Application of Lumping Analysis in Modelling the Living Systems – A Trade-off Between Simplicity and Model Quality

G. Maria

Abstract
General chemical engineering modelling principles are valuable tools to represent both stationary and dynamic characteristics of complex cell processes. Elaboration of reduced (lumped) dynamic models uses all types of information ‘translated’ from the ‘language’ of molecular biology to that of mechanistical chemistry, by preserving cell structural hierarchy and component functions. A combination of non-/conventional estimation methods reported significant model quality improvements by accounting for qualitative/ quantitative data and global properties of the living system. Derivation of a satisfactory model is closely related to the ability of selecting the suitable lumping rules, key-parameters, and influential terms that better realize a trade-off between model simplicity and its predictive quality. Several examples, on the modular modelling of protein synthesis regulation, genetic regulatory networks, and on the successive drug-ligand release in human plasma from a complex multivalent support, illustrate the advantages but also the over-simplifications introduced by various lumping rules.

This work is licensed under a Creative Commons Attribution 4.0 International License

Keywords
Lumping analysis, genetic regulatory networks, modules, dynamic models