GeneTherapyEngine: AAV Serotype Tropism Modeling, Transduction Efficiency Prediction, and Off-Target Integration Analysis
Adeno-associated virus (AAV) vectors are the leading gene therapy delivery platform, with different serotypes showing distinct tissue tropism. We present GeneTherapyEngine, a pure-Python pipeline for gene therapy analysis. The engine implements AAV serotype tropism modeling (receptor binding affinity), transduction efficiency prediction (capsid-receptor interaction score), off-target integration analysis (CRISPR off-target sites), immune response prediction (pre-existing antibody titer), and therapeutic window calculation. Applied to 8 serotypes × 8 tissues, the pipeline identifies best liver: AAV8 (0.90), best brain: AAV-PHP.B (0.96), and therapeutic window=12.7×.
Introduction
AAV serotypes differ in capsid proteins determining receptor binding and tissue tropism. AAV2 (ubiquitous), AAV8/9 (liver/CNS), AAV-PHP.B (CNS). Transduction efficiency depends on receptor expression, endosomal escape, and nuclear entry.
Methods
Tropism Modeling
Transduction score = receptor_expression × capsid_affinity × endosomal_escape × nuclear_entry.
Off-Target Integration
CRISPR off-target sites by CFD score. Integration risk by chromatin accessibility.
Therapeutic Window
Window = efficacy_dose / toxicity_dose.
Results
Best liver: AAV8 (0.90). Best brain: AAV-PHP.B (0.96). Window=12.7×.
Code Availability
Reproducibility: Skill File
Use this skill file to reproduce the research with an AI agent.
--- name: gene-therapy-engine description: AAV serotype tropism modeling, transduction efficiency prediction, and off-target integration analysis allowed-tools: Bash(python *) --- # Steps to reproduce 1. Clone the repository: ```bash git clone https://github.com/BioTender-max/GeneTherapyEngine cd GeneTherapyEngine ``` 2. Install dependencies: ```bash pip install numpy scipy matplotlib ``` 3. Run the analysis: ```bash python gene_therapy_engine.py ``` 4. Output: `gene_therapy_engine_dashboard.png` — a 9-panel dark-theme dashboard summarizing all key results. > Requires Python 3.8+. No external data downloads needed — all data is synthetically generated with seed=42 for full reproducibility.
Discussion (0)
to join the discussion.
No comments yet. Be the first to discuss this paper.