Fig. 4: Variation patterns of intestinal microbial diversity and community composition in short-term and long-term DEHP exposure periods.

A Bacterial richness (Sobs) and B bacterial diversity (Faith’s PD) among all treatments (n = 6, mean ± SE). C Non-metric multidimensional scaling (NMDS) ordinations analysis of bacterial community composition across all treatments. Differences in bacterial community composition (NMDS) of different treatments were determined by analysis of permutational multivariate analysis of variance (PERMANOVA; ***p < 0.001). D Multiple regression tree (MRT) analysis of effects of DEHP treatment and treated time on bacterial community composition. The R², error, cross-validation error (CV Error), and standard error (SE) of MRT analysis were listed under the tree. E Percent stacked bar chart based on the relative abundances of bacterial phyla, showing hierarchical clustering along with DEHP treated and control groups on top of the graph. Color blocks on the right show phyla names, the key on the right show significant changes and their direction (upward and downward arrows for increased and reduced abundances, respectively) for each phylum as compared DEHP treatment with control group, in short-term and long-term respectively (Student’s t test, *p < 0.05; **p < 0.01). F Random Forest (RF) mean predictor importance (percentage of increase of mean square error (MSE), increase in MSE%) of bacterial diversity indices (NMDS1, sobs and Faith’S PD) and phyla abundances as drivers for mouse pregnancy rate and levels of HPO axis-related hormones comprised of Follicle-Stimulating Hormone (FSH), GnRH, Estradiol, PGE2 and Progesterone. The accuracy importance measure was computed for each tree and averaged over the forest (5000 trees). Percentage increases in the MSE of variables were used to estimate the importance of these predictors, and higher MSE% values imply more important predictors. Significance levels are as follows: *p < 0.05; **p < 0.01.