Diverse classes of constraints enable broader applicability of a linear programming-based dynamic metabolic modeling...

Area of Science:

• Systems Biology
• Metabolic Engineering
• Computational Biology

Background:

• Existing metabolic modeling tools face limitations in scalability and simplifying assumptions.
• Linear Kinetics-Dynamic Flux Balance Analysis (LK-DFBA) was developed to capture metabolite dynamics and regulation within a scalable framework.
• The linearity of LK-DFBA constraints, while computationally efficient, may not fully represent complex biochemical system behaviors.

Purpose of the Study:

• To develop and evaluate new classes of constraints for LK-DFBA to more accurately model metabolite-reaction interactions.
• To assess the performance of these novel constraints across diverse synthetic and biological systems.
• To compare LK-DFBA predictions with experimental data for the first time.

Main Methods:

• Development of three novel constraint types for LK-DFBA.
• Testing these constraints on synthetic and biological metabolic models.
• Experimental validation of LK-DFBA predictions under wild-type and genetic perturbation conditions.

Main Results:

• No single constraint type demonstrated universal optimality across all tested systems.
• Optimal constraint choice varied even for systems with identical topology but different parameters.
• The optimal constraint for wild-type systems often remained optimal after genetic perturbations, irrespective of model topology or parameters.

Conclusions:

• The addition of multiple constraint approaches enhances LK-DFBA's capability to model a broader spectrum of metabolic systems.
• A single wild-type experimental dataset is sufficient to identify the most predictive constraint for a given system.
• This advancement improves the accuracy and applicability of dynamic metabolic modeling.