A Flux-to-AI Discovery Framework Reveals Carbon-Regulated Intermittent Non-Canonical Nitrification in Achromobacter xylosoxidans

Background Developing a data- and model-driven understanding of how carbon availability regulates non-canonical heterotrophic nitrification is critical to characterizing nitrogen cycling. Metabolic versatility across hosts and environments worldwide was identified using all available genomic data for Achromobacter xylosoxidans (Ax) by generating genome-scale metabolic and machine-learning models. Methods Here, we developed a model-driven pangenome analysis of all available Ax genomic data, adding 22% more new genomes and revealing an uncharacterized non-canonical nitrification pathway. Genome-scale metabolic modeling, experimental validation, and an expanded Flux-to-AI approach were integrated to investigate carbon-regulated nitrification and host-associated metabolic adaptations. Results Genome-scale metabolic model simulations provided quantitative insights into the shift to an alternative energy-generating strategy based on specific carbon utilization. Experimental validation showed that nitrification was intermittent and associated with the late exponential phase. The expanded flux-to-AI approach revealed nutrient-specific phenotypes associated with the metabolism of the dipeptide Gly-Met, acetoacetate, and L-cysteine. These host-dependent metabolic trade-offs appear to extend phylogenetically across bacterial and fungal taxa. Conclusions Beyond improving our understanding of host–microbe interactions, this systems-level view of microbial metabolism has broad applications in bioprocess optimization and environmental biotechnology. Overall, integrating modeling predictions with artificial intelligence methodologies has led to a flux-to-AI framework capable of predicting metabolic phenotypes across all high-quality A. xylosoxidans genomes available to date. Our findings demonstrate that human-associated isolates adapt to amino acid-rich environments and identify amino acid uptake and metabolism as key determinants of persistence and survival.