TABLE OF CONTENTS

• 1. Use Cases
• 2. Accessing the metadata editing module
  •     2.1 From the workspace 
  •     2.2 From plots 
• 3. Using the metadata module
  •     3.1 Locking columns
    •          3.1.1 Showing or hiding locked columns
  •    3.2. Copying columns
  •     3.3. Editing data
  •     3.4. Versioning
  •     3.5. Facets: edit all underlying
  •     3.6. Standard options
• 4. Appending metadata

NOTICE: only columns that are locked will be shown in your analyses

The metadata editing module is based on the OpenRefine software. Please review the OpenRefine reference page for details.

1. Use Cases

Metadata editing can be useful at multiple points during your analyses. Some examples are:

• Harmonize your data at the beginning of your project
  • Adjust your metadata after quality controls (such as missing values etc. 
  • Adjust the naming to a your lab standard
  • 
    
• Add additional data during your data analyses, based on the insights you are gaining
  •  You want to add extra metadata based on the certain analyses that you ran, such as cluster names.
  • You want to curate your dataset further

2. Accessing the metadata editing module

    2.1 From the workspace 

To start with editing your metadata, you go to the 'edit' button on the tile of your workspace, in this window you can see the 'Assign dataset' section. Next to the individual datasets, you can access the metadata editing module by clicking on the blue edit icon.

    2.2 From plots 

You can also access the metadata editing from individual plots by clicking on 'EDIT' on the plot tile and navigating to the edit metadata field.

3. Using the metadata module

The metadata module has an extensive repertoire of functionalities that can be explored by the clients. 

The most important ones are described here. 

    3.1 Locking columns

When clicking on the dropdown menu of a metadata variable, the user can select 'Lock' to lock a column.

This will grey out said column and will disable any further editing.

This step is irreversible, this column will not be able to be edited at any point in the future.

When the user does want to change something in this column, the remaining option is to copy the column and make the edits in the copied column. Once a column is locked it can be selected as input throughout the UniApp (e.g. color-coding for dimension reduction plots, observation selection, the design step for differential and marker gene analyses).

         3.1.1 Showing or hiding locked columns

You can choose to show or hide locked columns by hovering over 'locked'. This will help with keeping the interactive mode of your plots organized so that you only see the information that you need there. You can adapt his at any point.

   3.2. Copying columns

The user can copy columns using the dropdown menu  > edit column > Add column based on this column.

This can be used to add for example a clustering column based on an existing column, without losing the original input.

This is also helpful when the user would like to change input in the data from a locked column, as described above.

When adding a copy, the user can choose the naming of the copy of the column, the input will be exactly the same as the mother column.

    3.3. Editing data

The foremost functionality of the metadata editing is of course editing the existing data. 

For this, the user has to hover over the cell containing the data that needs to be changed. Here, the purple 'edit' box will appear. When clicked, this opens an editing screen where the desired changes can be made. When done, the user clicks 'apply' and the changes will be implemented. 

    3.4. Versioning

To be able to keep track of what changes were made to the data, there is the Undo/Redo tab where an oversight is provided of everything that was done in the metadata.

In this tab, all changes are listed in chronological order. 

When selecting a step (i.e. a previously made change), the metadata matrix will reboot to how it was implemented at said moment that this step was implemented.

    3.5. Facets: edit all underlying

Facets can be selected under the dropdown menu > Facet. 

When selecting the text facet, the text data from that column will be displayed in groups, with the number of datapoints that every group contains. 

This is also the case for the numeric facet. 

The Timeline facet shows the timeline of the datapoints.

Metadata variables containing only numeric values will only be able to showcase numeric facets. The same goes for text variables. 

When hovering over a certain facet (group), the user is able to edit the data. When implemented, this edits all datapoints in this facet. 

When for example a certain text datapoint is named 3 times 'NA' (capital letters) and 4 times 'na' (without capital letters) and this represents the same, this should be in the same facet. By editing the 'NA' facet to 'na', the user will create a facet 'na' with 7 values in it.

The same principle is applicable to numeric facets.

    3.6. Standard options

The standard options include but are not limited to:

   - Editing colums:

• Splitting columns
• Joining columns
• Renaming columns
• Removing columns
• Moving columns
  • to beginning
  • to end
  • left 
  • right

    - Transposing cells

   - View:

• Collapsing a column
• Collapsing all other columns
• Collapsing all columns to the left
• Collapsing all columns to the right
• Expanding all columns
• Expanding all columns to the left
• Expanding all columns to the right

4. Appending metadata

The metadata editor allows you to append new metadata columns to the existing metadata in the UniApp. To accomplish this, you first need to download the Append Metadata Template, by clicking the "Download template" button in the metadata editor.

On your PC, you will find the downloaded template as a CSV file. The naming convention for the file is:
template-file-[DATASET ID].csv 

Next, add the columns you want to append to this file, either manually or using tools like R or Python. 

In this example, we will manually add columns to the template using Excel. When you open the template file, it will look like this, containing only the Observation column.

It is critical not to modify anything in the first column, labeled as Observation. This column acts as the key linking the appended rows to the corresponding rows in the UniApp metadata. Do not change the column name, values, position, or delete the column.

Now, we can add new columns to the template. It is possible to add either categorical or numerical columns. The first row of the new columns should contain the column names, which will appear as such in the UniApp after appending. 

Once you finish editing, save the file and return to the UniApp. Press the "Append metadata" button. 

Select the template file with the columns you wish to append. Optionally, add a suffix if needed, and then click the "Confirm upload" button to append the metadata. 

If your template contains column names that already exist in the UniApp, you will be prompted to use different column names. This can be easily accomplished by adding a unique suffix.

The data upload will now be processed.

After a few minutes, you will see the appended columns appear in the UniApp metadata editor.