How to create cell selections

‍

Lasso tool

The lasso tool can be used to draw freely on the UMAP to select a group of cells and a shortcut to save the selection of cells.

‍

Add/subtract from selection

Variation of the lasso tool where you can draw around cells on the UMAP to add or remove cells from your current selection.

‍

Adjust

You can open the adjust selection panel to select groups of cells. See more on Adjusting selections.

‍

Adjusting selections

You can add or exclude cells from your initial cell selections, here we will show steps on how you can achieve this on the platform.

‍

Steps

Step 1. From the explorer page, you can open the Adjust selections panel from the button below the main plot.

Step 2. You will have the options to adjust your initial selected cells from the UMAP. You can add or remove cells by clusters/groups or by the expression value of a feature.

Step 3. You can use the checkboxes to add or exclude groups of cells from your selection

Step 4. From the values tab, you can select cells by "cell gating". This means you can search for a feature from the feature search bar then select cells based of their expression level of that feature. Using the sliders you can adjust the selection by the expression value of a feature.

Step 5. You can save the selection after your adjustments for other operations.

Step 6. These selections can be used for analyses such as differential gene expression, extract to a new dataset or use them to create a custom category.

‍

Manipulating saved selections

Saved selections of cells can be further managed to create new selections. These tools can be used exclude, intersect or unit selections of cells. A new selection will be created and the originals will still be available when you use Unite, Intersect or Exclude.

‍

Exclude

When you have multiple selections, you can use Exclude to create a new selection which excludes any overlapping cells.

Steps

Step 1. From the explorer page, open the Selections panel.

Step 2. Here, the Exclude function will combine the chosen selections and remove any overlapping cells, i.e. only cells that are unique in each of the chosen selection. Overlapping cells are cells that exist in at least two of the selections.

For example, if you have a selection of all B cells in a dataset, and you would like to exclude the B cells from sample_7. You can use a selection of sample_7 B cells and the Exclude function to create a new selection of B-cells with cells from sample_7 removed.

Step 3. Use the checkboxes to select the saved selections and click on Exclude.

‍

Intersect

You can create an intersection from the sets of cell selections. This can be a useful tool if you would like to find and create selection cells that that belong to specific clusters and groups.

Steps

Step 1. From the explorer page, open the Selections panel.

Step 2. With the intersect function, you can create a new selection of the cells that are overlapping in the chosen selections. Non-overlapping cells or cells that are unique to certain selections will not be included. You can use Unite if you want to include all the cells from the chosen selections.

For example, we want to find B cells that are from sample_7. Here we will use the intersect function to create a new selection of cells that exist in sample_7 and also exist in B cells.

Step 1. Use the checkboxes to choose the selections and click on the Intersect button

Step 2. You can also rename the selection, and or extract the selection into a new dataset.

‍

Unite

The Unite function will create a union of all cells in the chosen selections. This includes overlapping cells and cells that are unique to the selections.

Steps

Step 1. From the explorer page, open the Selections panel.

Step 2. You can create a union of selections using the Unite function. This will create a new selection which includes the set of all the selected cells.

In this example, we have selected 2 groups of cells, B cells 1 and B cells 2. We want to create a union of B cells from these 2 groups.

Step 3. Use the checkboxes to select your saved selections.

‍

Subset or extracting selections into a new dataset

When you have saved selections, you can extract the selections into a new dataset. Subsetting a dataset can be useful when you are interested in reclustering or reanalysing specific groups of cells in your dataset.

Steps

Step 1. From the explorer page, open the selections panel.

Step 2. Make sure you have saved the selection of cells you would like to extract, and click on the the extract button from the selection options.

Here, as an example, we are only interested in this group of "B cells" and would like to recluster just these cells in a new analysis. In this case, we can choose to extract these cells into a new dataset.

Step 3. A new dataset will be created for the subset of cells, you may find the new dataset in your Inbox. The original dataset that the subset was extracted from will remain the same.

Note: Extracting less than 50 cells into a dataset will most likely fail since there are not enough cells to run an analysis.

‍

Custom categories using selections

Steps

Step 1. Here in this example, we have saved selections of cells by their cell types. We will create a new categorical based of these selections.

Step 2. We have annotated our Leiden clusters and some clusters need to be merged. For example, we have clusters T cell 1 and T cell 2 which we would like to merge into T cells in our new custom category. Select the 2 clusters and save as "T cells".

Step 3. Now we can choose the selections we would like to include in our new category. When we create a new category, we categorize all the cells that were analysed in the dataset. You can create a category from just "B cells" and "T cells"  for example, but all other cells will be grouped into "None" or “Uncategorized”.

Step 4. Here, we named our new custom category "cell_annotations"

Step 5. Once the custom category has been created, you can view the selections as groups/clusters. You can easily edit the name and colours of your new category.