Synergising stoichiometric modelling with artificial neural networks to predict antibody glycosylation patterns in Chinese hamster ovary cells

• Published in: Computers & chemical engineering Vol. 154; p. 107471
• Main Authors: Antonakoudis, Athanasios, Strain, Benjamin, Barbosa, Rodrigo, Jimenez del Val, Ioscani, Kontoravdi, Cleo
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.11.2021
• Subjects: Artificial neural network; Constraint-based modelling; Flux balance analysis; Hybrid modelling; Monoclonal antibody; Protein glycosylation; Flux balance analysis; Artificial neural network; Monoclonal antibody; Protein glycosylation; Hybrid modelling; Constraint-based modelling
• ISSN: 0098-1354; 1873-4375
• DOI: 10.1016/j.compchemeng.2021.107471

•Stoichiometric model accurately describes CHO cell metabolism.•Artificial neural network used to model secreted antibody glycosylation.•Hybrid framework links extracellular culture conditions to recombinant protein quality.•Framework can be updated in-process with commonly monitored process variables.•Sets basis for bioprocess control. In-process quality control of biotherapeutics, such as monoclonal antibodies, requires computationally efficient process models that use readily measured process variables to compute product quality. Existing kinetic cell culture models can effectively describe the underlying mechanisms but require considerable development and parameterisation effort. Stoichiometric models, on the other hand, provide a generic, parameter-free means for describing metabolic behaviour but do not extend to product quality prediction. We have overcome this limitation by integrating a stoichiometric model of Chinese hamster ovary (CHO) cell metabolism with an artificial neural network that uses the fluxes of nucleotide sugar donor synthesis to compute the profile of Fc N-glycosylation, a critical quality attribute of antibody therapeutics. We demonstrate that this hybrid framework accurately computes glycan distribution profiles using a set of easy-to-obtain experimental data, thus providing a platform for process control applications.