KineFlux Predicts Genome-Scale Flux Distributions Accurately
Researchers Fayaz Soleymani, Zahra Razaghi-Moghadam, and Zoran Nikoloski publish March 16, 2026 in PLoS Computational Biology a method called KineFlux that combines machine learning with enzyme-constrained metabolic models to predict steady-state genome-scale flux distributions using only quantitative proteomic data. Using fluxomic and proteomic datasets from Escherichia coli and Saccharomyces cerevisiae, KineFlux aligns with classical flux estimates, shows transferability across conditions, and obviates the need for metabolomics or explicit enzyme kinetics.
Key Points
- 1Demonstrates KineFlux accurately predicts steady-state flux distributions from proteomics without metabolomics
- 2Combines machine learning with enzyme-constrained models to capture metabolite concentration effects on reaction fluxes
- 3Enables practitioners to predict genome-scale fluxes across conditions using only proteomic inputs and trained models
Scoring Rationale
Peer-reviewed method with practical code enables usable flux prediction, but novelty is incremental over existing enzyme-constrained approaches.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
