Table 1.
Demographic and clinical characteristics of patients and controls included in this study.
Figure 1.
The flowchart schematizes the experimental steps of the statistical analysis of microarray data.
Table 2.
The top 25 most significant genes from Limma analysis.
Figure 2.
Validation of gene expression by real-time PCR.
The expression of JAK2, IL-6, KITLG and CCL2 in the dissected aorta was validated by real-time PCR analyses. Expression levels in control aortas were set to 1. Data represent the mean ± SEM.
Figure 3.
Top Ranked ExpMods and JAK2 ExpMod.
Derivation of the JAK2 hotspot associated with aortic dissection: (A) The spin-glass module detection algorithm is applied to an integrated mRNA expression interactome to identify differential expression hotspots. The edges in the network are weighted according to the average of the absolute regularized t-statistics of the genes making up the edge (Methods). The edge weights are colour coded as indicated. (B) Detection of 8 non-redundant highly exclusive differential expression hotspots/modules (ExpMods) associated with aortic dissection, ranked according to their modularity values and all passing an adjusted P-value threshold of 0.05. The gene symbols of the seeds are shown. (C) Zoomed-in version of the JAK2 ExpMod. Both edge color and width encode the edge weights, as indicated. The Limma t-statistics of each gene (node) are shown as indicated. Observe how JAK2 defines the hub of this module, and forms dense subnetworks with other genes that are strongly differentially expressed e.g. IL-6/IL-6R, CCL2, KITLG and EPOR.
Figure 4.
JAK2 protein-protein interaction network by STRING analysis.
The network nodes are proteins. Seven differently colored lines represent 7 types of evidence used in predicting the associations. A red line indicates the presence of fusion evidence; a green line - neighborhood evidence; a blue line - coocurrence evidence; a purple line - experimental evidence; a yellow line - textmining evidence; a light blue line - database evidence; a black line - coexpression evidence.
Figure 5.
Validation of the JAK2 Expression Module.
Left panel: scatterplot of regularised limma t-statistics of JAK2 Expression Module members in our discovery data set (x-axis, t(D)) against an independent validation set (Weis-Mueller et al 2006) (y-axis, t(V)). Positive statistics indicate overexpression in AAD cases relative to normal controls. Blue dashed lines indicate the lines of nominal significance P = 0.05. Fisher’s exact test P-value, reflecting agreement between two sets is given. Right panel: heatmap of the differential expression statistics in discovery and validation set for the 22 JAK2 module members showing consistency between the two sets. Green: underexpressed in AAD. Red: overexpressed.