Mustafa Khammash

This paper introduces the Filtered Finite State Projection (FFSP) method to solve high-dimensional stochastic filtering problems in biochemical reaction networks, demonstrating its effectiveness with numerical examples and real biological data.

This paper develops a control-theoretic framework for engineering closed-loop cell-based therapies using nullgram profiling and a digital twin platform to achieve robust regulation of cellular behavior.

This paper introduces and analyzes an antithetic dual-rein integral feedback controller for stochastic chemical reaction networks, showing it can control first moments and maintain tunable variance bounds.

This paper introduces a multi-layer feedback control strategy for synthetic biology systems, demonstrating precise regulation in gene expression and population growth across scales.

This paper introduces cascaded antithetic integral feedback to improve dynamic performance and robust stability in synthetic biology systems.

This paper presents three biomolecular anti-windup topologies to prevent instability in integral controllers by linking control-theoretic strategies to chemical reaction networks.

The paper introduces a method using Mixture Density Networks to efficiently compute Fisher Information and perform policy search in stochastic chemical reaction networks with partial observations.

The paper demonstrates how nonlinear sequestration reactions in antithetic integral controllers can improve dynamic performance and reduce variability while maintaining robust adaptation.

This paper introduces a frequency domain method to estimate intrinsic eigenvalues of stochastic reaction networks using a small number of simulated trajectories, reducing computational burden.

This paper improves the light sensitivity of Magnets photosensors through directed evolution, enabling tunable optogenetic regulation in bacteria and mammalian cells.

The paper introduces a divide-and-conquer method to solve the Chemical Master Equation in high-dimensional gene expression systems, demonstrating its effectiveness in analyzing yeast transcription at single-cell resolution.

This paper identifies structural requirements for maximal robust perfect adaptation in biomolecular networks using linear-algebraic conditions and demonstrates their application through examples.

This paper presents a platform for dynamic control of bacterial co-culture composition using optogenetic feedback and develops a framework for modeling synthetic genetic circuits in E. coli.

This paper develops a framework using split inteins to design and test biomolecular controllers that achieve robust perfect adaptation in mammalian cells.

The paper implements and optimizes a novel optogenetic tool in mammalian cells using a split T7 RNA polymerase system for light-inducible gene expression control.

This paper develops a spectral theory and computational methods to analyze frequency spectra of noisy cellular networks, enabling topology discrimination and synthetic oscillator optimization.

The paper uses optogenetic actuator and ERK biosensor circuits to study MAPK network dynamics and identifies RSK2-mediated feedback as critical for ERK regulation and drug response.

This paper designs biomolecular PID controllers to achieve robust perfect adaptation and improved dynamic performance in synthetic biology systems.

The paper designs synthetic gene networks that decode dynamic signals into gene expression patterns, showing increased information transmission capacity through dynamic multiplexing.

This paper introduces a deep learning framework to solve high-dimensional chemical master equations using reinforcement learning and neural networks, enabling efficient approximation of solutions and their parameter sensitivities.

This work introduces log derivatives to analyze and design biomolecular circuits for robust stability and control through structural properties.

This paper introduces synthetic integral and proportional-integral feedback controllers in mammalian cells to achieve robust perfect adaptation, showing improved stability and reduced variability.

This paper introduces a likelihood-free nested sampling method for Bayesian inference in stochastic biochemical reaction networks, providing an unbiased evidence estimator and posterior samples with a novel termination criterion.

This paper proves that ACR systems contain an internal integral structure, enabling robustness to broader disturbances and facilitating modular synthetic biology design.

This study identifies gene expression burden in mammalian cells and introduces miRNA-based circuits to mitigate it, improving synthetic circuit performance.

The paper introduces a cell-in-the-loop approach to synthetic biology pattern formation, demonstrating checkerboard patterning through optogenetic signaling emulated in silico.

This paper analyzes delayed stochastic reaction networks and shows that phase-type delays do not affect ergodicity, while introducing a controller for stability.

The paper uses temporal perturbations of Erk dynamics to reveal the network architecture of FGF2-MAPK signaling and how it influences cell fate decisions.

This paper designs and implements an integral feedback controller in E. coli to achieve robust and tunable protein regulation.