Fig. 4: Geographical variables affecting the abundances of coexistent ARG-MRGs in agricultural soil microbiomes. | Nature Communications

Fig. 4: Geographical variables affecting the abundances of coexistent ARG-MRGs in agricultural soil microbiomes.

From: Organic fertilization co-selects genetically linked antibiotic and metal(loid) resistance genes in global soil microbiome

Fig. 4

a Partial least square-structural equation modeling (PLS-SEM) illustrating the direct and indirect effects on the coexistence of ARG-MRG (AMCC abundance). Blue, red, and gray indicate significantly positive, significantly negative, and nonsignificant effects, respectively. Path coefficients and coefficients of determination (R2) were calculated after bootstrapping (n = 100), and all path coefficients shown were statistically significant (p < 0.05). Different colors represented different types of variables. b Standardized direct (blue), indirect (pink), and total effects (green) from the PLS-SEM. c Mean predictor importance of different factors (same as PLS-SEM) on the abundance of AMCCs based on random forest modeling (Permutation test, n = 1000, *p < 0.05; **p < 0.01). Source data were provided as a Source Data file. ARG antibiotic resistance gene, MRG metal(loid) resistance gene, GOF goodness of fit, DG.rho Dillon-Goldstein’s rho, AVE average variance extracted, CD community diversity, EM elemental metabolism, MAT mean annual temperature, AP annual precipitation, GDP gross domestic product, HDI human development index.

Back to article page