Visualisations¶
The visualisations are shown by default when you open the somatic structural and copy number variant list page. To hide, adjust the width or make the visualisations full screen, see Variant List Somatic SV and CNV.
The visualisations allow you to:
- See the locations of somatic structural variants
- See an overview of all the affected genes
- See the large copy number changes and the data supporting those
- Navigate to specific regions to investigate the supporting data further
Info
Data has been hidden from these plots to protect patient identity.
Key
| # | Section | Description |
|---|---|---|
| 1 | Toolbar | The toolbar allows you to navigate the visualisations. You can zoom in or out at different intervals, or move left or right at different intervals when zoomed in, you can go back to the whole genome overview by clicking the refresh icon. |
| 2 | Cytobands | Shows the location of the chromosomes and related cytogenetic bands. |
| 3 | Plot arrows | The order of the plots can be moved up and down using the arrows. A greyed out arrow indicates the plot cannot be moved in the specified direction. All plots are moveable, including the gene track. |
| 4 | Tumour structural variants | Illustrates the somatic structural changes in domain 1, 2 and 3. This plot will show all the structural variants within the applied filter set on the variant list and will update what’s shown once new filters have been applied. |
| 5 | Gene Track | Illustrates the position of the affected genes. The blue arrow shows the direction. By default, we prioritise genes that are indicated for testing in the patient’s clinical indication. As you zoom to more specific regions and more horizontal space is available, more genes will appear in following priority: 1. Genes indicated for testing in patients clinical indication 2. Genes indicated for testing in patient’s clinical indication group 3. Genes present in Cancer Gene Census or National Genomic Test Directory 4. All other genes Hovering over the gene name will reveal a functional description from NCBI Refseq. As you zoom, the gene layout will become visible, exons are shown as blocks, intron regions as lines. Untranslated regions are half sized blocks shown in grey. You can hover the layout to see the exon count. |
| 6 | Tumour Coverage | For cases processed by Canvas, coverage for the tumour sample is calculated and the y-axis shows the actual depth of coverage. For cases processed by Dragen, the values on the y-axis aren't representative of true coverage for Dragen called CNVs and are now hidden. The data displayed is the highest resolution available from the Canvas/Dragen outputs and can be zoomed into a resolution of ~ 300bp windows. The expected values on the coverage and b-allele plots are plotted in black. To understand how they are calculated, please refer to the Cancer Genome Analysis Guide for details. |
| 7 | Tumour B-allele frequency | This plot evaluates how well the model predicts haplotypes by showing the observed B-allele frequency as well as the predicted based on the CN calls. The data is averaged across 100kb windows. To understand how they are calculated, please refer to the Cancer Genome Analysis Guide for details. |
| 8 | Tumour Absolute Allele count | On Absolute Allele Counts plot, the total allele counts are indicated with solid thick lines and minor allele counts are indicated with solid thin lines. The dashed line indicates overall ploidy estimate. For likely sub-clonal variants, the copy number (CN) is shown in black, the CN fraction (CNF) and minor CNF (MCNF) are represented by the coloured lines. The difference between the CN and CNF is highlighted in yellow, with a larger difference indicating a smaller sub-clonal fraction. The maximum copy number shown is capped, see the Cancer Genome Analysis Guide for details. |
| 9 | Germline Coverage | Coverage for the germline sample is calculated using Dragen, data is averaged across 1kb (1000bp) windows. The expected values on the coverage and b-allele plots are plotted in black. To understand how they are calculated, please refer to the Cancer Genome Analysis Guide for details. |
| 10 | Germline B-allele frequency | This plot shows the observed B-allele frequency data averaged across 500kb windows. To understand how this is calculated, please refer to the Cancer Genome Analysis Guide for details. |
| 11 | Germline Absolute Allele count | On Absolute Allele Counts plot, the total allele counts are indicated with solid thick lines and minor allele counts are indicated with solid thin lines. The dashed line indicates overall ploidy estimate. The maximum copy number shown is capped, see the Cancer Genome Analysis Guide for details. |
Tumour structural variant plots¶
This plot illustrates the prediction of structural variants called by the Manta SV caller. Deletions are shown in pink, inversions are shown in black, translations are shown in grey, insertions are shown in purple, duplications are shown in blue. The height of the structural variant is defined by the quality score, calcuated by Manta. Translocations called by Pelops with an estimated breakpoint (e.g., DUX4 fusions) are shown with a dotted line.
Info
The SV plot may appear crowded for tumours with many overlapping structural variants. We are currently investigating whether we can make the SVs in these plots easier to distinguish from each other. In the meantime, users can apply filters to the SV plot to remove any SVs that are not of interest.
Tumour copy number variant plots¶
These plots illustrate the predictions of large copy number changes as calculated by Canvas/Dragen as well as the raw data supporting those predictions. CNV losses are indicated in pink, CNV gains are indicated in blue, high CNV gains of five copies and above are indicated in green, copy-neutral LOH regions are indicated in orange. Copy number neutral areas are indicated in black and regions where no data is available are indicated in grey.
Likely sub-clonal variants¶
From the Dragen release onwards (NGIS release Orion), likely sub-clonal variants are displayed in yellow, which represents the difference between the copy number (CN) and CN fraction (CNF). For further information on how Dragen identifies likely sub-clonal variants, see the Cancer Genome Analysis Guide.
Warning
We have identified a bug in our plot generation, where genomic sex is being used rather than reported sex. If a cases genomic sex differs from the reported sex, the "expected coverage" line across chrX would display incorrectly.
An example case where the reported sex differed from the genomic sex is shown below:
As you can see, the black "expected coverage" line does not line up with the observed coverage values across chrX. This is because a higher level of coverage across chrX is expected in this case where the reported sex is female, but the genomic sex is male.
Plots available on the Interpretation platform show the correct "expected coverage" line due to utilising the genomic sex rather than reported sex.
It's expected that only a very small subset of cases will have a discrepant genomic and reported sex and are therefore impacted by this bug. A bug fix will be implemented in release 6.2.0, therefore cases ingested after the 6.2.0 release will be unaffected, regardless of their genomic and reported sex.
Info
Data has been hidden from these plots to protect patient identity.
Germline copy number variant plots¶
These plots illustrate the predictions of large copy number changes as calculated by Dragen as well as the raw data supporting those predictions. CNV losses are indicated in pink, CNV gains are indicated in blue, high CNV gains of five copies and above are indicated in green, recurrent or non-pass CNVs are indicated in purple. Copy number neutral areas are indicated in black and regions where no data is available are indicated in grey.
Highlighting variants¶
You can see the associated call of each structural variant on the plot by clicking it. This will show the associated variant in the table on the left. Clicking a structural variant on the list on the left, will highlight it on the plot by making the line bolder. Only structural variants can be highlighted.
Navigating the plots¶
You can navigate to specific regions in the plot in various ways:
- Click the coordinates of the breakpoints or regions in the table on the left
- Click the chromosome number in the chromosome track
- Click the cytogenetic band in the cytoband track
- Use the zoom, move or reset controls in the toolbar on top of the plots
- Scroll up or down with your mouse when your cursor is in one of the plots to zoom in
- Clicking and dragging the cursor left or right to move left or right
Moving Plots¶
The order of the plots can be moved up and down as desired using the corresponding plot arrows. A greyed out arrow indicates the plot cannot be moved in the direction indicated. All plots are moveable, including the gene track.
- To move a plot, simply click the plot arrow that corresponds to the desired direction.
- Continue to click the plot arrows until the plot is in the desired place.
- Refreshing the page will reset the plots to their original positions.
Potential errors¶
When the data for the plots can’t be loaded, you will see an error message as seen below. Try reloading the page, if that doesn’t solve the problem, get in touch with support via the Genomics England Service Desk.
