Fig. 5: A NHR gene regulatory network.
From: Worm Perturb-Seq: massively parallel whole-animal RNAi and RNA-seq
a Expression levels of nhr genes in adult animals. The TPM was quantified by the reference profile used in Supplementary Fig. 1c. b Tissue expression of nhr genes based on an adult stage single-cell RNA-seq data90. The maximal TPM of hypodermis and intestine expression is shown in the plot. c Summary of NHR WPS experiments. d Distribution of the number of DEGs in NHR perturbations (i.e., k-out) and the number of NHRs regulating the same gene (i.e., k-in). The red dashed line indicates the responsiveness threshold ( ≥ 5 DEGs). e Functional enrichment analysis of NHR RNAi conditions. Only the 54 NHRs (55 perturbations because nhr-25 was perturbed at both L1 and L2 stages) with more than 10 DEGs were analyzed to ensure power. Conditions without significant (Padj > 0.05) enrichment are not shown in the plot. The relative enrichment is defined as the −log10(Padj) normalized by its maximum in a given RNAi condition. f Proportion of testable RNAi (i.e. >10 DEGs) that showed significant enrichment (Padj < 0.1) in WormCat Level 1. The P values in (e, f) were derived from one-sided hypergeometric test with multiple testing adjusted by Benjamini-Hochberg method. Visualization of hub genes (i.e., regulated by >5 NHRs) of the stress response (g) and metabolic (h) categories. i Comparison of the activator score for metabolic (x-axis) and stress response (y-axis) DEGs. NHR perturbations with ≥5 DEGs in the corresponding categories were analyzed. j Overall comparison of activator scores for different DEG categories. Each dot represents an activator score for one NHR, calculated using DEGs from the category shown on the X-axis (All: n = 83; Stress: n = 26; Metabolic: n = 32). Boxes show the IQR (25th–75th percentiles) with median line; whiskers extend to 1.5×IQR. Two-tailed Wilcoxon tests were performed to calculate the P values. Source data are provided as a Source Data file.
