Microbial Communities and Biofilms

Microbial biofilms are critically important in medical, environmental and engineered biological systems. While foundational to the vast majority of microbial life on the planet, the basic design principles of microbial biofilms are still poorly understood. We are developing metabolic models of single species and multispecies biofilms that predict temporal and spatial variations with genome-scale resolution. The biofilm models are formulated by combining genome-scale reconstructions of species metabolism with transport equations that capture the relevant diffusional and convective processes within the biofilm. Our current models capture cellular growth and death, biofilm expansion and contraction, competition for nutrients, cross-feeding of metabolic byproducts, chemotaxis of motile species, secretion of small molecule inhibitors and antibiotic treatment. Simulations are performed to analyze the understand biofilm design principles and to develop improved antibiotic treatment strategies for biofilm eradication. This research is supported by NIH, NSF and ARO.

Renewable Biofuels and Biochemicals

An emerging conversion route for renewable biofuels and biochemical production is direct fermentation of waste gas streams and synthesis gas (syngas; mainly comprised of H2/CO/CO2) by specialized CO fermenting microbes. Commercial development favors bubble column reactors due to their high average mass transfer driving forces and long gas-liquid contact times. We are utilizing genome-scale metabolic reconstructions of CO fermenting bacteria to develop combined metabolic-transport models that capture the complex spatiotemporal behavior of syngas bubble column reactors. The bioreactor models are formulated by combining genome-scale reconstructions with transport equations that capture the key hydrodynamics of the multiphase gas-liquid system. Our current models provide both spatial and temporal predictions of cellular growth, dissolved gas consumption, byproduct synthesis, gas-liquid mass transfer and gas phase hydrodynamics. Simulations are performed to analyze the effects of column operating conditions and microbial engineering strategies on growth rates, product titers and volumetric productivities. This research is supported by NSF.

Biological Timekeeping

The circadian clock generates 24 hour rhythms that drive physiological and behavioral processes. Circadian disruptions have been associated with sleep loss, erratic wake times, reduced cognitive ability, obesity, diabetes, heart attacks, schizophrenia and Alzheimer’s disease. In mammals, circadian rhythms are generated by an internal self-sustained oscillator located in the hypothalamic suprachiasmatic nucleus (SCN) and consisting of approximately 20,000 heterogeneous neurons. We are developing multicellular network models to better understand neurotransmitter mediated intercellular communication within the SCN. Multicellular models are developed by coupling a heterogeneous collection of individual neurons according to a prescribed network connectivity pattern, which is derived from experiment to the extent possible. The resulting models are characterized by a high degree of heterogeneity with respect to single cell periodicity, neurotransmitter release and spatial organization, mimicking structural patterns within the SCN associated with distinct circadian functions. Dynamic simulations are performed to better understand the role of cellular and network heterogeneities on intercellular synchronization and entrainment to environmental cues. This research is supported by NIH.

Downstream Pharmaceutical Manufacturing

Downstream processing of Active Pharmaceutical Ingredients (APIs) typically involves several batch processing steps such as extraction, crystallization, filtration and drying. Removal of the solid API from liquid solvents via filtration and drying operations often represents a bottleneck in the manufacturing process. We are developing dynamic process models for API filtration and drying operations. Our initial work has focused on vacuum drying, including the formulation of combined heat-mass transport models and the development of a novel method for determining the drying end point based on on-line mass spectroscopy. The end detection algorithm was successfully tested on laboratory-scale and manufacturing-scale vacuum dryers by comparing model predictions to loss on drying measurements. Our current work is focused on incorporating the API particle size distribution, agitation rate and cake compaction effects into filtration and drying process models. Dynamic simulations are performed to provide insights into key design parameters and to enable process improvement and optimization. This research is supported by Sunovion Pharmaceuticals.