APOLLO: A genome-scale metabolic reconstruction resource of 247,092 diverse human microbes spanning multiple continents, age groups, and body sites

Abstract

SummaryComputational modelling of microbiome metabolism has proved instrumental to catalyse our understanding of diet-host-microbiome-disease interactions through the interrogation of mechanistic, strain- and molecule-resolved metabolic models. We present APOLLO, a resource of 247,092 human microbial genome-scale metabolic reconstructions spanning 19 phyla and accounting for microbial genomes from 34 countries, all age groups, and five body sites. We explored the metabolic potential of the reconstructed strains and developed a machine learning classifier able to predict with high accuracy the taxonomic strain assignments. We also built 14,451 sample-specific microbial community models, which could be stratified by body site, age, and disease states. Finally, we predicted faecal metabolites enriched or depleted in gut microbiomes of people with Crohn’s disease, Parkinson disease, and undernourished children. APOLLO is compatible with the human whole-body models, and thus, provide unprecedented opportunities for systems-level modelling of personalised host-microbiome co-metabolism. APOLLO will be freely available underhttps://www.vmh.life/.

Predicting microbial genome-scale metabolic networks directly from 16S rRNA gene sequences

“…For example, one could extend the training data substantially by building GEMs for larger collections of genomes, e.g., the full GTDB taxonomy [20], although adding genomes of lower quality could reduce the accuracy of ANN predictions. If one is specifically interested in high-quality predictions for the human gut microbiome, one could apply our approach to very large existing GEM collections of human gut microbes [8, 9].…”

“…Reconstructing eukaryotic metabolism still requires substantial manual curation [2], but automated reconstruction of accurate GEMs from genomes is possible for microbes [3, 4]. Fueled by rapidly growing genomic and biochemical databases [5, 6], automated metabolic reconstruction methods have produced large GEM collections, primarily of human gut microbes [7–9] but recently of unicellular fungi as well [10]. However, despite progress in metagenomics [11] and cultivation [12], many observed microbes lack genomic data and thus remain inaccessible to GEMs.…”

Personalised metabolic whole-body models for newborns and infants predict growth and biomarkers of inherited metabolic diseases

“…49 Finally, differences in gut microbial composition may also contribute to variations in the blood concentration of arginine and its precursors. Accordingly, we found that gut microbiomes from healthy infants were enriched in reactions associated with "Arginine and proline metabolism" compared with healthy adults (average 13.5% ± 2.9% vs 10.7% ± 1.2%) 54 (Table S13). Taken together, this application demonstrates how the infant-WBMs can be personalised based on newborn screening metabolomic data to investigate emerging metabolic properties.…”

Whole-body metabolic modelling reveals microbiome and genomic interactions on reduced urine formate levels in Alzheimer’s disease

“…To meet the scale of human gut microbiomes, a resource of genome-scale reconstructions of 773 human gut microbes, called AGORA 18 , was generated and subsequently expanded in size and scope (AGORA2), now accounting for 7,302 human microbes 19 . Additionally, a large-scale reconstruction resource based on metagenomically assembled genomes has been developed accounting for nearly 250,000 genome-scale metabolic reconstructions 20 . Using these microbial reconstruction resources, comprehensive, personalised microbiome models can be constructed using microbial relative abundances derived from metagenomics data through tools, such as the Microbiome modelling toolbox 21 , 22 .…”