Computing optimal factories in metabolic networks with negative regulation

Area of Science:

• Systems Biology
• Metabolic Engineering
• Computational Biology

Background:

• Metabolic networks are crucial for understanding cellular processes.
• Identifying metabolic pathways (factories) is fundamental in systems biology.
• Existing methods for finding factories do not optimize for pathway length or incorporate negative regulation.

Purpose of the Study:

• To introduce and solve the problem of finding optimal factories with minimal reaction steps.
• To incorporate both first- and second-order negative regulation into factory identification.
• To develop a computationally efficient method for optimal factory discovery.

Main Methods:

• Modeling metabolic networks and factories using directed hypergraphs.
• Proving the NP-completeness of the optimal factory problem.
• Solving the problem using mixed-integer linear programming.
• Employing an iterative approach to handle second-order negative regulation.

Main Results:

• The proposed optimization-based approach is highly efficient, finding optimal factories in seconds.
• The method successfully handles metabolic networks of significant size (tens of thousands of reactions/metabolites).
• Comprehensive experiments validated the approach across standard reaction databases.

Conclusions:

• The new method provides an efficient and effective solution for identifying optimal metabolic factories.
• Incorporating negative regulation and optimizing for reaction count advances metabolic network analysis.
• The developed tool, Odinn, offers a practical resource for researchers in systems biology and metabolic engineering.