Fig. 3: Weighted gene co-expression network analysis determines metastatic drivers.

a Biplot principal component analysis (PCA) shows groups along the PC1 axis that correspond to primary (blue triangles) and metastatic (green circles) PBC plus controls (red crosses). b WGCNA cluster dendrogram on all samples groups genes into distinct driver modules. Co-expression distance (TO, topology overlap) between genes (y-axis) and to genes (x-axis). Gene modules are colour coded. We selected the Green module where E2F1 is a hub gene for further analysis because of its relationship to TP53 and that MMP9 was a component of the module. c Heat map-based hierarchical cluster analysis of the Green module show clear and distinct expression patterns between tissue types. Z-score refers to high (red) and low (blue) gene expression using normalised values when compared to the mean of total sequencing reads. d GO analysis using REVIGO [52] scatterplot visualisation shows the cluster representatives in a two-dimensional space derived by applying multidimensional scaling to a matrix of the GO terms’ semantic similarities. Bubble colour indicates p value. Bubble size indicates the frequency of the GO term in the underlying gene ontology annotation GO term database. e Molecular functions significantly affected.