Thermodynamic and kinetic model for bacterial yield prediction

Bacterial yield, defined as the amount of biomass that can be formed per unit of substrate consumed, is a critical parameter in any biological systems. The estimation of bacterial net yield requires the consideration of the true yield and bacterial decay. Thermodynamic methods to predict true yield have been widely used in biotechnology and environmental engineering. However, true yield predictions are often inaccurate. The first part of this research identifies the causes of the high prediction errors of the current thermodynamic model. An improved true yield prediction model was developed and the energy utilization efficiency, K, was estimated as 0.41 from a priori analysis. The application of the expanded model was demonstrated in multiple growth situations. A large data set of reported true yields is presented. Significant causes of model uncertainty and literature-reported yield value uncertainties are described. Evaluation of the model with experimental yield values shows good predictive ability. However, the wide range in reported yields limits comprehensive validation. A more complete understanding of bacterial decay becomes necessary to overcome the uncertainty of experimental yield data.;Bacterial decay plays a very important role during biological processes, especially when the substrate level is low. However, its mechanism remains poorly understood. This research proposes a conceptual model of bacterial decay. Based on the conceptual decay model, a novel full net yield prediction model was developed and its four parameters were estimated. These estimated parameters work for any aerobic neutral systems and no fitting parameters are needed during model application. The absolute estimation errors of net yield prediction are 8.5+/-2% for the dataset presented. The low estimation errors show the net yield prediction model has very promising prediction ability.;More experimental bacterial yield data are still needed for further evaluating and developing bacterial yield prediction models. Thus, a relational database was built using Microsoft Access in order to organize the collected yield data. The expanded true yield prediction and the full net yield prediction model proposed in this research were coded into the database using Visual Basic for Applications. Tests show the database works efficiently.