protein sequence (and optional coordinates) to start a multimodal design flow.
High-intent search queries and answers about installing and using this skill.
To design a protein sequence using ESM3, provide your partial sequence and optional coordinates to the ESMProtein interface. The multimodal workflow uses GenerationConfig to condition sequence generation on your structural input.
You can run ESM3 protein workflows locally using ESMProtein interfaces, or use the optional Forge API for scalable cloud inference and batch processing. Forge access is required only for utilizing cloud-based models.
Protein embeddings in ESM C and ESM3 workflows are used for downstream modeling, clustering, and similarity search. They provide numerical representations of sequences to enable embedding-based screening and analysis.
The best way to compare protein embeddings across ESM3 and ESM C is by running cross-model comparisons within your workflow. Generate embeddings from both models and analyze the outputs for similarity and functional differences.
Yes, multimodal protein generation with ESM3 supports function prediction. The workflow generates sequences, structures, and functions simultaneously, allowing you to predict and optimize protein functions alongside structural design.
To scale batch processing for protein structure prediction using ESM3, use the Forge API for scalable cloud inference. This enables batch processing of multiple sequences and structures without local computational bottlenecks.
ovachieverGenerate protein sequences, structures, and embeddings using ESM3 and ESM C models.
dralkhConvert SMILES and RDKit molecules into numerical feature vectors.
SologaGenerate protein sequences and embeddings with ESM3 and ESM C.
PKU-YuanGroupConverts PDF documents to clean, structured Markdown.
