skills.rest
PharMolix
PharMolixCOMMUNITY
Explore all skills by this author →
GitHub

mutation-design-aav

Generate a 96-mutant AAV VP1 library with high fitness and diversity.

1.1k|132|Updated Apr 13, 2023
Search Tags:#design#protein#mutants#sequence-optimization#aav#oracle-model
One-click install
npx skills add https://github.com/PharMolix/OpenBioMed --skill mutation-design-aav
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: mutation-design-aav
Source: https://github.com/PharMolix/OpenBioMed/tree/main/skills/mutation-design-aav
Command: npx skills add https://github.com/PharMolix/OpenBioMed --skill mutation-design-aav

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Propose high-fitness and high-diversity mutants of the VP1 capsid protein of Adeno-Associated Virus (AAV) through multi-round iterative optimization.

Core Features & Use Cases

  • Automates multi-round design of a 28-amino-acid VP1 segment to identify variants with improved DNA packaging fitness and sequence diversity.
  • Downloads and loads an in-silico oracle AAV predictor and uses ESM2 embeddings to score candidate sequences.
  • Maintains population diversity using average pairwise Hamming distance and outputs a final 96-mutant library.

Quick Start

Run the mutation-design-aav skill to generate a final CSV of 96 AAV mutants with the highest predicted fitness and diversity.

Frequently Asked Questions about mutation-design-aav

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

FAQPage Schema
How do I design high-fitness AAV VP1 capsid mutants?▼

This skill designs AAV VP1 capsid mutants by proposing high-fitness variants through multi-round iterative optimization, scoring them with an in-silico oracle predictor and ESM2 embeddings, and outputting a 96-mutant library.

What is iterative sequence optimization for protein engineering?▼

Iterative sequence optimization is a multi-round process that proposes, evaluates, and diversifies protein mutant sequences to identify high-fitness variants while maintaining population diversity across up to 10 rounds.

Do I need PyTorch and Python 3.9 to run AAV mutation design?▼

Yes, AAV mutation design requires Python 3.9+, PyTorch, NumPy/Pandas, protein sequence tools, ESM2 embeddings, and access to the oracle predictor to evaluate candidate sequences.

How do I generate a 96-mutant AAV library with high diversity?▼

You generate a 96-mutant AAV library by running up to 10 rounds of iterative sequence optimization that maintains diversity using average pairwise Hamming distance, outputting a final CSV of sequences and predicted fitness.

What format is the final AAV mutant library output in?▼

The final AAV mutant library is output as a CSV file listing the proposed 28-amino-acid VP1 segment sequences and their predicted DNA packaging fitness scores.

Are there limitations to using an in-silico oracle model for AAV design?▼

The AAV design is limited to a 28-amino-acid VP1 segment and relies on an in-silico oracle predictor for fitness scores, requiring experimental validation to confirm actual DNA packaging fitness.

Related Skills

Explore More Education & Research Skills→
junior1pjunior1p
11

esm

Compute ESM2 embeddings and pseudo-log-likelihood scores for protein sequences.

Community
Intermediate
PharMolixPharMolix
1.1k

mutation-design-gfp

Design GFP mutant libraries through iterative optimization and export 96-sequence CSV files.

Community
Advanced
zongtingweizongtingwei
25

esm2-sequence-scoring

Score protein sequences and generate embeddings using the ESM2 model.

Community
Intermediate
jaechang-hitsjaechang-hits
298

adaptyv-bio

Submit cell-free protein expression and binding affinity experiments via Python SDK and REST API.

Community
Advanced