skills.rest
adaptyvbio
adaptyvbioOFFICIAL
Explore all skills by this author →
WebsiteGitHub

proteinmpnn

Design protein sequences from structural backbones using ProteinMPNN inverse folding.

151|20|Updated Jan 15, 2026
Search Tags:#inverse-folding#multi-state-design#sequence-design#protein-design#backbone-design#ProteinMPNN
One-click install
npx skills add https://github.com/adaptyvbio/protein-design-skills --skill proteinmpnn
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: proteinmpnn
Source: https://github.com/adaptyvbio/protein-design-skills/tree/main/skills/proteinmpnn
Command: npx skills add https://github.com/adaptyvbio/protein-design-skills --skill proteinmpnn

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables automated design of protein sequences using ProteinMPNN inverse folding, turning structural backbones into viable sequences.

Core Features & Use Cases

  • Inverse-folding sequence design: generate sequences that fit a given backbone.
  • Redesign and constraint handling: modify existing sequences and apply fixed-position constraints.
  • Multi-state and expression considerations: optimize sequences for stability and expression, including multi-state design.

Quick Start

Run a basic design for a given backbone with: python protein_mpnn_run.py --pdb_path backbone.pdb --out_folder output/ --num_seq_per_target 16 --sampling_temp "0.1"

Frequently Asked Questions about proteinmpnn

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

FAQPage Schema
How do I design protein sequences for a specific structural backbone?▼

Inverse folding sequence design uses ProteinMPNN to turn a structural backbone into viable sequences. You provide a PDB backbone and it generates sequences that fit the structure, controlled by a sampling temperature.

Can I apply fixed-position constraints when redesigning an existing protein sequence?▼

Yes, you can apply fixed-position constraints during sequence redesign to keep specific residues locked in place. This allows you to modify existing proteins while preserving critical structural or functional motifs.

Does ProteinMPNN multi-state design optimize sequences for stability and expression?▼

Yes, multi-state design optimizes protein sequences for stability and expression considerations. It generates sequences tailored to multiple structural states to improve overall folding viability and expression success.

Do I need a GPU and CUDA to run ProteinMPNN sequence design?▼

Yes, you need Python 3.8+, CUDA 11.x or higher, and a compatible GPU with sufficient VRAM. A Biomodals modal workflow is also required to execute the inverse folding scripts and generate output sequences.

How do I run a basic ProteinMPNN inverse folding task from a PDB file?▼

Run a basic inverse folding task by executing the Python script with arguments like --pdb_path for your backbone and --out_folder for the output. Use --num_seq_per_target to specify the number of generated sequences.

Related Skills

Explore More Education & Research Skills→
junior1pjunior1p
11

setup

Automate Modal CLI installation, authentication, and biomodals repository bootstrap for protein design toolchains.

Community
Intermediate
InternScienceInternScience
28

molclaw-sequence-valid-check

Validate protein sequences and return per-sequence boolean flags and messages.

Official
Basic
zongtingweizongtingwei
25

rfdiffusion

Generate de novo protein backbones with a diffusion-based generative model.

Community
Advanced
zongtingweizongtingwei
25

proteinmpnn

Design protein sequences via ProteinMPNN inverse folding on given backbones.

Community
Advanced