Fig. 5: Analysis of signalling networks and transcriptional host responses to leaf microbiota strains.

a, Regulatory networks in A. thaliana were inferred with ISMARA⁴⁷ using RNA sequencing data of responses to leaf microbiota strains (reported in Maier et al.²¹) (left) and to diverse MAMPs/DAMPs (reported in Bjornson et al.²⁹) (right), indicating that transcriptional reprogramming upon bacterial colonization and elicitor perception is predicted to be driven by overlapping WRKY TFs. The regulatory motifs (x axes, cut-off at 200) are ordered by significance (Z value, y axes). Regulatory motifs targeted by WRKY TFs are highlighted in yellow, and the top five motifs of the left panel are shown in bold. b, PCA of differentially expressed genes (P ≤ 0.01, two-sided Wald test with Benjamini–Hochberg correction, filtered for |log₂FC| ≥ 1 relative to axenic control in at least one condition) in response to Arthrobacter Leaf137 and Pseudomonas Leaf15 within genotypes (indicated at the top of each plot). Arthrobacter Leaf137 and Pseudomonas Leaf15 cause distinct transcriptional responses depending on the genetic background of the host. c, Heat map of log₂FCs of GNSR genes (bold) and selected PTI marker genes (non-bold) (vertical axis) in Col-0 wild-type plants or mutants colonized by the indicated strains (horizontal axis) relative to the corresponding genotype under axenic conditions. Arthrobacter Leaf137 and Pseudomonas Leaf15 consistently induced the GNSR in plants deficient in key immunity signalling components. Genes were filtered for P ≤ 0.01 (two-sided Wald test with Benjamini–Hochberg correction). Colour indicates the extent of fold change as indicated by the colour key. Conditions and genes are clustered by Ward’s method, as indicated by the trees. Main condition clusters are designated as I and II. In b,c, the data are from n = 5 independent experiments, each comprising 18 pooled plants per condition, sampled at nine days post inoculation.

Source data