Our lab employs a multidisciplinary approach to study the relationship between gene circuits and their dynamical properties in bacteria: quantitative fluorescence microscopy and microfluidics to capture temporal dynamics and cellular heterogeneities, molecular genetics to dissect biological mechanisms, machine learning to uncover patterns, and mathematical modeling to rebuild biological systems and predict experimental outcomes. Our current research topics fall into the following general categories:
Temporal signal processing
Much like consciousness emerges from interactions within networks of neurons, gene regulatory networks can give rise to novel emergent properties that cannot be understood or predicted from their individual components in isolation. Even simple circuits comprising just two or three genes can generate sophisticated temporal behaviors in bacteria, including stochastic pulsing, circadian oscillations, fold-change detection, noise filtering, and many others. We are investigating how various circuit attributes—such as network topology, reaction kinetics, and stoichiometry—shape temporal outputs and signal processing capabilities of gene circuits and to what extent these properties can be tuned experimentally.
Phenotypic heterogeneity
There is a prevailing notion in biology that phenotypes are determined by the interplay between genetics and environmental factors. Yet, bacterial cells with identical genomes and external conditions could exhibit strikingly different behaviors. A particularly relevant example is the spontaneous and transient emergence of antibiotic-tolerant cells within a genetically uniform population. We would like to uncover the gene regulatory circuits that are prone to drive phenotypic heterogeneities, unravel the key dynamical properties that enable stochastic cell-state transitions, and elucidate their physiological roles in the context of bacterial stress adaptation.
New methodologies
We are developing advanced microfluidic and imaging platforms to uncover and characterize dynamic processes in single bacterial cells and microbial communities. A key objective is to simultaneously achieve high temporal resolution, single-cell sensitivity, and high throughput, such that transient and heterogeneous events are faithfully and unambiguously captured under well-defined external conditions. In parallel, we are working on statistical and machine learning models for inferential and predictive analyses of circuit-driven gene expression dynamics, offering a scalable alternative to traditional differential equation-based approaches.
We are grateful for the trust and generosity of our funding agencies, whose support allows us to pursue cutting-edge research with curiosity and purpose.