Abstract
AbstractWe present a workflow for kinetic modeling of biocatalytic reactions which combines methods from Bayesian learning and uncertainty quantification for model calibration, model selection, evaluation, and model reduction in a consistent statistical framework. Our workflow is particularly tailored to sparse data settings in which a considerable variability of the parameters remains after the models have been adapted to available data, a ubiquitous problem in many real‐world applications. Our workflow is exemplified on an enzyme‐catalyzed two‐substrate reaction mechanism describing the symmetric carboligation of 3,5‐dimethoxy‐benzaldehyde to (R)‐3,3′,5,5′‐tetramethoxybenzoin catalyzed by benzaldehyde lyase from Pseudomonas fluorescens. Results indicate a substrate‐dependent inactivation of enzyme, which is in accordance with other recent studies.
Highlights
Modeling is a difficult task with many challenges
We have introduced a workflow for model calibration, selection and model reduction based on statistical Bayesian sampling, which was exemplified on the symmetric carboligation of 3,5-DMBA to (R)-3,30,5,50-TMB catalyzed by benzaldehyde lyase (BAL)
Overall plausibility of the stochastic model was investigated via residual analysis and parametric bootstrapping
Summary
Modeling is a difficult task with many challenges. A good model is predictive and helps to get deeper insight into the described system or phenomenon by, for example, explaining underlying mechanisms or giving raise to nonobvious hypotheses which can be tested in a subsequent step. We introduce a modeling workflow for parameter estimation, model selection, model reduction, and validation based on Bayesian statistics, which is tailored for consistent uncertainty quantification, and compare it to a similar workflow which uses local methods.[11] we discuss different ways to visualize outcomes of individual steps in the workflow. We regard such a workflow as a prerequisite for an automated data and model management system, which makes modeling results transparent and reproducible and facilitates standardization of processes. Results confirm previous recent findings about this reaction mechanism[4,5,11] and illustrate that global methods such as sampling-based analysis provide superior insights into underlying parameter dependencies compared to local approximations
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
