Reduction of a set of elementary modes using yield analysis

Area of Science:

• Systems Biology
• Metabolic Engineering
• Computational Biology

Background:

• Metabolic network analysis often involves understanding elementary modes (EMs).
• Extracting essential EMs for accurate metabolic behavior description remains a challenge.
• Existing methods for flux decomposition and EM identification have limitations.

Purpose of the Study:

• Introduce "yield analysis" (YA) as a novel method for extracting essential elementary modes (EMs).
• Address practical issues in metabolic network analysis: refining generating sets and decomposing measured fluxes.
• Develop a consistent approach for selecting minimal active modes.

Main Methods:

• Defined yield analysis (YA) as the analysis of metabolic pathways in yield space.
• Recalculated the minimal generating set of EMs, excluding trivial modes.
• Revisited flux decomposition among EMs, comparing a new method with existing approaches (Schwartz and Kanehisa, Provost et al.).
• Applied a two-staged reduction procedure to a metabolic network of recombinant yeasts.

Main Results:

• Reduced 369 EMs to 35 modes (91% reduction) through a priori reduction in a yeast fermentation case study.
• Identified minimal active sets of 3 and 4 modes for glucose and xylose fermentation, respectively.
• Demonstrated that the refined minimal generating set describes all possible yield states, while the active set represents experimental data.

Conclusions:

• Yield analysis provides an effective method for simplifying metabolic network models.
• The proposed approach enhances the accuracy and efficiency of elementary mode analysis.
• This technique offers a practical solution for understanding complex metabolic behaviors and experimental data.