Hands-on - Post Registration Operation¶

20 min
Moderate

Overview

This hands-on accompanies users navigating LabID to take a look at all items created during datasets registration. This includes of course (raw) datasets, but also sample, assays, and studies. This walkthrough involves

Finding an assay
Renaming an assay
Finding datasets
Finding samples
Merge samples
Annotate samples

Walkthrough

Step 1. Assays¶

Finding the Illumina sequencing assay registered during Register Datasets Hands-On 102. From the left menu of the application:

Click Assays then Illumina Sequencing

You now have been redirected to the Assay List page.

The list page displays an interactive table in which all assays are listed. This list can be filtered in different ways (ownership filter, global search, column filter). Here, considering trainees do not have many assays registered under their names, using the ownership filter should be enough to locate the assay.

By default all assays visible for the user are shown - the top ownership filter button All Group Personal is at the top-right. All includes all assays a user have permissions to see (mainly their own, the ones from their group and public ones).

To narrow down the list to only personal assays, click Personal

Assay list page with the ownership filter set on 'Personal'

Once located, click on the assay name to reach the assay detail page

You now have been redirected to the Assay Detail page.

The Assay Detail page lists all the information about the assay. Like other entities, assays can collect Annotations and Notes. The detail page also displays a lineage graph, exhibiting the Samples Assay Datasets relationships.

The Run Directory indicates where the datafiles are physically located on disks. The run directory is read-only to preserve data integrity i.e. you cannot move or rename this directory nor its content, but you can always navigate there to inspect its content e.g. using a traditional file browser. At EMBL, you can perform computation (including on the High Performace Cluster) directly on these files i.e. without making a copy.

Importantly, the assay detail page also provides an Output Datasets panel containing a list of all datasets that have been generated within this assay. This table allows you to easily navigate to the sample, study and learn about the datafile(s) included in the dataset.

This is the same list view as on the Datasets List page, but this one is filtered, displaying only the datasets belonging to this assay.

Step 2. Datasets¶

The datasets can be found on the Assay detail page, as seen above, but there is also a list page dedicated to datasets. To reach the Dataset List page:

Click Dataset Management then Datasets

You now have been redirected to the Dataset List page.

Datasets can be filtered based on their attributes. To see datasets for a given assay, use the Assay column filter. This is a free-text filter where the name of the assay of interest has to be inputted.

Step 3. Samples¶

Importantly, Samples can be retrieved (1) from the lineage graphs, and (2) from the Sample List page.

Step 3 - 1. Find details about a single sample¶

When interested in a single sample, one can go up through the Samples Assay Datasets chain, using the linage graph on either the Assay or Dataset detail pages.

Navigate to the detail page of your NGS Illumina Sequencing Assay
From the lineage viewer:
- Right-click on the xyz2 sample to open the contextual menu
- Click open details
  
  You now have been redirected to the Detail page of sample xyz2

Step 3 - 2. Find all samples¶

Samples do not list all the projects, assays or studies they are used in as attributes: only the primary project they were created for is listed. Therefore, standard column filters on Sample list pages cannot be used to restrict the list of samples to a given project (other than the primary one), study or assay. Instead, the Sample List page allows for setting a context filter.

Navigate to the Sequencing Library List page
- Click Biomaterials then Sequencing Library

You now have been redirected to the Sequencing Library List page.

Click the Context arrow button (top right of the page), and pick Set project context.

This opens a modal displaying a list page, where the project can be searched for. Select the Tea Project or Coffee Project (by clicking on the checkbox ) and validate. This closes the modal and loads the list page with the project context.

Sample list page with a project context set

What's the difference between the project context filter and the primary project table filter

At creation, samples are linked to a unique primary project which represents the project for which the sample was primarily generated. A sample only becomes associated with an assay, a dataset and thus a study when (raw) datasets are loaded (remember that a dataset must be associated with a study). Given that a sample can be subjected to different assays, a sample can therefore be associated with multiple datasets and studies. In addition, each study is associated with a project that may or may not be the same as the sample's primary project...

In conclusion, a sample:

has one primary project. You should use the usual table filter to find all the samples which primary project i e.g. Tea Project
gets associated with multiple assays, studies and projects once it is connected to dataset(s). You should use the context filter to find the samples associated (through datasets) to a particular assay, study or project.

This means that filtering the table using your projec e.g. Tea/Coffee Project might list different samples compared to when setting the project context to Tea/Coffee Project:

the context filter will list all the samples linked to datasets that belong to the Tea/Coffee Project
the table filter will list all the samples whose primary project is the Tea/Coffee Project project (no matter if they have been subjected to assays)

Step 3 - 3. Annotate samples¶

It is important to annotate samples with annotations expected for publication. To achieve this:

Go to the Sequencing Library List page
Set a context as seen above (Project: Tea Project or Project: Coffee Project)
Click the checkbox in the table header to select all samples.

Sample list page with a project context set and 4 samples selected, ready for batch editing

Open the batch editor clicking on the button (in the table toolbar at top right)

This opens up the batch editor pop-up

In the pop-up:

Select Annotations
Select the Annotation to be edited (here: GeneOfInterest)
Select the Operation (here: replace, to replace the current empty value)

Available operations differ depending on the selected annotation type. It is often only possible to "replace" a value. But when values accept a list, it is possible to add to, replace or remove from the existing value list
Enter the value BRCA1

Batch editor pop-up with a new annotation set

Click Save to validate.

A notification is displayed to confirm the annotation was added. The pop-up does not close automatically as it lets you add more annotations if needed.
Close the pop-up by pressing Quit

The list page is automatically reloaded to update the table information

When viewing items in a table, if certain items have annotations set, but the annotations are not currently visible (as columns) the Show all annotation icon displays a blue badge. Clicking the icon adds missing annotation columns

Add all relevant annotations to the table

Click the when it displays a blue badge to add all missing annotations columns to the table

Step 3 - 4. Merge samples¶

In certain situation, samples may be wrongly duplicated. This situation needs to be addressed and samples need to be deduplicated (merged). Samples merged back together aggregate relationship information. Merging samples is done from the Biomaterial List page. Here we will create a fake sample and merge it back to another sample.

Go to the Sequencing Library List page
Create a new sample:
- Click New sequencing library . On the new sample form, set:
  - the name: xxyyzz11
  - the project: Tea Project or Coffee Project
  - the organism: Zebrafish
  - the barcode: AAAAAAAAAA
- Click Save Item
Unset the context filter (if set from previous step)
Click the checkboxes to select the xxyyzz11 and xyz1 samples.
Click the button (in the table toolbar at top right)

Sample list page with 2 samples selected, ready for merging: This opens up the Merge Samples pop-up; This button is (1) disabled when less than one sample is selected, (2) orange when you do not have enough permissions on samples to perform a merge

Sample merge pop-up with a sample to keep selected

The pop-up displays the table of samples. The last column has a Keep toggle to select which sample should be kept - the other samples will be deleted - Pick one by clicking the toggle and validate by clicking the Merge samples button