A Mechanistic Digital Twin for Dual-AAV Bioprocessing and Multi-Gene Base-Editing Payload Delivery
Abstract Dual-AAV delivery of oversized editors couples vector manufacturing, product quality, co-delivery and gene correction, yet process models rarely carry this coupling or state what their data can identify. We built a modular computational digital twin spanning manufacturing, dual-vector reconstitution and gene-specific correction. An independent reconstruction of a published 21-state production model agreed across three stiff solvers (< 0.14%) but retained a 21.8% mean absolute gap to source observations at 46 h; its 14-parameter space was rank-deficient (12/14; condition number 6.4×10²¹), improving to 6.0×10³ after reduction. Fourteen scenario studies traced product quality to function: within the assumed surrogate equations and input distributions, variance-based sensitivity assigned 52% of functional variance to the A:B ratio (correlation screen: 2%), and titre-only release classified function at chance (AUC 0.49) versus 0.98 for a functional assay. All analyses are computational; no experimental validation is claimed.
Authors
- Faizan Ahmad (ORCID: https://orcid.org/0000-0002-7768-9411)
Institutions
- Jamia Hamdard (IN)
Publication Details
- Journal
- Research Square
- Published
- 2026-11-17
- DOI
- https://doi.org/10.21203/rs.3.rs-11747176/v1
- Primary Topic
- Virus-based gene therapy research
- Type
- preprint