multiTFA

multiTFA implements multivariate thermodynamics-based flux analysis (TFA) as a Python package to integrate multivariate treatment of thermodynamic variables with component contribution (group contribution) estimates for tighter thermodynamic constraints in metabolic flux models.

Key Features:

• Implementation: Provided as a Python package for integration with metabolic modeling workflows.
• Multivariate Treatment: Employs a multivariate approach to represent and constrain thermodynamic variables across reactions.
• Gibbs Free Energy Covariance Matrix: Utilizes the covariance matrix of reactions' Gibbs free energy to impose correlated thermodynamic constraints on fluxes.
• Component Contribution Methodology: Leverages component contribution (group contribution) estimates to refine Gibbs free energy values and reduce uncertainty.

Scientific Applications:

• Thermodynamics-constrained MFA: Applied to metabolic flux analysis (MFA) to tighten thermodynamic constraints and improve flux predictions under physiological conditions.
• Model evaluation and refinement: Demonstrated on a core E. coli model, achieving a median reduction of 6.8 kJ/mol in reaction Gibbs free energy ranges and changing directionality of three out of twelve glycolysis reactions from reversible to irreversible.

Methodology:

Integrates multivariate statistical techniques with thermodynamics-based constraints, using Gibbs free energy covariance matrices and component contribution (group contribution) estimates to constrain metabolic fluxes and assess reaction reversibility.