Additional file 1 of Co-localization of antibiotic resistance genes is widespread in the infant gut microbiome and associates with an immature gut microbial composition

Additional file 1: Figure S1. Emergence and co-selection of multidrug resistance in chromosomes or plasmids. A) Progressive acquisition of multiple resistances from mutations or gene transfer as a result of exposure to antimicrobial agents. Co-resistance represents the co-localization of multiple resistance genes in the same gene fragment. Cross-resistance represents a gene that confers resistance to multiple antimicrobials. B) Co-selection between multiple resistance genes upon exposure to any of the corresponding antimicrobials. This figure is adapted from the work of Cantón and Ruiz-Garbajosa [8]. Figure S2. Summary of antibiotic usage for 662 infants during the first year. Figure S3. Distribution of gene richness and abundance among bacterial species carrying (A) BRGs, (B) MRGs, (C) virulence genes, and (D) MGEs in the infant gut.Figure S4. Total abundance (sum of GCPM) of (A) virulence genes in log scale, (B) BRGs in log scale, (C) MRGs, and (D) MGEs in log scale in the infant gut. The labels on the x-axis describe (A) virulence factors, (B) biocides, (C) metals, and (D) categories of MGEs. The red dot in the scatter plot represents the median value of the total abundance.Figure S5. Co-localization bubble chart representing the drug classes related to 42 ABRGs. The size of the bubble is proportional to the abundance of the ABRGs. The drug classes in red represent those used in this cohort. Figure S6. The pecent of varience explained by different phyla in the co-localization of ARGs and BRGs/MRGs. Figure S7. Distribution and Co-localization of ARGs, BRGs/VGs, and MGEs on plasmids or Chromosome in the infant gut. A) The respective abundance (percentage) of contigs, resistance genes, VGs, and mobile genetic elements in chromosomes and plasmids. B) Venn diagram depicting the number of contigs with resistance genes and MGEs in chromosomes and plasmids. Figure S8. The representative co-localization contigs carrying the different types of integrons on plasmids. Based on the ARGs on the 22 representative contigs, 16 representative contigs carrying class 1 integrons were chosen and listed in the figure (each representative contig represents a group consisting of 1 to 16 contigs). The complete integron in each contig is highlighted in gray. For ease of viewing, we only marked the approximate location of the attC recombination site, the constitutive promoter Pc for the gene cassettes, the promoter for the integrase gene PintI, and the integration site attI. To target potential mobile ARGs, we only focused on the co-localization contigs carrying MGEs on the plasmids. The location and length of genes are proportional to the actual conditions. Figure S9. The mean relative abundance of E. coli in the gut of infants at one week, one month, one year, four years, five years, and six years of age based on 16S sequencing data. Figure S10. The percent of the 45 most abundant bacterial species in the infant gut. Figure S11. E. coli drives the bimodal distribution of BRGs, MRGs, and virulence genes in the infant gut. A) Density plot of gene richness among infants and Spearman correlation analyses between the richness of ARGs and other genes. Y-axis (not shown in the density plot) represents the number of infants. The linear regression curve and confidence interval are plotted as an illustration; inference was performed with the Spearman correlation coefficient R and the corresponding p-value (p < 0.05 as the significance cutoff). B) Average silhouette width associated with PAM clustering when different numbers of clusters were used, and the total abundance and richness of genes in the two clusters. A high average silhouette value indicates strong clustering. k = 2 was the optimal number of clusters. C) Importance of the top five bacterial species to the grouping of PAM clusters, as determined with a Random Forest-based approach using the mean decrease in accuracy and the relative abundance of E. coli in the two clusters. Species are ordered top-to-bottom as most-to-least important.