Fig. 4

A selection of the default outputs from the Multiple Differential Expression workflow, using the demonstration dataset. Using three sample groups—Lamina Propria (LP) (red), Mesenteric Lymph (ML) (green) and Mesenteric Lymph Node (MLN) (blue), and two differential comparisons ML versus LP and MLN versus ML. Significance for differential genes was adjusted p < 0.01 and absolute log2 fold change > 1. a Bar chart of the number of up and downregulated genes for each comparison. b Gene expression heatmap of the 3,220 significant genes from either comparison. Colour denotes row scaled (Z-score) expression values, with darkest blue as lowest expression and darkest red as highest. The Y-axis has been hierarchically clustered using Spearman Correlations, UPMG agglomeration and mean reordering. c Fold versus Fold scatterplot comparing MP versus LP (Y-axis) to MLN versus ML (X-axis) at each gene. Each dot is one gene, with black dots being non-significant, blue being significant in MLN versus ML only, green in ML versus LP only and red in both. The Spearman correlation coefficient (SCC) is − 0.25. d Gene expression heatmaps for four of the identified differential expression signatures (1–4). Further plot details are as (c). e Differential expression signature meta-gene violin plot with jitter values for signature number 1. The mean expression (Z-score) across all genes in the signature is given on the Y-axis. Black dots denote individual samples. The red dot and whisker denote mean and standard deviation respectively. f Bar chart of the five most enriched gene-sets (GO Biological Processes) for the genes in signature number 1. The X-axis shows the − log10 p value and the data labels the number of significant genes in each gene-set. All gene-sets are significant at p < 0.05. g As (e) however for signature number 4. h As (f) however for signature number 4