Bottom-up parameterization of enzyme rate constants: Reconciling inconsistent data.

Area of Science:

• Metabolic Engineering
• Systems Biology
• Computational Biology

Background:

• Kinetic models are crucial for understanding metabolic systems and designing production strains.
• Assembling kinetic models enzyme-by-enzyme ('bottom-up') faces challenges like data gaps, complex mechanisms, and in vitro-in vivo discrepancies.

Purpose of the Study:

• To develop a robust computational workflow for estimating kinetic parameters in detailed mass action enzyme models.
• To create a software package (MASSef) for handling diverse kinetic parameters and reaction mechanisms.
• To reconcile inconsistent kinetic data and enable the construction of scalable, in vivo-consistent pathway models.

Main Methods:

• Developed a computational workflow for robust kinetic parameter estimation, accounting for parameter uncertainty.
• Implemented the workflow in the MASSef software package, supporting macroscopic and microscopic kinetic parameters.
• Utilized three enzyme case studies to demonstrate data reconciliation and model assembly.

Main Results:

• MASSef successfully identified and reconciled inconsistent kinetic data from in vitro experiments and between in vitro and in vivo conditions.
• Parameterized enzyme modules were effectively used to assemble pathway-scale kinetic models that align with in vivo behavior.
• The workflow demonstrated robustness in estimating kinetic parameters for detailed mass action enzyme models.

Conclusions:

• The MASSef workflow provides a robust method for parameterizing enzyme kinetic models at scale.
• This approach effectively utilizes historical literature data and machine learning estimates to overcome limitations in kinetic modeling.
• Enables the creation of more accurate and predictive metabolic models for research and strain design.