The Quality Control Extension allows you to set Cell QC parameters for filtering away low quality cells before using the Gating Extension or the Unsupervised Clustering Extension for cell classifications.

Once you Launch the module, click on the + icon to Create New Configuration. Add a custom configuration name, and a Description (optional).

From there, select a Biomarker Expression Version, the select the regions that you’d like to QC (with the option to use the Regions Filter if needed), an Existing Configuration (optional), and then click Stage Configuration. Once you have staged all of the region level configurations you would like in your new Cell QC configuration, click Submit All Configurations.

Once you click create, you will see the following page layout (it may take a few moments to load):

The four metrics that can be adjusted for QC are Cell size, Signal sum, DNA signal, and Signal coefficient of variation (CV). These parameters can be adjusted using the slider bars on the left hand side.

• Cell size indicates the size in pixels of the dilated segmentation masks for each cell object. Poor quality objects will have abnormally large or small size.
• DNA signal indicates the mean pixel intensity of the DNA channel for each cell object.
• Signal sum indicates the total raw signal intensity across all channels for each cell object, where signal intensities are normalized by object area. Artifacts will tend to have abnormally high signal sum, as they stain non-specifically across all channels. Poorly stained areas may have abnormally low signal sum.
• Signal coefficient of variation indicates the coefficient of variation (standard deviation divided by the mean) of the vector of biomarker intensities for each cell object. Artifacts will typically have abnormally low signal CV, since they have intense stains across many channels.

By default, these are log10 transformed, but you can view the raw metric values by un-selecting the Log10 Transform Metrics checkbox at the top of the page.

In the Metric drop-down menu, select the metric you would like to adjust, and in the Region drop-down menu, select the region that you would like to view. This will pull up a histogram plot of the selected metric, as well as a spatial coordinate plot of passing and failing cells, for the selected region.

Use the slider bars to adjust the metrics as necessary. You will notice that as you restrict the parameters for a passing cell, gray dots indicating failing cells will appear in the right panel.

Changes to the metric thresholds can be applied to a specific region, to all regions within an experiment, or even all regions in the study.

When you apply threshold changes to an experiment, cells within every region in that experiment will be filtered based on the current thresholds.

There are five buttons on the right side of the spatial coordinate plot:

• Panning tool allows you to click and drag across the image.
• Lasso tool and Box selection tool both allow you to focus your analysis on a subset of cells in a region. The Lasso tool allows you to draw a freeform region of interest, while the box selection tool allows you to draw a rectangular region of interest. In both cases, the cells in the region of interest become highlighted, and their distribution on each of the QC metrics can be shown in the histogram on the left.