Tutorial: Jump start on the UCSC Cancer Genomics Browser

Step-by-step instructions to start a Cancer Browser session

1. Begin by choosing a dataset you would like to view, and select "Heatmap" from the dropdown menu. Here we have selected the Neve et al. cell line CGH dataset:


The heatmap will load and appear above the list of datasets. The heatmap on the left represents the copy number values of each probe across the genome (labeled below) with each pixel row representing a sample. The feature information of each sample is represented by a heatmap on the right, with each feature labeled below it. Clicking on a feature will sort the samples (i.e. rows) accordingly. Secondary sorting using shift-clicking breaks ties in the primary sort.

2. To access feature configuration options, click on the blue bar to the right of the feature information. This will show a new panel below the dataset:

The currently displayed features are listed and can be removed by clicking the "X" symbol to the right of the text. They can be reorganized by dragging and dropping the features in order. Additional available features can be added through the dropdown menu above the list. Clicking "Update Image" will refresh the feature heatmap above with the current features in the list in order. For example, if we remove all but "Her2", the feature heatmap will show only that feature:

3. To define subgroups, click the name of a feature in the list, which will display subgrouping controls for that feature (the red and green selection boxes in the image above). By selecting the subset of features in each box and clicking the "Add Subgrouping" button, we can define two subgroups based on Her2 status:

You can see the current subgrouping defined in the list on the right, and a new red/green bar appears next to the feature information allowing you to identify which samples belong to which subgroup. The option to run statistics is displayed at the bottom.

4. By selecting Studen t-test from the statistics dropdown and Bonferroni from the next dropdown, we are able to run genome-level statistics to determine if there are any regions of particular interest (with multiple hypothesis correction).

5. We can see a small significant peak in chromosome 17. By clicking on the heatmap at that position, we are able to zoom in to investigate the region further.

Below the heatmaps, we can see the current coordinates of the zoomed image, as well as a button to jump directly to the UCSC Genome Browser:

If we click into the UCSC Genome Browser, we can see that the Her2 amplicon is represented in that region, which explains the Her2 significance between the subgroups of samples we selected.

6. Shift-clicking allows us to zoom back out to the full genome level, and we can switch between heatmap and summary view for the dataset:

The summary view allows us to view the distribution of amplifications and deletions across the genome for a dataset. Since we have subgroups defined, we will see a summary image which is split top and bottom for the two subgroups, as well as the statistic track below:

We can remove the subgrouping and see a single summary view for the dataset:

These views allow us to quickly display large-scale genomic amplifications or deletions, as well as examine possible amplifications or deletions for specific subgroups of samples.