AI-Driven Metabolic Modeling and Engineering in Microorganisms

Microbial metabolism is governed by complex reaction cascades, non-linear enzyme kinetics, and multi-layered regulatory circuits. While traditional constraint-based and kinetic models have provided essential foundations, navigating cellular regulation and vast combinatorial design spaces for strain engineering remains a key bottleneck. Rapid developments in artificial intelligence (AI), machine learning (ML), and generative architectures now offer powerful, data-driven frameworks to predict, optimize, and rationally engineer microbial metabolic behavior.This Research Topic focuses specifically on the development and application of machine learning, deep learning, and hybrid data-driven modeling to decode metabolic networks and accelerate the Design-Build-Test-Learn (DBTL) cycle. We welcome contributions that use computational intelligence to predict enzyme function and novel biocatalytic pathways, reconstruct and refine genome-scale metabolic networks, predict intracellular fluxes, and optimize strain performance for bioproduction. Studies integrating multi-omics datasets (transcriptomics, proteomics, metabolomics) with AI-driven models to reveal metabolic control and physiological responses are particularly encouraged.By centering on AI and machine-learning methodologies as the core predictive engine, this collection emphasizes pathway-level metabolic engineering, biocatalysis, and synthetic biology. It is distinctly separated from classical thermodynamic modeling, kinetic frameworks, and macroscopic population or evolutionary dynamics, focusing instead on data-driven strain design and molecular metabolic behavior.We welcome Original Research, Reviews, Mini Reviews, Perspectives, and Methods/Technology and Code articles addressing, but not limited to:-Deep learning and generative AI for enzyme discovery, catalytic turnover rate (k_cat) estimation, and substrate specificity prediction.-Machine learning-guided metabolic network reconstruction, automated gap-filling, and flux prediction.-Hybrid modeling frameworks integrating mechanistic models with ML to predict metabolic phenotypes and trade-offs.-Data-driven strain design and pathway optimization for high-yield bioproduction and bioprocess efficiency.-Machine learning architectures for multi-omics data integration to elucidate metabolic regulation and intracellular fluxes.-Active learning and Bayesian optimization workflows for microbial cell factory engineering. Topic editors: Hena Dhar, Yichao Han, Xinglong Wang Source: https://www.frontiersin.org/research-topics/85550/ai-driven-metabolic-modeling-and-engineering-in-microorganisms