Bayesian filtering for model predictive control of stochastic gene expression in single cells.

Area of Science:

• Cellular and Molecular Biology
• Systems Biology
• Biophysics

Background:

• Gene expression is inherently stochastic, leading to cell-to-cell variability.
• Controlling protein production at the single-cell level is crucial for understanding cellular dynamics and developing synthetic biology applications.
• Previous methods often rely on population averages, limiting precision in individual cells.

Purpose of the Study:

• To develop and validate a method for precise control of protein production in individual cells.
• To compare the efficacy of single-cell stochastic control with traditional population-based approaches.
• To address discrepancies between deterministic and stochastic models of gene expression.

Main Methods:

• Utilized modern microscopy and optogenetics for targeted light application to individual cells.
• Employed a finite state projection based stochastic model of gene expression.
• Integrated Bayesian state estimation for real-time control of protein copy numbers.
• Compared the developed method against population-based control strategies.

Main Results:

• Successfully controlled protein copy numbers within individual cells with high precision.
• Demonstrated superior performance of the single-cell stochastic control method compared to population-based approaches.
• Showcased the ability of the control strategy to reconcile differences between deterministic and stochastic gene expression models.

Conclusions:

• The developed stochastic control method offers precise regulation of protein production at the single-cell level.
• This approach enhances the accuracy of gene expression control and model predictions.
• The findings have implications for synthetic biology, cell-based therapies, and fundamental research in gene regulation.