Cancer is by nature very heterogeneous: The genomic alterations of one cancer patient can be very different compared with another patient. This is the case for different cancer types AND more importantly for this tutorial cancer samples can be very heterogenous although they are from the same tumor site. For this reason it is important to obtain detailed clinical information about a sample.
Having a the expression data of a cohort of patients, it is not very convenient to compare the expression differences on a sample-gene basis. In a first step the genes may be grouped into modules like f.ex. pathways, so they can be analysed as one unit. This is exactly what we propose with the Sample Level Enrichment Analysis (or short SLEA). With this analysis we can asses the transcriptional status of each pathway (or other modules) for each sample. In a second step we can relate the enrichment status with the clinical annotation, like cancer subtypes.
To see how it is done, watch the video tutorial and/or read the instructions below
Make sure you have the data provided in step 1 of this use case. If you don’t have it yet, download this zip-file , which includes:
Then follow the steps below.
A wizard will open.
A new tab will open with the analysis summary.
Select to open the heatmap in the Results section.
To discriminate between over- and under-expressed pathways we select to display the Z-Score value (in the Cell Properties tab)
Now if some specific pathway has gotten your interest, you can click the button on the top right of the heatmap to show all the genes of that pathway with the original expression data associated.