(368g) Digital Twin in Microbial Process Optimization

Microbes inhabit chemically and hydrologically dynamic world. Their spreading and proliferation are regulated by nutrient availability and flow dynamics, which can be better understood using chemical engineering concepts. This study combines laboratory-scale experiments with mechanistic modeling of multiphysics (including mass transfer, reaction kinetics, and fluid mechanics) to quantify bacterial transport and metabolism processes. Our work provides a quantitative analysis of critical parameters influencing bacterial activities, guiding strategies to maximize microbial efficiency in various applications.

Thorough mixing between microbes and chemicals is a prerequisite for processes such as bioremediation. We modeled bacterial spreading using a modified conservation equation that incorporates bacterial motility, chemical landscapes, and fluid velocity at two scales: a 15 cm-long chromatography column and a microfluidic porous network. Our experimentally tuned model identified critical parameter ratios explaining the balance between opposing effects on bacterial distribution. These findings led to the determination of an optimal flow velocity that enhances mixing efficiency between bacteria and chemicals.

The metabolism of sugars to by oral microbes to produce lactic acid is the primary cause of dental plaque and cavities. We formulated a kinetic model based on Monod-type equation to describe the impact of flowing media on bacterial growth. This kinetic model, tuned by data from a static batch system, predicted temporal changes in bacterial viability under a dynamic system with continuous medium flow. In static conditions, nutrient depletion and toxin accumulation caused cell viability to peak at a specific time, depending on sugar type and concentration. Comparatively, continuous fresh media flow resulted in a slower but sustained increase in cell density. These findings shed light on the tradeoff between nutrient availability and convection flushing in continuous cell cultures.

Research Interests: FDA’s Quality by Design guidelines and demand for biosimilars give rise to process optimization and intensification, which requires a comprehensive understanding of cell metabolism and bioreactor-related parameters. In my future career, I aim to leverage my expertise in microbiology and mechanistic modeling, or digital twin, to identify the relationship between critical process parameters and quality attributes to optimize process design. Additionally, I seek to expand my knowledge in process analytical technology (PAT) and artificial intelligence (AI), as PAT offers real-time, nondestructive monitoring of biologics process, and AI is emerging as a powerful tool for processing large, multivariate data sets. Integrating PAT and AI with mechanistic modeling will facilitate the construction of a network of dependencies between real-time data and underlying meta-information, advancing process development and manufacturing.