MetaFlux: A Pure Python Genome-Scale Metabolic Network Analysis Engine

Abstract

MetaFlux is a lightweight, dependency-free genome-scale metabolic network analysis engine implemented entirely in Python using only NumPy and SciPy. It provides Flux Balance Analysis (FBA), Flux Variability Analysis (FVA), single-gene knockout screens, pairwise synthetic lethality detection, and 13C Metabolic Flux Analysis (13C-MFA). Unlike existing tools, MetaFlux has zero compiled dependencies.

Introduction

Genome-scale metabolic models (GEMs) are fundamental to systems biology. However, popular tools like COBRApy require external LP/MILP solvers (GLPK, Gurobi, CPLEX), and the TriTensor-based approaches require CUDA and custom kernels. MetaFlux fills the gap: a pure Python implementation that runs anywhere.

Methods

Flux Balance Analysis (FBA)

FBA solves the linear program:

$$\max_\mathbf{v} \ \mathbf{c}^T \mathbf{v} \ s.t. \ \mathbf{S}\mathbf{v} = \mathbf{0}, \ \mathbf{lb} \leq \mathbf{v} \leq \mathbf{ub}$$

using scipy.optimize.linprog with the HiGHS backend.

Flux Variability Analysis (FVA)

FVA computes the min and max flux for each reaction while maintaining the optimal objective:

$$f_j^{\min} = \min_{\mathbf{v}} v_j \quad f_j^{\max} = \max_{\mathbf{v}} v_j \ s.t. \ \mathbf{c}^T\mathbf{v} = f^*, \ \mathbf{S}\mathbf{v}=\mathbf{0}, \ \mathbf{lb}\leq\mathbf{v}\leq\mathbf{ub}$$

Gene Knockout Analysis

Gene knockout is propagated through Boolean GPR (Gene-Protein-Reaction) rules. Each gene's active/inactive state is evaluated against the GPR expression to determine if a reaction is knocked out.

E. coli Core Model

The built-in E. coli core model (Orth et al. 2010, Molecular Systems Biology) contains 72 metabolites, 95 reactions, and 91 genes. Validated against COBRApy textbook model: growth = 0.8739 h⁻¹ (aerobic glucose minimal medium).

API

from metaflux import build_ecoli_core_model, run_fba, run_fva

# Load model
model = build_ecoli_core_model()

# Run FBA
result = run_fba(model)
print(f"WT growth: {result['objective']:.4f} h⁻¹")
# WT growth: 0.8739 h⁻¹

# Run FVA
fva_df = run_fva(model)
print(f"Essential reactions: {fva_df['essential'].sum()}")
# Essential reactions: 48

# Gene knockout
ko_model = model.knockout_gene("b0720")  # citrate synthase
ko_result = run_fba(ko_model)
print(f"CS knockout growth: {ko_result['objective']:.4f}")
# CS knockout growth: 0.0000 (essential)

Installation

pip install numpy scipy pandas plotly
git clone https://github.com/junior1p/MetaFlux.git
cd MetaFlux && pip install -e .

Benchmark

Conclusion

MetaFlux provides a production-quality pure Python implementation of core metabolic analysis methods. It is ideal for environments where compiled solvers are unavailable, for education, and for integration into lightweight pipelines.

GitHub: https://github.com/junior1p/MetaFlux License: Apache 2.0