Self-regulating microfluidic system for lipid nanoparticle production.

Area of Science:

• Biotechnology
• Nanotechnology
• Computational Science

Background:

• Lipid nanoparticles (LNPs) are crucial for gene delivery, enabling advanced vaccines and cancer therapies.
• Optimizing LNP formulation is complex due to the vast design space.
• In silico methods offer a cost-effective approach to LNP formulation optimization.

Purpose of the Study:

• To demonstrate data-driven automation for rapid lipid nanoparticle formulation parameterization.
• To integrate computational fluid dynamics (CFD) and microfluidic screening with machine learning.
• To enable fully automated LNP formulation optimization with minimal human intervention.

Main Methods:

• Utilized low-cost microfluidic hardware for rapid formulation parameterization.
• Employed computational fluid dynamics (CFD) simulations and microfluidic screenings.
• Developed machine learning algorithms to predict critical quality attributes like size and encapsulation efficiency.

Main Results:

• Successfully derived machine learning models from CFD and microfluidic data.
• Deployed a self-regulating microfluidic device for automated LNP formulation.
• Achieved on-the-fly optimization of lipid nanoparticle formulation.

Conclusions:

• Data-driven automation combined with microfluidics accelerates LNP formulation optimization.
• This approach bridges computational and experimental work for efficient LNP design.
• The developed method promises to speed up the discovery of optimized nanoparticles for therapeutic applications.