Identification and automatic segmentation of multiphasic cell growth using a linear hybrid model.

Area of Science:

• Mathematical Biology
• Biotechnology
• Microbiology

Background:

• Multiphasic cell growth is common in microorganisms but challenging to model.
• Existing nonlinear models can be complex and difficult to interpret.
• A need exists for simpler, interpretable models of dynamic biological processes.

Purpose of the Study:

• To introduce a novel linear hybrid mathematical model for multiphasic cell growth.
• To demonstrate its equivalence to a Switched affine AutoRegressive model with eXogenous inputs (SARX).
• To provide a simpler alternative to existing complex nonlinear models.

Main Methods:

• A linear hybrid model was developed to represent continuous growth phases with discrete parameter switches.
• The model was shown to be equivalent to a Switched affine AutoRegressive model with eXogenous inputs (SARX).
• The model was applied to infer growth phases and parameters from microbial datasets.

Main Results:

• The proposed hybrid model effectively captures multiphasic growth dynamics.
• Inferred growth segments and parameters closely matched expert determinations for Lactococcus lactis, Streptococcus pneumoniae, and Saccharomyces cerevisiae.
• The model demonstrated robustness across different microorganisms and experimental conditions.

Conclusions:

• The developed hybrid model offers an interpretable and simpler approach to describing multiphasic cell growth.
• This framework is effective for analyzing microbial growth data and potentially other phased biological processes.
• The model's success highlights the utility of hybrid systems in biological modeling.