Fig. 7: WMS genome differential abundance in the children, based on comparisons of high-vs-low Social Disadvantage (SD) scores (N = 35 and N = 41 biologically independent samples, respectively) and high-vs-low Psychosocial Stressors (PS) scores (N = 36 and N = 32 biologically independent samples, respectively). | Nature Communications

Fig. 7: WMS genome differential abundance in the children, based on comparisons of high-vs-low Social Disadvantage (SD) scores (N = 35 and N = 41 biologically independent samples, respectively) and high-vs-low Psychosocial Stressors (PS) scores (N = 36 and N = 32 biologically independent samples, respectively).

From: Social and psychological adversity are associated with distinct mother and infant gut microbiome variations

Fig. 7

a Taxonomy and relative abundance per sample for genomes with the highest predictive value for SD in the random forest (RF) model based on the GMs from the children (ranked by mean decrease in accuracy of the RF model; MDA). Also shown are -Log of the Kruskal-Wallis test FDR-corrected P-values from LEfSe (when effect size ≥ 2), FDR-corrected P-values from ANCOM-BC2 and the relative abundance of the taxa when present. On violin plots, thick black lines indicate median values, dashed black lines indicate data range quartiles, shaded area heights represent the proportion of data points for the abundance values, and the width of shaded areas spans from the minimum to the maximum value. Violin plots represent data from all samples in the comparison. b Taxonomy, relative abundance and differential abundance statistics for the genomes with the highest predictive value for PS in the RF model based on the GMs from the children. Data visualization and statistics are the same as for a.

Back to article page