Variant Detail Page¶
Users can access more information about a particular variant using the Variant Detail page (see clip below). To access this page, click anywhere (except on a hyperlink) within that variant's row in the Omics Explorer table. This is available in the Omics Summary view or in the Filtered View view.
At the top of the Variant Detail page is a header (as seen below) that includes info based on the selected variant (such as Amino Acid Change, Nucleotide Change, & Variant Class)
The details page lays out the data of the Omics Explorer table in a grouped format. The details shown will depend on the variant selected.
- CKB - Cancer Knowledge Base
- Gene - Gene name
- AA Change - Amino Acid Change
- NC Change - Nucleotide Change
- Position: Refers to the numeric chromosome position
- Variant Class
- Coding Effect - These are the types of mutations that affect the RNA and protein coding
- DBSNP RS ID: This ID number will link out to the dbSNP (database of Single Nucleotide Polymophism) website.
- EXON number
- Gene ID
- Transcript ID
- Variant Allele Freq
- VCF Filter
Population Allele Frequency:
- Genomes Freq
- Genomes HOM
- Exomes Freq
- Exomes HOM
- Max Freq
- Min Freq
- ClinVar Significance
- ClinVar Disease
- ClinVar ID
- ClinVar Review
- ClinVar Submission
- ClinVar Nearby: This feature was added by LifeOmic. If there’s a variant for a gene, this will look in ClinVar to see if there is anything two or three amino acids in either direction that may be of significance.
- COSMIC Count
- COSMIC Status
- COSMIC Histology
- COSMIC Site
- COSMIC ID: This is the numeric COSMIC ID of a mutation. This ID number links to the COSMIC website.
- COSMIC Nearby: If COSMIC Count is one and COSMIC Nearby is one, it means the variant has been seen once in COSMIC and there is a variant nearby Population Allele Frequency
In Silico Prediction:
- Damaging %
- Damaging Rank Score
- MUT Taster
Genes can have many different transcripts. While they are all the same gene, they have different splicing/different exons. If you click on the Transcripts tab, you will see an Alternate Transcripts table. Here we denote the canonical transcript with a blue "i" info icon beside the transcript id. All other transcripts are listed beneath it (see image below).
- Transcript ID – Most canonical transcripts have a blue info dot beside it. If there is no dot, it means there is no canonical transcript identified because the way it was annotated did not specify it was part of the same gene.
- NC Damage
- AA Change
- Variant Class
- EXON Number
The CKB tab is where you can view any drug matches that correlate with the tumor variants.
CKB/Drug Association Table Columns:
- CKB - In the Cancer Knowledge Base
- Gene – Gene name
- Feature - Mutation
- Source – Source external database (civic, cosmic_resistmut, docm, pmkb, cgi)
- Drug – Name of drug associated with it
- Response Type – Tells how responsive it is to the drug. (Resistant, Responsive, better outcome, poor outcome)
- Evidence Level – AB, A (clinical practice/standard of care), B (some sort of clinical trial/off-label use data), C (early clinical trials/one case showed benefit), D (preclinical/mouse studies)
- Source Evidence – 1, 2, 3, Clinical evidence, Preclinical evidence, Validated association, Case study, Early trials
- Disease – Disease information (location/type)
- Publication – Links out to pubmed article regarding any gene drug association
- Comment - Any additional information is noted here
The Gene tab gives more information about the gene. This information is brought in from an external source (referenced in the summary).
- Summary - Tells more information about the gene
The bar graph shows details of published cancer studies where the gene is predicted to act as a cancer driver. This data comes from an external database. The bar chart shows the gene mutation frequency or where its a driver gene in various cancer types (as shown below). Hovering over the info button (i) in the graph title will reveal a popup stating "This bar chart and table show details of published cancer studies where the gene is predicted to act as a cancer driver. All studies are listed in the table and those cohorts with at least 30 samples are plotted in the bar chart."
The data found in the table shows details of published cancer studies where the gene is predicted to act as a cancer driver. This data comes from an external database called Intogen. Intogen curates the frequency of mutated genes in different cancer types (see image below). The table cannot be filtered or interacted with. In this external reference data set, they’ve taken different cancer types and looked at the samples with mutations compared to the overall in their study cohort and derived a frequency (% samples). They include information such as if the sample is WGS (Whole Genome) or WXS (Whole Exome), whether they were treated or untreated when they were sequenced (before or after any cancer treatment), and "Reference" refers to where the data came from (PMID = PubMed ID). Users can copy and paste the reference info into a browser for more info on the study where that data came from.
Data in this table includes:
- Cohort Details
- Cohort Cancer Type
- % Samples: Found by taking different cancer types, looking at the samples with mutations compared to the overall in their study cohort, and deriving a frequency.
- Samples with Mutation
- Cohort Samples
- Platform (WGS = Whole Genome, WXS = Whole Exome)
- Treated: Treated or untreated when sequenced or before/after cancer treatment
- Reference: Where the data came from.(PMID = PubMed ID) You can copy and paste the reference info into a browser for more info on that study and where the data originated.
Needle Plot Tab¶
The Needle Plot is a visual representation of where that mutation occurs in the protein. Needle Plots frame how important this variant is by giving a clinician or researcher a quick reference point of how often this variant has been seen in cancer. It lets them know if it’s common and if it’s not, is it near a common variant? Or is it clustering near other mutations?
If you look at the Needle Plot example below, for gene PIK3CA, you see high peaks and clusters of needles, showing something is going on in these spots of the gene.
Hovering the cursor over any needle will tell you which mutation is at that needle point.
A needle plot is only for coding variants in the exons of a gene (somatic variants). On the top left of the plot is external reference data brought in from TCGA (Cancer genome atlas resource). The needle plot shows you left to right, the position in the protein. It shows you where the variant is for the patient with the blue flag.
There is a key to the right corresponding with the flag color:
- black = truncating
- green = missense
- blue = synonymous
- purple = inframe
The other needles on the plot come from TCGA’s tumor sample pool. You can hover over each one and it will show you the mutation found and how many have been found in TCGA. Beneath the line with the needle plot are colored bars with info in them. These are protein domains (a special part of the protein with a known/presumptive function – this data comes from another database, PFFam). You can click and drag from top left to bottom right of a section of the needleplot to zoom in.
Reference Tissue Expression Tab¶
For Expression variants only, you can click on this tab on the Variant Details page to see the Reference Tissue Expression in:
- TCGA Cancer Type Tumor Samples (as seen in image below)
- TCGA Normal Samples
- GTEx Normal Broad Tissues
- GTEx Normal Specific Tissues
- TreeHouse Tumor Samples (Pediatric)
In the example shown in the image above, the purple horizontal line is the Expression for this patient in this sequencing test. The plot allows you to see how this expression relates to other cancer types. In the center of the plot, the chartreuse bar is the sum of All Diseases/Tissues.
If the gene we were looking at is highly expressed such as KIT is in this example, we can return to the Filtered View, click on Outliers to filter the genes that are highly expressed – greater than 2 std. deviations above the mean for TCGA expression. To do this we:
- Scroll down the list of filters and click on Outliers
- Under Compare Against Source select "TCGA Tumor Samples", and select All Diseases/Tissues
- Set #of std deviations to "2", and select "Above the mean"
Click Add Filter
In the example below, we clicked on the Germline Variant tab, and clicked the checkbox for CKB. The MAPK1 gene came back as a Gene Match, but the Variant Detail page shows no info under Drug Association. This means this gene has a known gene drug association, but this particular variant in this patient sample didn’t have an exact match to a drug.
Standalone Blue Dots
Blue dots appearing alone in a field of the Omics Explorer table means the canonical transcript does not meet that column criteria, but it's alternative transcript does meet the criteria. It lets you know there is more information about this gene.
For instance, in the clip below, we filter for missense variants. The results in the Omics Explorer table shows a variant with a blue dot in a blank field under both AA Change and Variant Class. This is because the canonical transcript does not have a missense variant. However, if we hover over the blue dot, it shows the alternate transcript which does have a missense variant. This is why it came through as a result for that missense filter.
Affects Alternate Transcript Example
In the example below, we filtered for missense variants. The Omics Explorer table shows a variant with a blue dot beside "missense" under the Variant Class column and a blue dot beside the "A128V" under the AA Change column.
The canonical transcript is a "Missense", but if you hover over the blue dot beside the word "missense", the words “intron variant” appear, meaning there is another transcript that labels this mutation as an intron variant and not a missense. Likewise, hovering over the blue dot beside the "A128V" AA Change, we’ll see 2 different AA changes from the canonical AA change shown in the table, A65V and A85V.
If we click on the row to view the Variant Detail (see image below) we see 10 transcripts, there is a beside the canonical transcript and is the one displayed in the Omics Explorer table. There are eight total transcripts with a missense mutation. And there are two intron variant mutations (seen when we hovered over the blue dot beside "missense" in the Variant Class column) but since it’s an intron, there is no AA Change.