 |
Page History
| Scrollbar | ||
|---|---|---|
|
| Page info | ||||
|---|---|---|---|---|
|
This chapter describes search results that caIntegrator returns after queries.
Topics in this chapter include the following:
| Table of Contents | ||
|---|---|---|
|
Query Results Overview
After you launch a search of a caIntegrator study, the system automatically opens the Query Results tab showing the results of your search. If you have not configured the column and sort display parameters before launching the search, by default the tab shows only the subject identifiers and a column that allows you to select each row of the data subset.
To display and/or sort additional data, you must return to the Columns Results Type Tab and/or Sorting tabs Tab to set display parameters, then re-run the search. The new search results will display the additional information, with the columns and data sorted as you specified. See https://wiki.nci.nih.gov/x/B4DnAg#3-SearchingacaIntegratorStudy-ResultsTypeTab.
caIntegrator paginates search results into pages of configurable size (default 20) with standard paginated navigation controls. To sort columns by ascending or descending parameters for on any displayed field, click on the underlined column header.
The query results that can display depend upon the criteria you established for the search. Follow the links below for more information about the category of data you searched.
You can download search results as a CSV file. The file contains the You can download search results as a CSV file. The file contains the annotations, columns and data sort configurations you specified in the search query. See on page 63 Exporting Data.
See also .
Browsing Query Results
The query results that can display depend upon the criteria you established for the search. Follow the links below for more information about the category of data you searched.
See #Subject Annotation and Imaging Data, #Gene Expression Data, and #Expanding Imaging Data Results.
Subject Annotation and Imaging Data
| Wiki Markup |
|---|
If you run the search before configuring column and sort display parameters, only the \[subject\] ID that meet the criteria and a column allowing you to select each row appear on the table, shown in the following figure.
!imaging subj ID only75.png|vspace=4, alt="Query Results page"! |
You can add details for one or more subjects by configuring them on the Results Type tab. Annotations listed there are the column headers in the CSV file(s) that were uploaded to the study. For information about using the Results Type tab, see on page 60.
Gene Expression Data
If after defining gene expression criteria on the Criteria tab, you select the Gene Expression result type on the Results Type tab, genomic data search results display in a gene expression data matrix. Because the data was downloaded from caArray, the data permissions granted there still apply. In other words, if you have been given access to the data in caArray, you can see it in caIntegrator.
You can select on the Results Type tab a preferred orientation for displaying the results: genes in rows and subjects in columns, or genes in columns and subjects in rows.
For Gene criteria, the cells display the median gene expression value for each gene. By each gene symbol, caIntegrator displays an icon (
) which you can click to open the Cancer Genome Anatomy Project (CGAP) showing data for the gene, shown in the following figure.
If you have selected Gene Expression on the Results Type tab, then the column headers are a clickable label which sorts the entire table on that column. If you selected Reporter ID on the Results Type tab, the Reporter ID is clickable (and the gene is not clickable).
For fold-change criteria, the cells display the normalized signal-based value for a given reporter for a given sample. In the results matrix, caIntegrator highlights matrix values for fold change results that meet fold change criteria. Red represents upregulated values and blue indicates downregulated values. The following two figures display gene name search results with gene reporter type display in the first and reporter ID reporter type display in the second.
- Genomic data does not display in tandem with subject annotation and imaging data; it only displays when you select the Gene Expression result type on the Results Type tab. Genomic data is however, filtered by subject annotation and imaging query criteria configured on the Criteria tab.
- Click the Export Options CSV link to download the CSV file whose data displays on the Search Results tab. When you do so, the CSV file opens automatically in MS Excel or similar applications for working with spreadsheets, showing the columns and sorting as you defined them in caIntegrator on the appropriate tabs.
You can save genes identified in the search results as a gene list. For more information see .
Copy Number Data*
If after defining copy number criteria on the Criteria tab and running a copy number query, (see on page 55), you should select the Copy Number result type on the , and rerun the query. Copy number data search results display in a data matrix containing samples vs. genomic regions.
Subject Annotation and Imaging Data Results, Gene Expression Data Results, and Expanding Imaging Data Results.
Subject Annotation and Imaging Data Results
If you run the search before configuring column and sort display parameters, only the [subject] ID that meet the criteria and a column allowing you to select each row appear on the table, as shown in the following figure.
You can add details for one or more subjects by configuring them on the Results Type tab. Annotations listed there are the column headers in the CSV file(s) that were uploaded to the study. For information about using the Results Type tab, see Results Type Tab.
Gene Expression Data Results
If after defining gene expression criteria on the Criteria tab, you select the Gene Expression result type on the Results Type tab, genomic data search results display in a gene expression data matrix. Because the data was downloaded from caArray, the data permissions granted there still apply. In other words, if you have been given access to the data in caArray, you can see it in caIntegrator.
You can select on the Results Type tab a preferred orientation for displaying the results: genes in rows and subjects in columns, or genes in columns and subjects in rows.
For Gene criteria, caIntegrator finds and summarizes at the gene level all reporters that match criteria for the gene you defined on the Criteria tab. When displaying results at the gene level, the mean is displayed if there are two probes, the median if there is an odd number of probes greater than two and the mean of the two median probes if there is an even number of probes greater than two. Next to each gene symbol, caIntegrator displays an icon () which you can click to open the Cancer Genome Anatomy Project (CGAP) showing data for the gene. Gene symbols and icons are identified in the following figure.
If you have selected Gene Expression on the Results Type tab, then the column headers are a clickable label which sorts the entire table on that column. If you selected Reporter ID on the Results Type tab, the Reporter ID is clickable (and the gene is not clickable).
For fold-change criteria, the cells display the normalized signal-based value for a given reporter for a given sample. In the results matrix, caIntegrator highlights matrix values for fold change results that meet fold change criteria. Red represents upregulated values and blue indicates downregulated values. The following two figures display gene name search results with gene reporter type display in the first and reporter ID reporter type display in the second. Note the left hand column in each example.
Gene reporter type display
Reporter ID reporter type display
- Genomic data does not display in tandem with subject annotation and imaging data; it only displays when you select the Gene Expression result type on the Results Type tab. Genomic data is however, filtered by subject annotation and imaging query criteria configured on the Criteria tab.
- Click the Export Options CSV link to download the CSV file whose data displays on the Search Results tab. When you do so, the CSV file opens automatically in MS Excel or similar applications for working with spreadsheets, showing the columns and sorting as you defined them in caIntegrator on the appropriate tabs.
You can save genes identified in the search results as a gene list. For more information, see Creating a Gene or Subject List.
Copy Number Data Results
After defining copy number criteria on the Criteria tab and running a copy number query, you should select the Copy Number result type on the Results Type Tab, and rerun the query. Copy number data search results display in a data matrix containing samples vs. genomic regions.
- Gene symbols display parallel to chromosome regions on the matrix.
- Sample ID column headings display the Subject ID/Sample ID (for example, E09262/E09262) because each calculation is based on a comparison of a tumor and matched blood sample from the same subject.
The values in the Sample ID
- Gene symbols display parallel to chromosome regions on the matrix.
- Sample ID column headings display the Subject ID/Sample ID (for example, E09262/E09262) because each calculation is based on a comparison of a tumor and matched blood sample from the same subject.
- The values in the Sample ID
columns are mean segment values as calculated by the DNAcopy algorithm, shown in the following figure. These are expressed as log2 (test/reference, as in tumor/normal). For more information about the algorithm, see Bioconductor
.
.Multiexcerpt include nopanel true MultiExcerptName ExitDisclaimer PageWithExcerpt wikicontent:Exit Disclaimer to Include DNAcopy ouput values can be
DNAcopy ouput values can be negative. If the test and the reference genomic samples both have two copies of a chromosomal region, the ratio of test/reference is '1', and the log2(1) = 0. That is, if there is no change in the chromosomal structure, then the value is 0. If there are more copies in the test sample (amplification of the chromosomal segment), the ratio of test to reference is greater than 1, and the log2(test/reference) is greater than 0. For example, if the test sample has 6, the ratio or test/reference is 6/2 = 3; log2(3) = 1.58. In a deletion, the test is less than the reference, for example 1. The DNAcopy output value would be log2(1/2) = log2(0.5) = -1.0. Values below -0.6 are often considered a deletion.
Creating a Gene or Subject List
From any page in caIntegrator that shows such a group, you can save a list of genes or subjects so you can use it for searches or analyses. This functionality can also be used where a gene or subject list was created outside of caIntegrator, for example, a list of subjects with validated mutation such as from TCGA projects, or a list of subjects with high EGFR expression or any lists of subjects with genomic or clinical characteristics determined with other tools.
...
- Click the Create New List link in the left sidebar. This opens the Manage List page, shown in the following figure.
- Enter a name for the gene or subject list.
- Enter a description (optional).
- Select Make Visible to Others if you want the list to be visible to anyone who views the study. This selection places the list in the Global List folder in the left sidebar under Saved Lists. In any box where you can select lists, the term 'Global' will identify any list so identified when the list is created.
- Select the List Type, Subject or Gene.
- If you select Subject, enter the Subject IDs in the text box that appears. Proceed with step 7.
- If you select Gene, proceed with step 6.
- For Gene Symbol, enter one or more gene symbols in the text box or click the icons to locate genes in the following databases. If you enter more than one gene in the text box, separate the entries by commas.
caIntegrator provides three methods whereby you can obtain gene symbols for creating a gene list:. For more information, see #Choosing Choosing Genes. - If you so choose, you can upload a gene or subject ID list. For the Upload File field, click the Browse button to navigate to a .csv file made up of gene symbols. caIntegrator converts the comma-separated content to a gene list.
Wiki Markup
See #Editing a Gene or Subject List for information about the edit feature.
- figure.
See Editing a Gene or Subject List for information about the edit feature.
| Info | ||
|---|---|---|
| ||
When you perform a GISTIC analysis, caIntegrator automatically saves the retrieved genes in the Saved Copy Number analysis in the left sidebar. For a query or plot analysis, they also appear in the Gene Picker dialog box described in #Choosing Choosing Genes. |
...
...
Editing a Gene or Subject List
To view a gene list or subject list in caIntegrator, under Study Data in the left sidebar, click Saved Lists > Global Lists, or My Lists. Select the list/analysis you want to open. The system displays gene or subject lists that have been saved for the open study.
You can initiate the following functions on this page:
- Click on any of the list names or the list icon (
) to rerun the query from which the gene or subject list was first created. If the list is a gene list, in the query results, you can click on the gene information icon () to open the Cancer Genome Anatomy Project (CGAP) showing metadata for the gene. - Click the edit icon (
) to open an Edit Gene/Subject List dialog box. On this page you can review the list of gene symbols or subject IDs included in the list shown in the following figure.
...
- In
...
- the
...
- Edit
...
- [List
...
- Type
...
- ]
...
- dialog
...
- box,
...
- you
...
- can
...
- perform
...
- the
...
- following
...
- tasks:
...
...
- To
...
- rename
...
- the
...
- list
...
- in
...
- the
...
- [List
...
- Type
...
- ]
...
- List
...
- Name
...
- text
...
- box,
...
- enter
...
- the
...
- new
...
- list
...
- name.
- You can change the visibility of the list in the appropriate check box.
- To delete the list, click the Delete button.
- Click Save to save your changes or Cancel to leave the page without making changes.
Once a list is created, you cannot edit the list contents.
| Include Page | ||||
|---|---|---|---|---|
|
| Include Page | ||||
|---|---|---|---|---|
|
Expanding Imaging Data Results
In reviewing imaging search results, it is important to understand the hierarchy of submissions in NBIA. For more information, see Relationship of Subject to Study to Series to Images.
If you run a search before configuring column and sort display parameters, only the Subject Identifiers for the patients/images that meet the criteria and a column containing one check box per row display by default. An example displays in the following figure.
If your annotation choice on the Columns Results Type page identifies annotations such as tumor size or tumor location, the search results display image series subsets that have those annotations, or any annotations you check on the Results Type page. The check boxes work in conjunction with buttons at the bottom of the results page, shown in the following figure. By expanding display parameters, you can view complete details for image search results.
You can add more details for images by configuring image annotations on the Results Type tab. Annotations listed there are the column headers in the image series CSV file(s) that were uploaded to the study. Examples of image details include the following:
...
You can set display parameters for the results on the Columns and Sorting tabs. For more information, see on page 60 Results Type Tab.
See also #caIntegrator caIntegrator and NBIA, [#Retrieving Retrieving DICOM Images and #Example Example of Retrieving Images.
caIntegrator and NBIA
Images can be accessed in NBIA if you see buttons on the Search Results page. See the Imaging Note in on page 60 Results Type Tab. You can click links on the Search Results tab to view or download image data.
- View in NBIA – This link corresponds to each Image Series listed in the results table. If you click the link, NBIA opens to the login page. After you log in, NBIA brings up the first image in the corresponding image series; an example is shown in the following figure. You must log into NBIA to see the data. On the NBIA page that opens, you can opt to view the entire series containing this image, or you can display the image as a large JPEG-formatted image. You can also add the image to the NBIA basket. For more information, see the NBIA online help or user's guide accessible from NBIA.
- Forward to NBIA – search results:forwarding imaging results to NBIA;NBIA:forwarding imaging results toThis This button is linked to results you have selected by row. Click the button to open NBIA, where the image series you select are loaded in the NBIA image basket. In the event that the caIntegrator study was NOT configured with image annotation for an image series, caIntegrator sends NBIA a list of Study Instance UIDs, for which NBIA will add all corresponding image series to the basket. In the event that the caIntegrator study was configured with annotations for an image series, the system sends NBIA a list of Image Series IDs, for which NBIA adds all corresponding image series to the basket.
Retrieving DICOM Images
On the Imaging caIntegrator imaging data Search Results page, you can click the Retrieve DICOM Images button which is linked to results you have selected by row. caIntegrator retrieves the corresponding image(s) from NBIA through the grid. NBIA organizes the download file by patient ID, StudyInstance UID, and ImageSeries UID, and compresses it into a zip file. When caIntegrator notifies you that the file is retrieved, the DICOM Retrieval page indicates whether the retrieved files are Study Instance UIDs or Image Series UIDs, shown in the following figure. For more information, see the note below.
Click the Download DICOM link to download and save the file. caIntegrator unzips the file and displays the list of images in the file. To open the DICOM images, you must have a DICOM image viewer application installed on your computer. For more information about one such viewer, see http://www.e-dicom.
In the search results, not all of the patients in the data subset may be mapped to image series IDs. If you select a mixture of patients that have image annotations as indicated by an image series ID and patients that do not have image annotations (no image series ID), when you click the Retrieve DICOM Images button, NBIA retrieves the images for the entire NBIA study instance UID that includes the image seriesIDs you checked.
If on the Search Results tab you select only patients that have image annotations as indicated by an image series ID, when you click the Retrieve DICOM Images button, NBIA retrieves images for the NBIA image series that were matched in the search. If the results are a mixture, but you select one specific row with a valid image annotation, caIntegrator aggregates to the image series. If results are a mixture and you select multiple rows, caIntegrator aggregates to the NBIA study in which multiple image series you have selected in the search results are found.
If your query does not have image annotations and all check boxes are selected, results will go up to image series UID and gives all image series in it. Search results may ultimately depend on how the study was created. For example, if no image series display in query results, it means they were not mapped in the study. In that case, the results "move" up to Study Instance UIDs. See
To best understand this, it is important to review the hierarchy of submissions in NBIA. For more information, see .
See also
Example of Retrieving Images
If you are searching a study that has image data and image annotation(s) for at least one image series, you would follow these steps:
- Open a study that has imaging data associated with it that points to the production NBIA server.
- Make a query that will have image series or patients who are associated to Image Studies and select a few of those patients in the check box.
- Click the Retrieve Dicom Images button. Note that it aggregates to the image study.
- Now go back to Results Type tab, select all image annotations and run the query again.
- Select an image series type column and click the Retrieve Dicom Images button. caIntegrator now aggregates to the Image Series that were selected and not the Image Study.
- Select a row that doesn't have image series data, and a row that does, and push the button. This should aggregate to the study for the rows selected.
- Click Forward to NBIA. You should see the same types of aggregation for these tests.
When the image Study is in the checked boxes (regardless of image series being there or not), the system aggregates up to the Image Study level.
Relationship of Patient to Study to Series to Images
This flowchart illustrates the relationship of patient to study to series and lastly to images.
For example, the Study Instance UID is the set of images resulting from one patient office visit. When you upload a spreadsheet of an image series, the hierarchy of images in an image series might look like this:
Study Instance UID (one office visit):
Brain (image series)
- Brain image 1
- Brain image 2
- Brain image 3
Leg (image series)
- Leg image 1
- Leg image 2
- Leg image 3
You can add details for images by configuring image annotations on the Results Type tab. Annotations listed there are the column headers in the image series CSV file(s) that were uploaded to the study. Examples of image details include the following:
- All image details (name, size, etc.)
- The series that the image belongs to
- Image feature attributes
- The subject ID. Click the subject ID under Annotations on the Results Type tab to display this.
Exporting Data
| Multiexcerpt include | ||||||
|---|---|---|---|---|---|---|
|
In the search results, not all of the subjects in the data subset may be mapped to image series IDs. If you select a mixture of subjects, some of which have image annotations as indicated by an image series ID and some of which do not have image annotations (no image series ID), when you click the Retrieve DICOM Images button, NBIA retrieves the images for the entire NBIA study instance UID that includes the image seriesIDs you checked.
If on the Search Results tab you select only subjects that have image annotations as indicated by an image series ID, when you click the Retrieve DICOM Images button, NBIA retrieves images for the NBIA image series that were matched in the search. If the results are a mixture, but you select one specific row with a valid image annotation, caIntegrator aggregates to the image series. If results are a mixture and you select multiple rows, caIntegrator aggregates to the NBIA study in which multiple image series you have selected in the search results are found.
If your query does not have image annotations and all check boxes are selected, results will go up to image series UID and gives all image series in it. Search results may ultimately depend on how the study was created. For example, if no image series display in query results, it means they were not mapped in the study. In that case, the results "move" up to Study Instance UIDs.
To best understand this, it is important to review the hierarchy of submissions in NBIA. For more information, see Relationship of Subject to Study to Series to Images.
Example of Retrieving Images
If you are searching a study that has image data and image annotation(s) for at least one image series, you would follow these steps:
- Open a study in caIntegrator that has imaging data associated with it that points to the production NBIA server.
- Run a query that will have image series or subjects (patients) who are associated to Image Studies and select a few of those subjects in the check box.
- Click the Retrieve Dicom Images button. Note that it aggregates to the image study.
- Now go back to Results Type tab, select all image annotations and run the query again.
- Select an image series type column and click the Retrieve Dicom Images button. caIntegrator now aggregates to the Image Series that were selected and not the Image Study.
- Select a row that doesn't have image series data, and a row that does, and push the button. This should aggregate to the study for the rows selected.
- Click Forward to NBIA. You should see the same types of aggregation for these tests.
When the image Study is in the checked boxes (regardless of image series being there or not), the system aggregates up to the Image Study level.
| Include Page | ||||
|---|---|---|---|---|
|
Exporting Data
You search results:exporting data;exporting:dataYou can choose to download tabular search results as a CSV file. Click the Export .csv link at the bottom of the page. You may need to scroll the page to see it. The file contains the annotations, columns and data sort configurations you specified in the search query.
| Info | ||
|---|---|---|
| ||
You will not see the Export option when gene expression data displays as query results. |
| Scrollbar | ||
|---|---|---|
|
