...

Filter low-quality cells

An important step in analyzing single cell RNA-Seq data is to filter out low-quality cells. A few examples of low-quality cells are doublets, cells damaged during cell isolation, or cells with too few coutns to be analyzed. In a CITE-Seq experiment, protein aggregation in the antibody staining reagents can cause a cell to have a very high number of counts. These are low-quality cells that can be excluded. Additionally, if all cells in a data set are expected to show a baseline level of expression for one of the antibodies used, it may be appropriate to filter out cells with very low counts or a low number of detected features. You can do this in Partek Flow using the Single cell QA/QC task. 

We will start with the protein data.

• Click the Antibody Capture data node
• Click QA/QC in the toolbox
• Click Single Cell QA/QC

This produces a Single-cell QA/QC task node (Figure ?)

...

• Double-click the Single cell QA/QC task node to open the task report

The task report lists the number of counts per cell and the number of detected features per cell in two violin plots. For more information, please see our documentation for the Single cell QA/QC task. For this analysis, we will set a maximum counts threshold to exclude potential protein aggregates and, because we expect every cell to be bound by several antibodies, we will also set a minimum counts threshold. 

• Click the Single cell QA/QC node once it finishes running
• Click Task report on the task menu

The Single cell QA/QC report opens in a new data viewer session. There are interactive violin plots showing the most commonly used quality metrics for each cell from all samples combined (Figure ?). For this data set, there are two relevant plots: the total count per cell and the number of detected genes per cell. Each point on the plots is a cell and the violins illustrate the distribution of values for the y-axis metric. Because mitochondrial transcripts are not present in protein data, this plot is not informative for this data set.

...

• Click Image Modifiedin the Filtering card on the right
• Click Apply filter...
• Select the Antibody Capture data node as input in the pipeline preview (Figure ?)
• Click Select

...

• Click OK to dismiss the message
• Click the project name at the top to go back to the Analyses tab
• Your browser may warn you that any unsaved changes to the data viewer session will be lost. Ignore this message and proceed to the Analyses tab

A new task, Filter counts, is added to the Analyses tab. This task produces a new Filter counts data node.

Next, we can repeat this process for the Gene Expression data node. 

• Click the Gene Expression data node
• Click the QA/QC section in the toolbox
• Click Single Cell QA/QC

This produces a Single-cell QA/QC task node

• Double-click the Single cell QA/QC task node to open the task report

The task report lists the number of counts per cell, the number of detected features per cell, and the percentage of mitochondrial reads per cell in three violin plots. For this analysis, we will set a maximum counts threshold maximum and minimum thresholds for total counts and detected genes to exclude potential doublets and a maximum mitochondrial reads percentage filter to exclude potential dead or dying cells. 

• In the Selection card on the right, set the Counts threshold to keep cells between 1500 and 15000 
• Set the Detected features to keep cells between 400 and 4000
• Set the % Mitochondrial counts to keep cells between 0% and 20% (Figure ?)

• Click Image Modifiedin the Filtering card on the right
• Click Apply filter
• Select the Gene Expression data node as input in the pipeline preview
• Click Select
• Click OK to dismiss the message about the task being enqueued
• Click the project name at the top to go back to the Analyses tab
• Your browser may warn you that any unsaved changes to the data viewer session will be lost. Ignore this message and proceed to the Analyses tab

A new task, Filter counts, is added to the Analyses tab. This task produces a new Filter counts data node (Figure ?)

Normalization

After excluding low-quality cells, we can normalize the data. 

We will start with the protein data.

• Click the Filtered counts data node produced by filtering the Antibody Capture data node
• Click Normalization and scaling in the toolbox
• Click Normalization
• Click the green Image Modified button
• Click Finish to run (Figure ?)

The recommended normalization for protein data includes the following steps:  Add 1, Divide by Geometric mean, Add 1, log base 2. This is a variant of Centered log-ratio (CLR), which was used to normalize antibody capture protein counts data in the paper that introduced CITE-Seq [1] and in subsequent publications on similar assays [2. 3]. CLR normalization includes the following steps: Add 1, Divide by Geometric mean, Add 1, log base e. Normalizing the protein data to base 2 instead of e allows for better integration with gene expression data further downstream. If you would prefer to use CLR, click and drag CLR from the panel on the left to the right. 

Normalization produces a Normalized counts data node on the Antibody Capture branch of the pipeline. 

Next, we can normalize the mRNA data. We will use the recommended normalization method in Partek Flow, which accounts for differences in library size, or the total number of UMI counts, per cell and log transforms the data.

• Click the Filtered counts data node produced by filtering the Gene Expression data node
• Click the Normalization and scaling section in the toolbox
• Click Normalization
• Click the Image Modified button 
• Click Finish to run (Figure ?)

...