v10 Documentation
Search

Sample Quality Control

Sample Quality Control (QC) is crucial in data analysis.

Despite all care during samples preparation and instrument setup, issues can arise when running the acquisition on the instrument (i.e. most common ones are clogs and air bubbles). Those issues may result in the alteration of the measurement of the parameters which can lead to false discovery. The sample quality control platform can evaluate the fcs files and remove cells that were contained in areas of irregularity that occurred during acquisition. 

A graph showing a graph of a graph Description automatically generated with medium confidence

Figure 1. Importance of running QC on samples

There are many great R based algorithms available, but FlowJoTM  Version 11 has its own integrated QC tool.  In the navigation bar, click on the “Quality control” icon (Figure 2.1).  Select  the parameters that will be checked (Figure 2.2) and adjust quality control settings in the “Properties panel” (Figure 2.3, 2.4, 2.5). Information regarding each setting can be found below. 

 A screenshot of a computer Description automatically generated

A table of text with black text Description automatically generated with medium confidence

Figure 2. Sample quality control (QC) 

 

Signal stability

  • Changepoints maximum: A changepoint is a time point where a significant change occurs in the mean or variance in the data across the time of acquisition. For instance, a changepoint will often occur if the flow rate suddenly changes. The changepoint maximum is the maximum number of change points that can be detected for each channel. 
  • Changepoints penalty: The changepoint penalty controls the amount of change in the mean or variance required to label a timepoint as a changepoint. Raise the penalty value to be less strict in the anomalies detection. 
  • Remove Outliers: If selected, the Remove outliers option in our Quality Control platform will remove small spikes in the flow rate even if the signal stabilizes afterwards.
Figure 3. The effect of selecting Remove Outliers. With Remove Outliers selected there are many tiny yellow regions in the middle of the acquisition that are getting removed because they briefly had high variance in their signal. With Remove Outliers deselected these areas are not removed because their signal quickly normalizes after these tiny patches of irregularity.  


Flow rate 

MAD range: The median absolute deviation (MAD) is used in a Extreme Studentized Deviate test to identify outliers. The value controls the deviations from the peak range that are allowed. Lower the value to be more strict (Default 3.5 – Range 0.25 to 5).

 

Dynamic range 

Select whether the dynamic range check should be performed on the upper limit, the lower limit, or both. The Dynamic range option will remove any events with a parameter that contains either the maxmimum (upper limit) or minimum (lower limit) value recorded in the fcs files by the cytometer. 


On completion

After running QC on your data, a green check mark will be on displayed on the group “Experiment Data”, a set of Good Events/Bad Events populations will be created  and displayed in the Populations panel (Figure 4).  

Figure 4. Green check mark after performing QC. Bad Events and Good Events populations are now shown in the populations panel. 

A full QC report will be displayed in the Discovery panel. The Overlay shows Bad Events and Good events vs Time plot, where blue are the Good events and red are the events removed after QC (Bad Events). You can also chose to display a parameter of your choice on the Y axis (Figure 5). 

A screenshot of a computer Description automatically generated

Figure 5. QC Report