Flow Cytometry Tutorials: All About Compensation

Flow Cytometry Tutorials: All About Compensation

Published on 2 July 2020

Learn principles of compensation for your Flow Cytometry data analysis. By the end of this tutorial, you should be able to understand: What is compensation? Why we need to do compensation? How to do compensation? How to do manual correction of compensatio

  • Facebook
  • Twitter
  • Linkedin
JM
Transcript
00:00
Flow Cytometry Tutorials: All About Compensation
00:04
Objectives:
00:06
By the end of this tutorial, you should be able to understand: 
00:09
- What is compensation?
00:12
- Why we need to do compensation?
00:15
- How to do compensation?
00:18
- Manual compensation correction
00:21
- Some compensation tips
00:26
What is Compensation?
00:29
-It's the process of correcting spectral overlaps between channels
00:34
Spectral overlap
00:37
Why do we need to do compensation? 
00:41
In Flow Cytometry, target-specific signal is detected  in the form of Fluorescence
00:47
Some common sources of Fluorescence for Flow include:
00:53
Fluorochromes (aka Fluorophores)conjugated to antibodies
00:60
Fluorescent proteinsused as reportersof gene expression
01:07
Fluorescent dyessuch as DNA-bindingdyes. eg.7AAD
01:13
-Any given Fluorescent molecule exhibits characteristic excitation   and emission properties.
01:18
For example:
01:22
FITC and PE can be excited using 488nm (Blue laser)
01:28
The emission maxima of FITC is ~520nm
01:33
and the emission maxima of PE is ~575nm
01:40
Notice how FITC emission peak overlaps with the PE emission peak? 
01:48
This FITC overlap contributes false positive signal in PE channel
01:53
Therefore, compensation is needed to correct spectral overlap b/w channels 
01:58
How to do Compensation? 
02:01
-Using proper compensation controls, and optimal instrument   settings one can generate an accurate compensation matrix 
02:08
-Modern (digital) Flow Cytometers come with acquisition software    that allows Auto-compensation
02:15
-However, how good an auto-compensation wizard works depends   on how good the compensation controls are
02:22
So what "proper" controls are needed?
02:25
There are mainly two types of controls needed for compensation:
02:30
An Unstained control(cells only)
02:32
&
02:34
Single Stained Controls(one for each color)
02:38
-Unstained control takes into account intrinsic autofluorescence of a given cell-type
02:43
Therefore, highly recommended to match  unstained control to that of experimental cell type
02:49
-For compensation calculation itself, internal negative(negative peak of the single stains) can be used
02:55
But, unstained control is still required for voltage setup
02:59
-When experimental cells cannot be used as unstainedcontrols (eg. when dealing with very few cells),compensation beads may be used.
03:08
Several commercial beads are available
03:13
Single Stained Controls
03:15
-Single stained controls are needed one for eachfluorochrome in the panel
03:19
-Must be the same fluorochrome as in the samples
03:23
eg. If you have FITC, only FITC can be used butnot GFP or AF488 (even when they fall in same channel)
03:29
-Preferably the same antibody, but can be switched  with another antibody (if marker expression is low)
03:35
substitute antibody must target the same cellsas experimental cells 
03:40
-Compensation beads may be used if short on cells   or marker expression is low.
03:45
-Both beads and cells can also be combined
03:49
eg. when using viability dye, you can use cells for dye, and beads for Fluors  
03:54
Last but not the least, single stained controls must be at least as bright as samples
04:01
How to set the optimal PMT voltages?
04:05
-An optimal voltage is where:
04:08
-The positive signal is detected above background
04:13
-The positive peak is not off scale (because of too high voltage)
04:21
-The resolution of the positive and/or dim population is not lost   (because of too low voltage)
04:34
How to calculate compensation?
04:38
-Following are the 3 requirements to calculate compensation
04:42
1. Once set, the optimal voltages need to be the same across channels
04:46
Unstained tube and all the single color tubes need to be recorded at thesesame voltages
04:52
2. Same gated population needs to be used for compensation calculation 
04:57
For example:
04:59
If you are analyzing lymphocytes,then same lymphocyte gate needs to beapplied to all the comp' controls
05:08
In FACSDiva software, you can do thisby right-clicking on the gate and then selecting "Apply to all..." option  
05:14
3.
05:15
Positive gate on each single color control needs to be adjusted toencompass true positive
05:21
This gate can be different for different single color controls
05:25
Once these requirements are met, you can calculate compensations
05:29
In FACSDiva, you can do this by selectingExperiment>Compensation Setup>Calculate Compensations
05:36
How to do manual compensation correction? 
05:39
Often times, auto-calculated compensation values are not accurate
05:43
This is more true when single color controls lack a distinct positive peak
05:47
Manual compensation correction can be done by :
05:51
- Statistically, by comparing MFIs
05:54
while adjusting compensation matrix created by auto-comp wizard
05:58
Manual correction by comparing Median Fluorescence Intensities (MFIs):
06:02
* This is the most accurate method 
06:04
Let's assume you have FITC & PE Fluors
06:07
FITC-A
06:07
PE-A
06:09
The four possible populations would be...
06:12
DoubleNegative
06:14
FITC+
06:16
DoublePositive
06:17
PE+
06:19
In a well compensated sample:
06:21
The PE MFI value of FITC+ populationaligns closely with the double negative
06:29
PE MFI=9
06:30
PE MFI=8
06:32
The FITC MFI value for the PE+ populationaligns closely with the double negative
06:39
FITC MFI=10
06:40
FITC MFI=11
06:43
When FITC is UNDERcompensated:
06:45
 PE MFI for FITC+ population looks higher(as FITC signal bleeds into PE channel)
06:51
FITC+ UNDERcompensatedPE MFI= 153
06:53
When FITC is OVERcompensated:
06:56
 PE MFI for FITC+ population looks lowerthan the double negative population
06:60
FITC+ OVERcompensatedPE MFI=  in negative values
07:04
Same compensation rules apply for thePE+ population (with only difference is -we compare FITC-MFIs for PE+ population)
07:13
Summary 
07:14
-Compensation is a process of correcting spectral overlaps between channels
07:18
-Modern (digital) Flow Cytometers provide auto-compensation wizards 
07:22
-Auto-compensation wizard works only as good as compensation controls 
07:26
-Proper controls & optimal voltage settings result in an accurate comp' matrix
07:30
-Auto-comps can be corrected manually (if needed) by looking at MFI values
07:34
-Manual correction involves comparing MFIs of single positive populations  with double negative population within a sample
07:41
-3 words to remember: UNDERcompensated, OVERcompensated, and  well-compensated
07:49
Some compensation tips... 
07:51
In FACSDiva software, compensation matrix can be adjusted underCytometer Settings option
08:01
To show MFIs in FACSDiva software, right click on the plot showingparameters that need to be compensated, and select "Create Statistics"
08:08
Here is an example statistics view showing MFIs for FITC and PE plot:
08:12
Compensated Sample
08:14
UNDERCompensated for FITC
08:16
Notice how the PE-MFIs are higher for FITC+ (Q4) vs double negative (Q3)
08:24
When doing manual compensation correction, always make sure touse Biexponential display to show events below zero
08:31
On regular Log Scale
08:31
With Biexponential display
08:34
Without Biexponential display, noticehow the events below zeroare hidden - making it tricky to knowwhere the populations are
08:41
In FACSDiva software, Biexponential display can be turned onin the inspector window.
08:51
What looks uncompensated is NOT always uncompensated...
08:55
Sometimes it could be just biology rather than compensation 
08:59
For example: "Steric hindrance" causes populations to look UNDERcomp'd
09:05
Notice how the populationlooks undercompensated?
09:09
This phenomenon occurs when twoantibodies compete to occupy same space
09:14
Anti-TCRαβ (clone WT31)binds to a conformational epitope formed by the TCR and epsilon chain.
09:19
CD3 antibody binds to theepsilon chain of theTCR complex
09:24
Further Reading...
09:27
- Compensation by Dr. Mario Roderer: http://www.drmr.com/compensation/
09:29
- Technical bulletin on Compensation by BD:https://www.bdbiosciences.com/documents/Compensation_Multicolor_TechBulletin.pdf
09:31
-Compensation guidelines by Excyte:https://expert.cheekyscientist.com/guidelines-for-setting-compensation-controls-in-flow-cytometry-experiments
09:34
CreditsConceptualized and Created ByJavid MohammedBackground Music www.bensound.com
09:40
Please Subscribe For More...
09:42
The Flow Cytometry Networkwww.thefcn.org