Supplementary data to this article: Supplementary Document 1. Detailed description of the databases construction. Supplementary Figure 1. Processing and construction workflow of reference databases. The final databases used in the comparison are highlighted in red. Supplementary Figure 2. Determination of the optimal number of clusters for average abundance clustering analysis. The Elbow method was used to set the optimal number of clusters for the k-means algorithm at genus (A) and species (B) level. Supplementary Figure 3. Determination of the optimal number of clusters for z-scored average abundance patterns analysis. The Elbow method was used to set the optimal number of clusters for the k-means algorithm at genus (A) and species (B) level. Supplementary Figure 4. Bracken results comparison between the enriched filtered-MAG and original NT databases. From top to bottom, the performance of the databases was examined in terms of classification capacity for all samples as a whole, by host lifestyles and geographic origin at genus (A) and species (B) level. In the box plots, the black line within the box marks the median and the red triangle the mean, outliers are presented as red dots. Supplementary Figure 5. Relative abundances of top gut microbiota at the genus level. Only the 25 most abundant taxa in all samples were represented. Taxa are sorted by average relative abundances and their NCBI’s taxID is indicated in parentheses. In the box plots, the black line within the box marks the median and the red triangle the mean. Supplementary Figure 6. Universal core species. (A) Intersections between cores of interest. (B) Taxonomic relationships between universal core species. The number of core taxa assigned to a particular level is indicated inside the square brackets. (C) Prevalence-Abundance Heatmap. Taxa are sorted by average relative abundances and their NCBI’s taxID is indicated in parentheses. Supplementary Figure 7. Taxonomic relationships between different intersect core species in a prevalence gradient. Additional intersect cores were computed defining soft and medium prevalence species cores, which were compared to the corresponding universal species core, working in a prevalence range between 0.5 and 1. Supplementary Figure 8. Relative abundances of top gut microbiota at the species level. Only the 25 most abundant taxa in all samples were represented. Taxa are sorted by average relative abundances and their NCBI’s taxID is indicated in parentheses. In the box plots, the black line within the box marks the median and the red triangle the mean. Supplementary Figure 9. Relative abundances of individuals for the universal core taxa. Abundances are shown for the genus (A) and species (B) core taxa. Groups and taxa were clustered using the Complete method and Euclidean distances. In the case of the groups, this was done for every single group combination, and each taxon NCBI’s taxID is indicated in parentheses. Supplementary Figure 10. Abundance patterns of individuals for the universal core taxa. Z-scored abundances are shown for the genus (A) and species (B) core taxa. Groups and taxa were clustered using the Complete method and Euclidean distances. In the case of the groups, this was done for every single group combination, and each taxon NCBI’s taxID is indicated in parentheses. Supplementary Table 1. Summary information for the microbiome project genomes used in the construction of the enriched databases. Supplementary Table 2. Summary information for the metagenome studies used in this work. Supplementary Table 3. Summary of statistics for samples quality control. The different results generated by FastQC were aggregated with MultiQC for each study. Supplementary Table 4. Metadata for each human gut metagenome sample. Supplementary Table 5. Core genera membership for the different cores of interest. Presence-absence data is presented for the different cores of interest at 0.9, 0.7, and 0.5 prevalence thresholds. Prevalence values for the intersect core taxa at 1e−4 relative abundance threshold are also shown. NCBI’s taxIDs are indicated in parentheses. Supplementary Table 6. Core species membership for the different cores of interest. Presence-absence data is presented for the different cores of interest at 0.9, 0.7, and 0.5 prevalence thresholds. Prevalence values for the intersect core taxa at 1e−4 relative abundance threshold are also shown. NCBI’s taxIDs are indicated in parentheses. Supplementary Table 7. Universal core genera information and known characteristics.

Peer reviewed