If you have already created a gating hierarchy, skip to Editing an existing hierarchy

If there are no existing gating hierarchies, start by clicking on CREATE GATING HIERARCHY.

You’ll have the option to create the hierarchy manually or by running our automated classification pipeline.

Creating an automated hierarchy

Provide a name for the new hierarchy you wish to create.

You can then choose how many cells to sample for the computation. A higher number will produce more accurate results, but will take a longer time to generate results. We have found that performance increase tends to fall off around 2000, so that is likely a good starting point.

You can also configure the median absolute deviation (MAD) factor, which determines how high the cutoff thresholds will be set. A higher MAD value will raise the cutoff, and vice versa. We have found that a MAD factor of 5 is usually appropriate.

If you have any quality control filters that have been exported from the Cell Quality Control Extension in the Analyzer page, you can choose one to apply here. Any cells that are failing QC in the selected filter will be omitted from the plot view and any downstream analysis.

Hierarchy Input

The Hierarchy input must be provided in JSON format under the following guidelines:

• Each key must contain the biomarker name.
• Any “not” gates should use the not prefix in front of the biomarker name, e.g. notPD1.
• In general, it’s best practice to include a “not” gate for every gate. This allows you to retain information on cells that have uncertain classification (e.g. cells under the CD45+ gate can’t be confidently assigned to any of the immune populations can be called “Other immune”)
• Once a biomarker is used in the hierarchy, subsequent references to that biomarker in different gates should be suffixed with a double underscore and number so that all gate names are unique, e.g. PD1__1, PD1__2, etc.

Staging Region Level Configurations

We also provide the ability to select specific regions for analysis in case you want to focus your analysis on a subset of your study.

First select a Biomarker Expression Version and a Quality Filter (optional). These selections will be applied to each of the regions that you select below.