Abstract Constructing high-quality genome-scale metabolic models (GEMs) for less-studied species remains challenging. To address this, we developed AlphaGEM, a versatile toolbox leveraging proteome-scale structural alignment and deep-learning-based predictions for efficient genomic mining to generate GEMs ready for applications. Our findings show that the structural alignment or protein-language-model-based prediction (i.e., PLMSearch), could identify more homologous protein relationships than sequence-blast-based alignment, contributing to the accurate profiling of metabolism from target organisms. Additionally, AlphaGEM encompasses an ensemble procedure empowered by multiple deep learning toolboxes to effectively mine the dark metabolic functions encoded by nonhomologous proteins, significantly expanding species-specific metabolic networks. We demonstrate AlphaGEM’s accuracy by building GEMs for eukaryotes (e.g., S. pombe , C. albicans ) and prokaryotes (e.g., K. pneumoniae , B. subtilis ), achieving predictions comparable to manually curated models while outperforming existing tools. AlphaGEM also successfully reconstructs GEMs for M. musculus and C. griseus , showcasing its great potential for uncovering dark metabolism in complex mammals. Lastly, we demonstrate that AlphaGEM could facilitate the automatic GEMs reconstruction for 332 distinct yeast species with high prediction fidelity. In conclusion, AlphaGEM provides unprecedented opportunities for the precise, rapid construction of GEMs across diverse domains, which sets a solid foundation for universal functional analysis of non-model organisms having genome sequences available.
Han et al. (Thu,) studied this question.