View Source

The significant CpG loci detected in the previous step actually form a methylation signature that differentiates , which we shall now visualise by a heat map.

Select the (CpG of interest) spreadsheet in the spreadsheet pane on the left
Select Cluster Based on Significant Genes from the Visualization panel of the Illumina BeadArray Methylation workflow
Select Hierarchical Clustering for Specify Method (Figure 1)

Genomics Suite Documentation > Obtain methylation signatures > figure29.png

Select OK
Verify that (CpG of interest) is selected in the drop-down menu
Select Standardize for Expression normalization (Figure 2)

Genomics Suite Documentation > Obtain methylation signatures > figure30.png

Select OK

The heat map will be displayed on the Hierarchical Clustering tab (Figure 3).

Genomics Suite Documentation > Obtain methylation signatures > figure31.png

The experimental groups are rows, while the CpG loci from the (CpG of interest) spreadsheet are columns. Methylation levels are compared between Naive shPOU5F1, Naive shNANG, and Primed shCTRL vs. Naive shCTRL. CpG loci with higher methylation are colored red, CpG loci with lower methylation are colored green.

As expected, the Naive shPOU5F1, Naive shNANG, and Primed shCTRL experimental conditions cluster together; however, we can also see regions of similarity between Primed shCTRL and Naive shCTRL. This suggests that treatment with shRNA against POU5F1 or NANOG brings the methylation profile of naive hPSCs closer to that of primed hPSCs, but also causes additional changes not observed in either primed or naive hPSCs.