cBioPortal Tutorial #3:
Patient View
Investigate individual patients or samples in detail
Last update: December 22, 2021
Tutorial Objectives
Option # 1 to get to patient view:
Anywhere you see a patient or sample ID, that ID is a link to patient view for that case.
See next slide for examples.
Click on any of these sample/patient IDs
Option #2 to get to patient view:
Use the study summary page to filter down to cases of interest. Then click the “view the selected patients” button.
See next slide for example.
1. Filter to a subset of patients, if desired
2. Click on this button to view the selected patients
No matter how you get to patient view, you will be taken to the summary tab.
Depending on the study, the other tabs in patient view may or may not be present.
In this tutorial we will look at patient view in two different studies to highlight the different kinds of data that may be available.
Example 1: Brain Lower Grade Glioma (TCGA, PanCancer Atlas)
This is the same query that we used in the single study query tutorial. Hover over a case of interest and then click on the patient ID.
Link to this page
Patient View, Example 1: Summary
Basic details about the patient and sample(s). Hover over the patient ID or sample ID to see more information.
Figure showing where called CNA and mutations are across the genome. Hover over any of these for more details.
Lists of all called mutations, structural variants and CNAs (amplifications and deep deletions only).
Link to this page
Copy, download, add/remove columns or search.
Patient View, Example 1: Pathways
Explore the alterations listed in the Summary tab in the context of frequently altered pathways defined by TCGA. For more detail on this tab, refer to the Pathways Tutorial.
Link to this page
Patient View, Example 1: Clinical Data
All available patient-level clinical information
Below the patient-level information is sample-level information. Patients with multiple samples will have multiple columns in this table.
Link to this page
Patient View, Example 1: Pathology Report
Original pathology report, de-identified.
Note: Pathology Reports are only available for TCGA studies.
Link to this page
Patient View, Example 1: Tissue Image
Zoomable image of the tissue. When available, additional images can be selected from the list on the left.
This tab integrates the Cancer Digital Slide Archive.
Note: Tissue images are only available for TCGA studies.
Link to this page
Example 2: Low-Grade Gliomas (UCSF, Science 2014)
1. Filter the study to a subset of patients, if desired
Link to this page
2. Click on this button to “View selected cases”
Patient View, Example 2: Patient Summary
This study has multiple samples per patient and extensive clinical data to generate this enhanced patient timeline.
Link to this page
Patient View, Example 2: Patient Summary
List of all mutations called. The first column (“Samples”) shows which samples had a particular mutation. The Allele Freq column depicts the mutation frequency in each sample by the height of the bar.
Patient timeline showing surgeries, radiographic progression and treatments. Hover over any feature for additional information. Click the to expand the timeline.
Click to view the next patient
Figure showing distribution of mutations across the genome for each sample.
Link to this page
Patient View, Example 2: Patient Summary
Link to this page
List of all samples for this patient. Hover on a sample ID for more details or click to get to a sample summary page (we’ll do this in a few slides)
Hover over a sample ID to see the plot for just that sample
Hover to see an enlarged version. It shows a histogram with overlaid density estimation of the allele frequency in each tumor sample.
Patient View, Example 2: Genomic Evolution
Link to this page
The Genomic Evolution tab is present for any patient with 2 or more samples. This tab provides visualizations to examine how mutation allele frequencies vary among samples and change over time. The Timeline (on the Summary tab) can also be shown on this tab to put allele frequency changes in context.
Allele frequencies can be displayed as a Line Chart or Heatmap
Click to view the timeline above the allele frequency visualization
Patient View, Example 2: Genomic Evolution - Line Chart
Link to this page
Each dot represents the allele frequency of a mutation in a sample. Lines connect mutations that are detected in multiple samples. Options above the chart enable customization.
Patient View, Example 2: Genomic Evolution - Line Chart
Link to this page
Hover or click on a mutation in the table to see it highlighted in the chart above.
Change x-axis to show all samples equally spaced (below) or samples in real time (aligned with timeline, see first Genomic Evolution slide)
Click on mutations in the table below and then check this box to only see those mutations in the chart.
This chart and the mutation table are linked - hover or click on a mutation in the chart to see it highlighted in the table below.
Click on mutations in the chart above and then check this box to see only those mutations in the table.
Patient View, Example 2: Genomic Evolution - Heatmap
Link to this page
Each box is colored according to the allele frequency of a mutation in a sample. Options above the chart enable customization.
Hover or click on a mutation in the table to see it highlighted in the chart above.
Patient View, Example 2: Clinical Data
All available patient-level clinical information
All available sample-level information
Link to this page
When available, the data used to populate the timeline in the Summary tab is shown here.
Patient View, Example 2: Sample Summary
Clicking on a sample ID on one of the previous pages brings up this sample-specific page.
Link to this page
Ok, now that we’ve seen what data is present in Patient View, we can start asking some fun question!
Let’s look at RAS mutations in Uterine Corpus Endometrial Carcinoma (TCGA, Nature 2013).
Example 3: Run the query
Link to this page
Example 3: OncoPrint
Link to this page
In general, mutations in these genes are mutually exclusive. However, there’s one case with driver mutations in both KRAS and NRAS. Let’s look at that patient in greater detail by clicking on the patient ID (“TCGA-B5-A0JV”).
Example 3: Patient View
3. Could this be related to differences in clonality? Perhaps the PIK3CA mutation is clonal while the NRAS & KRAS mutations are in two distinct subclones. If that theory is correct, we would expect to see other mutations with similar variant allele frequencies. Indeed,we can see that is true by looking at the histogram of variant allele frequency.
2. Note that all three genes are diploid, so the differences are unlikely to arise from copy number alteration.
Link to this page
1. Look at the Allele Freq column for each mutation. NRAS Q61K (19%) and KRAS G12D (21%) have similar variant allele frequencies, but PIK3CA E542K is twice as high (38%).
Summary of Example 3: Using Patient View, we can infer the clonality of mutations and understand how two mutations, which are usually mutually exclusive, can be present in the same tumor sample. In this case, the KRAS and NRAS mutations appear to be present in two distinct subclones of a single tumor.
Questions?
Check out our other tutorials
or email us at: cbioportal@googlegroups.com