IGM: Integrated Gene-expression Modeling for multi-condition flux-preserving genome-scale metabolic models.

Area of Science:

• Systems Biology
• Metabolic Engineering
• Computational Biology

Background:

• Genome-scale metabolic models (GEMs) are crucial for studying cellular metabolism.
• Conventional methods like Flux Balance Analysis (FBA) often produce ambiguous results due to inconsistent integration of condition-specific data.
• Existing gene expression integration methods are limited to single conditions, use arbitrary thresholds, or reduce flux interpretability.

Purpose of the Study:

• To develop a novel framework, Integrated Gene-expression Modeling (IGM), for robustly integrating multi-condition gene expression data into GEMs.
• To preserve flux units and enhance biological relevance in metabolic modeling.
• To improve the accuracy and consistency of flux predictions across different experimental conditions.

Main Methods:

• IGM utilizes a mixed-integer linear programming (MILP) framework.
• It integrates relative gene expression via gene-protein-reaction (GPR) rules without binarization.
• Flux Variability Analysis (FVA) defines feasible flux ranges, and the model minimizes discrepancies between fluxes and gene expression.

Main Results:

• IGM significantly improves correlation with experimentally measured fluxes in *E. coli* models.
• The framework reduces flux solution ambiguity and enhances predictive consistency across multiple conditions.
• IGM variants, particularly with L1 norm regularization, demonstrate high accuracy and preserve transcriptomic patterns.

Conclusions:

• IGM provides a robust framework for condition-specific and consistent integration of transcriptomic data into metabolic modeling.
• It enables biologically grounded predictions of metabolic adaptation and dynamic changes.
• IGM enhances the interpretability and predictive power of GEMs for multi-condition studies.