In flow cytometry, compensation is the process of correcting spillover from one fluorescent channel to another. When you label your samples with multiple antibodies, the fluorescent probes on the antibodies may have similar emission spectra, meaning they will emit fluorescent light at similar wavelengths (See this handy emission spectra chart to find your antibody's spectra). Your cytometer will therefore record the fluorescence for both at similar levels, meaning that your populations will ‘look’ similar on your flow plot. This will make it difficult to properly gate your desired cell population. By compensating between fluorescent channels, you can correct for this spillover and more easily separate out your target population.
How do I know if I will need compensation?
To know if you will need to compensate in your flow panel, you will need to know both the fluorescent markers you intend to use and your cytometer’s laser configuration. Your flow core or an online manual for your cytometer should be able to tell you the laser configuration. In general, if two fluorescent markers are read by the same laser, they will need to be compensated for one another. This overlap in spectra is often called "spillover." For example, if your flow cytometer reads out the fluorophores FITC and PE on the same laser, you will need to compensate between the two of them because they emit at a similar wavelength. However, if your flow cytometer reads them out on different lasers, you should not need to compensate between them.
Another rule of thumb is if the names of the fluorescent markers are similar, you will need to compensate between them (although there are exceptions). For example, on most flow cytometers you will need compensation between APC and APC-Cy7; PE and PE-Cy7; APC-Cy7 and PE-Cy7; or BV421 and BV51. It is a good idea to look up new fluorescent markers before deciding to use them, using a tool like BioLegend’s Fluorophore Guide Chart.
Pro tip! Some antibodies go by different names depending on your flow cytometer’s setup. For instance, BV510 and AmCyan have almost identical emission spectra but may be called BV510 on one cytometer and AmCyan on another.
What does spillover look like and how do I fix it?
When plotting two channels against each other, you should be able to draw perpendicular lines that intersect your populations. If you see a diagonal population, this is a sign of spillover and indicates that it is necessary to apply a compensation value between the two channels (Figure 1A). Applying compensation means you are telling your cytometer to adjust the fluorescent signals by a specific value between the two channels. It is something your computer will do for you to better analyze your results; it does NOT change anything in your samples. This means that, unlike voltages, you can adjust compensation values later in your analysis, unlike voltages.
Figure 1A shows a single-color APC control, with APC and APC-CY7 plotted against each other. Since it is a single-color control, you know the beads should only be positive for APC, yet it seems as if the beads are also positive for APC-Cy7. You therefore would say that APC is “bleeding,” or spilling over, into APC-Cy7 channel and needs to be compensated out of APC-Cy7. After increasing the compensation values for APC against APC-Cy7, the populations are in more of a straight line and are only positive for APC (Figure 1B).
For single-color samples, you can only change the compensation values for whichever fluorophore is present in the sample (APC in the case of Figure 1). In Figure 1A, the population needs to be shifted towards the x-axis, so you would increase the compensation value for APC. If the population needed to be shifted away from the x-axis, you would decrease the compensation value.
 |
| Figure 1: Compensation example with single color control. Flow plots showing an APC single-color control sample before (A) and after (B) compensation with APC-Cy7. The positive APC population is circled in red in both panels. |
Pro tip! To determine which way your population needs to be shifted, consider your perpendicular lines. In this case, the population on the left is a horizontal line, so our circled population on the right needs to shift towards the X axis to become a perpendicular vertical line.
In general, most spillover between samples can be fixed with fairly low compensation values (0-50). Inputting compensation values differs between machines or the program you’re using, but usually you can click on the correct box and either manually change the compensation values or adjust the values with up/down arrow buttons.
For anyone just starting to use flow cytometry, it’s generally recommended to start compensating with single-color controls before moving onto your experimental samples. With time and practice, you may be able to do your compensation setup on your experimental samples without needing to use your single-color controls as a starting point.
Applying compensation to your experimental samples
In order to do compensation on your experimental samples, you will first need to gate out your main population and singlets. Then, you can start to compensate on your experimental samples. Let's look at an experimental example in Figure 2.
Figure 2A shows the sample prior to compensation. APC and APC-Cy7 are both bleeding into each other, as is evident by the two diagonal populations in the plot. A compensation value of five was applied to APC-Cy7 against APC to correct for this bleed. A compensation value of 3.7 for APC into APC-Cy7 applied next, which fixed the bleed in this population. Note that in the corrected flow plot in 2A, the populations are not perfectly straight, but the center of each population is in a line with the other populations.
 |
| Figure 2: Compensation example based on experimental sample. Flow plots showing a sample before (A) and after (B) compensation between APC and APC-Cy7, with compensation panels below each plot. |
The general rules of compensation apply to your experimental samples as they did with your single control samples: if the population needs to be shifted towards an axis, increase the compensation value, whereas if it the population needs to be shifted away from an axis, decrease the compensation value.
If neither of these helps shift your population into the correct spot, you may need to look at the other channels you’re using first. A different fluorophore channel may be affecting your populations and may need to be “comped” first before you apply compensation values to your current channels.
A notable exception
If you have a sample you are trying to compensate in the top right corner of a flow plot, meaning that it should be double-positive for the two markers on your plot, and the population does not move even after applying very large compensation values (100+), consider what two markers your fluorescent markers are staining for. If a cell is positive for both markers, it will show up as a diagonal line in the top right corner of the plot and large compensation values will not move it (Figure 3). If the population is not a double-positive population and is instead due to bleeding, you will see the populations move, as in Figure 2, following compensation.
 |
| Figure 3: Flow plot for a double positive sample. The cell population in the top right corner of the plot is positive for both markers, so it shows up on a bit of a diagonal line in the plot. However, because we can see straight lines in the populations in the other corners of the dot plot, compensation is not required. |
In Figure 3, we also know that on most cytometers, FITC and APC should not require compensation between each other. We can therefore conclude that this population is double-positive for both fluorophores and that there is no issue with compensation here.
Conclusion
Compensating between your fluorescent markers can be challenging! But we hope this guide can help you get started. Remember, you can always change your compensation values during the analysis portion of your experiment. So, as long as you are able to distinguish your populations from each other on your flow cytometer, you should be able to run your experiment and finalize your compensation values later. This can be helpful as you're learning!
Ashlyn has a research background in CD8+ T cells and cancer immunotherapies. She is currently a research technician in Arlene Sharpe's lab at Harvard Medical School.
More resources on the Addgene blog
Introduction to Gating in Flow Cytometry
Topics: Antibodies, antibodies 101
Leave a Comment