In bioprocess engineering the Qualtiy by Design (QbD) initiative encourages the use of models to define design spaces. However, clear guidelines on how models for QbD are validated are still missing. In this review we provide a comprehensive overview of the validation methods, mathematical approaches, and metrics currently applied in bioprocess modeling. The methods cover analytics for data used for modeling, model training and selection, measures for predictiveness, and model uncertainties. We point out the general issues in model validation and calibration for different types of models and put this into the context of existing health authority recommendations. This review provides a starting point for developing a guide for model validation approaches. There is no one-fits-all approach, but this review should help to identify the best fitting validation method, or combination of methods, for the specific task and the type of bioprocess model that is being developed.
Highlights
Novel machine learning method based on recurrent neural networks for the accurate prediction of upstream cultivation processes.
Insights into the temporal evolution of main upstream parameters and outcomes.
High predictive capability, straightforward implementation and broad transferability of the proposed approach.
New insights into the influence of parameter changes for process outcomes.
Introduction of new analysis methods like autocorrelation functions for the detailed study of experimental and computational data.
Pilot and prototyping scale investigations were undertaken in order to evaluate the technical feasibility of producing value-added biopolymers (polyhydroxyalkanoates (PHAs)) as a by-product to essential services of wastewater treatment and environmental protection. A commonly asked question concerns PHA quality that may be expected from surplus biomass produced during biological treatment for water quality improvement. This paper summarizes the findings from a collection of investigations. Alongside the summarized technical efforts, attention has been paid to the social and economic networks. Such networks are needed in order to nurture circular economies that would drive value chains in renewable resource processing from contaminated water amelioration into renewable value-added bioplastic products and services. We find commercial promise in the polymer quality and in the process technical feasibility. The next challenge ahead does not reside so much any more in fundamental research and development of the technology but, rather, in social-economic steps that will be necessary to realize first demonstration scale polymer production activities. It is a material supply that will stimulate niche business opportunities that can grow and stimulate technology pull with benefit of real life material product market combinations.
This paper presents a systematic investigation into monomer development during mixed culture Polyhydroxyalkanoates (PHA) accumulation involving concurrent active biomass growth and polymer storage. A series of mixed culture PHA accumulation experiments, using several different substrate-feeding strategies, was carried out. The feedstock comprised volatile fatty acids, which were applied as single carbon sources, as mixtures, or in series, using a fed-batch feed-on-demand controlled bioprocess. A dynamic trend in active biomass growth as well as polymer composition was observed. The observations were consistent over replicate accumulations. Metabolic flux analysis (MFA) was used to investigate metabolic activity through time. It was concluded that carbon flux, and consequently copolymer composition, could be linked with how reducing equivalents are generated.
We present explainable machine learning approaches for the accurate prediction of solvation free energies, enthalpies, and entropies for different ion pairs in various protic and aprotic solvents. As key input...
The coupling of individual models in terms of end-to-end calculations for unit operations in manufacturing processes is a challenging task. We present a probability distribution-based approach for the combined outcomes of parametric and non-parametric models. With this so-called Bayesian predictive ensemble, the statistical moments such as mean value and standard deviation can be accurately computed without any further approximation. It is shown that the ensemble of different model predictions leads to an uninformed prior distribution, which can be transformed into a predictive posterior distribution using Bayesian inference and numerical Markov Chain Monte Carlo calculations. We demonstrate the advantages of our method using several numerical examples. Our approach is not restricted to certain unit operations, and can also be used for the more robust interpretation and assessment of model predictions in general.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.