immunebuilder

Predicts immune-protein structures from sequence or FASTA inputs using GPU-accelerated Docker models.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/yjcyxky/biominer-skills --skill immunebuilder
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: immunebuilder
Source: https://github.com/yjcyxky/biominer-skills/tree/main/immunebuilder
Command: npx skills add https://github.com/yjcyxky/biominer-skills --skill immunebuilder

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill accelerates accurate structure prediction for immune proteins (antibodies, nanobodies, and T-cell receptors) using specialized models in ImmuneBuilder.

Core Features & Use Cases

  • Fast, accurate predictions for heavy-light antibodies (ABodyBuilder2), single-domain antibodies (NanoBodyBuilder2), and TCRs (TCRBuilder2).
  • GPU-accelerated Docker workflows that run images like ghcr.io/open-prophetdb/immunebuilder:arm64-blackwell on supported platforms.
  • Use Case: Pharmaceutical researchers can rapidly iterate designs by predicting 3D structures from sequence inputs or FASTA files to guide optimization and selection.

Quick Start

Launch the ImmunBuilder Docker image with GPU support and provide sequences or FASTA inputs to generate PDB models. Example: docker run --rm --gpus all -v $(pwd):/data ghcr.io/open-prophetdb/immunebuilder:arm64-blackwell ABodyBuilder2 -H "HEAVY" -L "LIGHT" -o /data/output.pdb

Frequently Asked Questions about immunebuilder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I predict antibody structure from a FASTA sequence?

To predict antibody structure from FASTA sequences, provide heavy and light chain sequences to ABodyBuilder2 within the GPU-accelerated Docker container to generate PDB output files.

Can I use GPU-accelerated Docker for nanobody structure prediction?

Yes, you can run nanobody structure prediction using GPU-accelerated Docker images like ghcr.io/open-prophetdb/immunebuilder:arm64-blackwell with NanoBodyBuilder2 to generate 3D PDB models.

What is needed to predict TCR 3D structures from sequence inputs?

Predicting TCR 3D structures requires running the TCRBuilder2 model inside the Docker image with GPU support, accepting sequence or FASTA inputs to output PDB files.

Does immune protein structure prediction work for both single-domain and heavy-light antibodies?

Yes, immune protein structure prediction supports both single-domain antibodies using NanoBodyBuilder2 and heavy-light paired antibodies using ABodyBuilder2, outputting PDB format structural models.

Are there limitations when running immune protein prediction without GPU support?

This immune protein prediction workflow requires Docker images with GPU support, meaning environments lacking GPU acceleration cannot execute the specialized structure prediction models.