Identification of nonlinear kinetics of macroscopic bio-reactions using multilinear Gaussian processes

Abstract

In biological systems, nonlinear kinetic relationships between metabolites of interest are modeled for various purposes. Usually, little a priori knowledge is available in such models. Identifying the unknown kinetics is, therefore, a critical step which can be very challenging due to the problems of (i) model selection and (ii) nonlinear parameter estimation. In this paper, we aim to address these problems systematically in a framework based on multilinear Gaussian processes using a family of kernels tailored to typical behaviours of modulation effects such as activation and inhibition or combinations thereof. Using one such process as a model for each modulation effect leads to a much more flexible model than conventional parametric models, e.g., the Monod model. The resulting models of the modulation effects can also be used as a starting point for estimating parametric kinetic models. As each modulation effect is modeled separately, this task is greatly simplified compared to the conventional approach where the parameters in all modulation functions have to be estimated simultaneously. We also show how the type of modulation effect can be selected automatically by way of regularization, thus bypassing the model selection problem. The resulting parameter estimates can be used as initial estimates in the conventional approach where the full model is estimated. Numerical experiments, including fed-batch simulations, are conducted to demonstrate our methods.

An integrated approach for modeling and identification of perfusion bioreactors via basis flux modes

“…(i) the difficulty in guaranteeing global optimality of the parameter estimates due to their computation via numerical optimization algorithms that may either attain globally suboptimal solutions or result in intractable problems; and (ii) the identification of the correct model structure among a set of candidate model structures. Hence, the proposed procedure is in line with the purpose of other recently published methods but extends these methods in the sense that it circumvents the need to perform linear reparameterizations or to use modeling frameworks without a physical meaning [18,19]. To this end, the rational structure of the kinetic model is used to express the parameter estimation problem as a polynomial optimization problem where the cost and constraints are explicitly written as quadratic functions that involve only a few decision variables, which can be reformulated as a convex semidefinite program via the concept of sum-of-squares polynomials and sparse semidefinite relaxations [20].…”

A decade in review: use of data analytics within the biopharmaceutical sector

“…Another journal article with a small number of experiments was by Wang et al who applied Gaussian process (GP) multilinear regression to infer the modulation effect of four metabolites using six experiments as a training set and one for validation. A smaller number of experiments was sufficient as GP uses the generated estimates of modulation effects to estimate parametric models, which generate more data which resulted in more accurate predictions [ 37 ]. These applications demonstrate that smaller numbers of experiments can produce robust and accurate models, but the data needs to be representative of the whole process being modelled.…”

Safe model-based design of experiments using Gaussian processes

“…Polynomial chaos expansions have been used in stochastic model predictive control for the purposes of uncertainty propagation through a nonlinear model [67]. Similarly, Gaussian process (GP) [51,65] has been proposed as an alternative to PCE, as they are non-parametric models.…”