EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

EINN: An enzyme-informed neural network guided by an enzyme-constrained genome-scale metabolic model

Ray Steven, Kai Kimata, Denis Chegodaev, Lilies Handayani, Kenji Satou

Synthetic and Systems Biotechnology · 2027-03 · 17 · 66-84

DOI: 10.1016/j.synbio.2026.07.009 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Synthetic and Systems Biotechnology, 2027-03. Verfasst von Ray Steven und 4 weiteren Autor:innen. Beteiligte Einrichtung: Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa, 9201192, Japan. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Predicting cellular metabolism from molecular data is a central challenge in systems biology, with direct implications for engineering microbial cell factories. Genome-scale metabolic models (GEMs) provide a mechanistic foundation for such predictions, but their accuracy is limited by an inability to account for the finite catalytic capacity of the proteome. Enzyme-constrained GEMs (ecGEMs) address this by incorporating enzyme turnover numbers and proteome allocation constraints. Hybrid neural-mechanistic models like Artificial Metabolic Network (AMN) and Metabolic-Informed Neural Network (MINN) have sought to combine the flexibility of machine learning with the structure of GEMs, yet they rely on conventional, unconstrained metabolic networks, allowing flux predictions to violate enzyme capacity constraints and pushing models toward biologically unrealistic solution spaces. Here, we introduce the Enzyme-Informed Neural Network (EINN), a conceptual framework which relies on ecGEMs to constrain the neural network. We implement two variants of EINN, ecAMN and ecMINN, using Escherichia coli ( E. coli ) ecGEMs built with the GECKO 3.0 toolbox, and systematically evaluate their performance. Compared to their unconstrained counterparts, ecAMN significantly improves training stability and predictive accuracy by eliminating convergence to poor local minima, while ecMINN reduces overfitting and achieves lower error rates through mechanistic integration of proteomic data as flux bounds. Together, these results show that constraining the neural network's solution space with enzymatic capacity limits is a more effective hybrid modeling foundation than relying on conventional GEMs alone.

Schlagwörter

Stichwörter der Autor:innen

Genome-scale metabolic modelHybrid modelMachine learningMulti-omicsNeural networks

Autor:innen und Institutionen

  • Graduate School of Natural Science and Technology, Kanazawa University, Kanazawa, 9201192, Japan.
  • Faculty of Transdisciplinary Sciences for Innovation, Institute of Transdisciplinary Science for Innovation, Kanazawa University, Kanazawa, 9201192, Japan.

Zitieren

APA

Steven, R., Kimata, K., Chegodaev, D., Handayani, L., & Satou, K. (2027). EINN: An enzyme-informed neural network guided by an enzyme-constrained genome-scale metabolic model. Synthetic and Systems Biotechnology, 17, 66-84. https://doi.org/10.1016/j.synbio.2026.07.009

Vancouver

Steven R, Kimata K, Chegodaev D, Handayani L, Satou K. EINN: An enzyme-informed neural network guided by an enzyme-constrained genome-scale metabolic model. Synthetic and Systems Biotechnology. 2027;17:66-84. doi:10.1016/j.synbio.2026.07.009

BibTeX
@article{Steven2027,
  author = {Steven, Ray and Kimata, Kai and Chegodaev, Denis and Handayani, Lilies and Satou, Kenji},
  title = {EINN: An enzyme-informed neural network guided by an enzyme-constrained genome-scale metabolic model},
  journal = {Synthetic and Systems Biotechnology},
  year = {2027},
  volume = {17},
  pages = {66--84},
  doi = {10.1016/j.synbio.2026.07.009}
}
Kostenlos starten →