Figure 2

Screenshot of Epiviz. It shows the main elements available within the tool: 1) the main toolbar, featuring all UI controls; 2) a scatter plot, showing two computed measurements: the average and difference between colon gene expression for normal and cancer tissues; the code for this plot is customized in the UI to show a line at y=0, that separates genes with positive and negative differences; 3) a heatmap, showing values from the Gene Expression Barcode [4] comparing the normal and cancer expressions for different tissues. Using its clustering feature, we notice that tumors tend to group separately from normal tissues; in addition, the clustering result seems to be determined by a small number of genes, namely MMP1, MMP3 and MMP10; 4) a stacked plot, showing two columns for normal and cancer gene expression; it uses the color by transformation, to highlight genes with various expression differences. This plot offers several insights: first, that overall expression tends to be higher for cancer than normal tissues; second, it allows us to immediately spot the differentially expressed genes, by brushing over the blocks colored in deep red, corresponding to them; 5) a custom track defined in a plugin hosted on GitHub Gist, showing blocks aligned to the genome, with height corresponding to the expression of the genes; 6) a stacked track, showing a computed measurement, corresponding to the difference between normal and cancer methylation; this track offers an insight over the hypo/hyper-methylated blocks; 7) a lines track, showing DNA methylation for normal and cancer colon tissues; the track uses the group by transformation to aggregate three normal samples and three tumor samples, and displays error bars to show the variation of methylation for each group at each data point; in addition, it uses basis interpolation to smoothly connect the available data points; 8) a genes track, showing human genome genes fetched from the UCSC database[10] using a data provider plugin stored externally on GitHub Gist; 9) a tooltip showing details on demand for the gene MMP1. The highlighted items correspond to the brushing feature, triggered while hovering over the MMP1 gene in the genes track. The feature links all visualizations together by genomic location. http://epiviz.cbcb.umd.edu/?gist[]=160e8b84795603961b9f&gist[]=5a88f39caa801e58b8ae&ws=GJU2bfURaUd