Fig. 3: Rank I ARGs copy number and richness in the genomes of 8388 E. coli isolates.

a Mean of E. coli Rank I ARGs (per genome), mobile Rank I ARGs (MRank I ARGs), and the occurrence frequency of MRank I ARGs (genomes existing MRank I ARGs/ total genomes) over time in soil (1404 genomes). Each point represented the mean of all samples for that year. Gray shading denotes the 95% confidence intervals. Pearson’s correlation (two-sided) with loess regression for data smoothing was used to test the statistical relationship between variables (mean copy number for Rank I ARGs: p < 0.00018, mean copy number for MRank I ARGs: p < 6.7e-07, occurrence frequency of MRank I ARGs: p < 1.9e-07). b Richness of E. coli Rank I ARGs (per genome) and MRank I ARGs over time in soil (1404 genomes). The richness was calculated in gene level. Each point represented the average of all samples for that year. Gray shading denotes the 95% confidence intervals. Pearson correlation (loess regression) was conducted to obtained the p value and r. Richness refers to the number of Rank I and MRank I ARGs gene type. Pearson’s correlation (two-sided) with loess regression for data smoothing was used to test the statistical relationship between variables (diversity for Rank I ARGs: p < 0.00015, diversity for MRank I ARGs: p < 1.3e-07). c Mean copy numbers of Rank I ARGs and MRank I ARGs in different habitats across various periods (using 8,388 genomes collected from various habitats). 999 rounds of random sampling based on the minimum habitat and period counts were conducted to eliminate the impact of sample size differences across different years on the results. 38 genomes per habitat per period were included in the analysis. Each point represented the average of all samples for that year. “Before 2007” represented genomes collected prior to 2007, and the remaining period groupings were consistent with that in metagenomic analysis (i.e., Period A: 2008–2010, Period B: 2011–2013, Period C: 2014–2016, Period D: 2017–2019, Period E: After 2020). In the boxplots of panels, hinges indicate the 25th, 50th, and 75th percentiles, whiskers indicate 1.5× interquartile ranges, dots indicate the average value from each random sampling. Significant comparisons (two-sided t test) between different temperature and different groups were calculated and detailed p value and T stat were provided in the Supplementary Data 9. d Richness of Rank I ARGs and MRank I ARGs in different habitats across various periods (using 8388 genomes collected from various habitats). 999 rounds of random sampling were performed and 38 genomes per habitat per period were included. Significant comparisons (two-sided t test) between different temperature and different groups were calculated and detailed p value and T stat were provided in the Supplementary Data 10. In the boxplots of panels, hinges indicate the 25th, 50th, and 75th percentiles, whiskers indicate 1.5× interquartile ranges, dots indicate the average value from each random sampling. Richness refers to the number of Rank I and MRank I ARGs gene type. e The Rank I ARGs copy number, MRank I ARGs copy number, and the occurrence frequency of MRank I ARGs (genomes existing MRank I ARGs/ total genomes) in soil collected from different period. The Rank I ARGs copy number, MRank I ARGs copy number, and the occurrence frequency of MRank I ARGs were normalized to a range of 0 to 1 in each period. R² is calculated from the linear model, and r is calculated from the Pearson correlation (liner regression). * indicated p < 0.05, ** indicated p < 0.01. The correlation was statistically tested using both Pearson’s correlation test (two-sided) and a linear model (two-sided). Rank I ARGs: catII: p < 0.048, erm(42): p < 0.048, mef(B): p < 0.0074, AAC(3)-IId: p < 0.028, mph(A): p < 0.0086, cmlA5: p < 0.0062, linG: p < 0.0090, aadA5: p < 0.0065, AAC(6’)-Ib7: p < 0.0078, tet(M): p < 0.0074, sul3: p < 0.010, floR: p < 0.0078, aadA2: p < 0.018. MRank I ARGs: AAC(3)-IId: p < 0.013, mph(A): p < 0.0052, cmlA5: p < 0.0096, linG: p < 0.0087, aadA5: p < 0.0053, tet(M): p < 0.047, sul3: p < 0.015, AAC(3)-IIe: p < 0.034, floR: p < 0.0095, aadA2: p < 0.0049, tet(B): p < 0.021. occurrence frequency of MRank I ARGs: aadA8: p < 0.041, lnu(G): p < 0.041, AAC(3)-IId: p < 0.045, erm(B): p < 0.036, tet(M): p < 0.036.