Saturday, March 26, 2016

What Is Flow Cytometry Light Scatter And How Cell Size And Particle Size Affects It

Written by Mike Kissner


Light scatter is unfortunately one of the more misunderstood concepts in cytometry.


When you first learn flow cytometry, you were likely told the misleading phrase, “Forward scatter signal intensity is proportional to cell size, and side scatter signal intensity is proportional to cell granularity.”


The short story here is that, yes, this can be the case when we’re dealing with typical cells (consider blood granulocytes, which are bigger than lymphocytes, have more intense FSC signals; granulocytes are more “granular” than lymphocytes and thus have more intense SSC signals). However, the truth is that light scatter complexity is belied by this perennial but unsatisfactory introductory explanation.


While particle size (particle radius) certainly does influence light scatter signal, its intensity is a function of a combination of factors, including:



  •   wavelength of laser illumination

  •   collection angle, and

  •   refractive index of the particle and flow medium (sheath) 


It turns out that light scatter intensity has a strong dependence on the relationship between the size of the particle to the wavelength of the laser. More specifically, particles with diameters that are larger than the wavelength of the laser will scatter light with a different pattern than particles that are smaller than the wavelength of the laser.


Keep in mind that we most typically measure scatter using 488 nm excitation (and sometimes 405 nm excitation), so particles or cells with diameters larger than about 0.5 microns will behave differently than particles with diameters significantly smaller than 0.5 microns.


It turns out that there is a physical theory, named after the German physicist Gustav Mie (1869-1957), that predicts and explains the behavior of light scattering of particles larger than the wavelength of illumination. Essentially, Mie’s theory predicts that the intensity of scattered light has a strong angular dependence.


In other words, the intensity of the signal generated from scattered light depends on the angle at which we collect and direct the light towards a detector. Mie’s theory explains why we use a forward scatter (FSC) detector to measure the light scatter signal of typical mammalian cells, which are much bigger than the wavelength of the illumination source (typically the 488 nm laser).


How Small Particles Affect Forward Scatter


Forward scatter detectors collect light at small angles relative to the incident beam and can take advantage of the fact that cells preferentially scatter light in this “forward” direction.


Additionally, because cells scatter so much light in the forward direction, we can save money and use a less sensitive detector to measure this light. As such, forward scattered light is traditionally and often effectively measured with a photodiode, rather than the more sensitive photomultiplier used to measure fluorescence and side scatter.


However, the story is quite different for particles that are smaller than the illumination wavelength (<488 nm or <~0.5 um), like microvesicles and ectosomesIt turns out that light scatter by particles of this size range is NOT dependent on the angle at which it is measured. This has very significant implications for small particle analysis. Most importantly, because small particles do not preferentially scatter light in the forward direction as do cells, the resolution in this detector may not be sufficient to measure or even identify these particles above background.


How Small Particles Affect Side Scatter


Side scatter, given its orientation in the quieter fluorescence collection path, as well as its traditional detection by the much more sensitive photomultiplier tube, will likely have much better resolution than forward scatter.


Similarly, we often use side scatter as the “trigger” or “threshold” parameter when measuring small things like bacteria, microparticles, or microvesicles for the same reason—better sensitivity allows us to better distinguish and measure small particles above background.


Interestingly, scatter gets dim VERY quickly when particles have diameters below the wavelength of illuminating light, considering that scatter intensity decreases with a dependence on r6 of the particle. The bottom line is that small things like extracellular vesicles can be incredibly difficult to detect using scatter signals. This can make publishing flow cytometry data on small particles very difficult.


What Is An Obscuration Bar?


Instrument manufacturers and operators often take advantage of this property of small particle light scatter by installing an adjustable obscuration bar on the forward scatter detector.


The forward scatter obscuration bar is a universal component of this detector that helps to diminish background in the FSC detector by blocking laser light from interacting with the detector. When no particle is present in the laser beam, scattered laser light hits and is blocked by the obscuration bar. On the other hand, when a particle is present in the laser beam, laser light refracted (scattered) by the particle passes over the bar and triggers signal associated with that particle.


An adjustable obscuration bar can be rotated to expose a wider or narrower surface to the laser beam, blocking more or less laser light, respectively, from hitting the detector. To resolve smaller particles from optical noise, it can be helpful to block more laser light from hitting the detector, which can be accomplished by widening the bar.


Moreover, because small particles do not preferentially scatter light in the forward direction, the proportion of signal blocked by the bar of the total signal is less significant than it would be for larger particles, which do preferentially scatter in the forward direction. This strategy is most effective when a photomultiplier tube is used for forward scatter detection than a photodiode, given that the former is much more sensitive than the latter.


This relationship between particle size, wavelength of illumination, and scatter angle can also help explain the side scatter properties of cells. Typical cellular side scatter signal is much more robustly correlated to granularity than forward scatter is to cell size. In fact, the cytoplasmic “granules” that influence side scatter signal are often smaller than 0.5 um and will thus scatter in a non-Mie pattern. Side scatter of mammalian cells and side scatter of small particles are not terribly different after all.


What Is A Refraction Index?


In addition to the wavelength of the laser and the collection angle of the scatter optics, another factor that significantly contributes to the light scattering intensity of a particle is the refractive index of the suspension medium (sheath, which is essentially water) and the particle itself.


For particles that obey Mie’s predictions, light scattering is largely composed of laser light refraction. When a particle or cell is absent at the interrogation point, the laser’s intersection with the particle or cell produces a characteristic refraction, the “ring of diffraction,” that we are most likely acquainted with on sense-in-air cell sorters. This ring of light spreads outwards from the stream in all directions, in the same plane as the laser beam, and is blocked from entering the forward and side scatter detection paths by the obscuration bars in front of each.


However, the presence of a cell in the laser beam changes the composition of the medium through which the light travels. Laser light now passes through the cytoplasm—which contains protein, lipids, and carbohydrates rather than simply the water it passed through in the absence of a cell—which therefore causes the light to bend differently than it does when it passes through the sheath fluid alone. This refraction of light by a cell is a function of the difference in refractive indices (RI) between the media through which light passes, and it causes laser light to bend in such a way that it passes over the obscuration bar and interacts with the detector, generating scatter signal.


Every type of material has an associated refractive index. When light passes from one material to another—say, from the saline of the sheath fluid to the material of the cell and then back through the sheath fluid again—the amount of light that bends due to this transition is proportional to the difference between the refractive indices between the media. The bigger the difference, the more the scatter.


Why Beads Are Not Good For Calibrating Particles


It is precisely this property of light scatter that makes it a terribly unreliable measurement of cell size.


Two particles or cells of exactly the same size may have different refractive indices, due to their composition (e.g. cytoplasmic proteins), and will therefore generate scatter signals with different intensities. This is also precisely why synthetic beads are also terrible size calibrators.


The refractive index of a typical polystyrene bead (1.59) can be significantly different than a cell’s, resulting in very different scatter signals between a bead and a cell of the same diameter. Given the fact that cells are composed of much more water than a polystyrene bead is and that water’s RI is 1.333, polystyrene beads are going to scatter a lot more light than a typical cell would.


According to studies published in Current Protocols In Cytometry, this situation can be even more severe for microvesicles. The study estimates an average and typical RI of a microvesicle to be approximately 1.39. Given the RI of a polystyrene bead at 1.59 and water at 1.333, a microvesicle’s scatter signal may be one to two orders of magnitude lower than that of the polystyrene bead. Here’s the bottom line—beads are not good at calibrating the scale of a scatter parameter in terms of particle size, regardless of whether the particle is a big one or a small one.


However, all is not lost when it comes to identifying microvesicles. Rather than use scatter as a trigger/threshold parameter to identify these kinds of particles and to measure them, the study suggests that fluorescence is a better choice. Small particles can be labeled with a universal dye that causes all membrane-bound particles in a suspension to fluoresce, allowing discrimination of microvesicles from other particles in the solution. There are nuances and caveats to this kind of labeling, but it can provide a much more robust way to identify membrane-bound microparticles than scatter alone.


Light scatter is a fundamental topic to flow cytometry and understanding how cell size and particle size affect light scatter is critical to performing proper flow cytometry experiments. By understanding how small particles affect forward scatter and side scatter, you can collect better data. Knowing what an obscuration bar is and what a refractive index is, as well as how refractive indices are affected by small particles, will help you design experiments that produce publishable data.


To learn more about getting your flow cytometry data published and to get access to all of our advanced materials including 20 training videos, presentations, workbooks, and private group membership, get on the Flow Cytometry Mastery Class wait list.


Flow Cytometry Mastery Class wait list | Expert Cytometry | Flow Cytometry Training

Saturday, March 19, 2016

What Is A Statistical Analysis T-Test And How To Perform One Using Flow Cytometry Data

Written by Tim Bushnell, Ph.D.


Designing an antibody panel and running samples on a flow cytometer are not the only steps in a flow cytometry experiment.


After you run your experiment, you have to analyze the data. In particular, you need to perform statistical analyses of the data. This is especially true if you’re hoping to publish your data.


Once all the experiments are concluded and the preliminary analysis of the data performed, you must perform statistical analyses on the data to determine if there is significance in the data.


There are several different statistical tests that can be performed depending on the type of data and the comparisons being made. In the case of either making a comparison against a hypothetical mean, or comparison between two populations, the gold standard test is the Student’s T-Test.


What Is A Statistical T-Test?


The T-Test was developed by chemist William Sealy Gosset, who developed the test while working at the Guinness Brewery as a way to monitor the production of their most famous product.


Since he wasn’t allowed to publish his work directly, the paper was published under a pseudonym in the journal, Biometrika.


Before getting into the details of how the T-Test is performed and how the results are interpreted, there are several factors that need to be kept in mind…


The T-Test makes several assumptions about the data:



  • The data is from a Gaussian distribution

  • The data is continuous

  • The sample is a random sample of the population

  • The variance of the populations is equal (If not, there are variations on the theme to address this.) 


There are three major variations on the T-Test:



  • One-sample T-Test – compares the mean of the experimental sample to a hypothetical mean.

  • Unpaired T-Test – compares the mean of the control and experimental samples.

  • Paired T-Test – compares the mean of two samples where the observations in one sample can be related to the observation in the second sample. (For example, the effects of treatment on patients where there is a before treatment and after treatment measurement.) 


The three pieces of information needed to perform a T-Test:



  • The mean of both samples

  • The standard deviation of both samples

  • The number of observations


The T-Test compares the differences between the means of two populations to determine if the null hypothesis should be rejected. At a minimum, to perform the T-Test, one needs the means and standard deviations of both populations, and the number of measurements.


The researcher also needs to set the threshold value, also termed the α. We will compare this threshold to the P-value. If the P-value is greater than the α, there is no significance in the data. However, if the P-value is less than the α, there is significance in the data.


What Is A Null Hypothesis (HO)?


Simply stated, this is a statement about the relationship of the above two populations.  Mathematically, this can be expressed as:


μA = μB


The null hypothesis makes the assumption that our experimental results are from random variation. If, during the statistical analysis, the data is sufficient to show that random variation is not a sufficient explanation for the data, the alternative hypothesis (HA) must be accepted. 


A One-Tailed Versus A Two-Tailed T-Test


A T-Test can either be one-tailed or two-tailed. The above example would be an appropriate null hypothesis for a two-tailed T-Test—that is, when the investigators do not know if the treatment will cause an increase or decrease in the measurement. If the investigators expect the treatment will cause an increase OR a decrease, a one-tailed T-Test is more appropriate. 


How To Run A T-Test


In the following example, the researchers sought to determine if the percentage of CD4+ T-cells in patients who had Irumodic Syndrome was increased after treatment with Byphodine.


The percentage of CD4+ T-cells was measured on PBMCs before treatment and one week after treatment. Considering this information, this is how you would proceed to run a T-Test… 


1. Establish the null hypothesis.


“In patients with Irumodic Syndrome, treatment by Byphodine either decreased or caused no change in the percentage of CD4+ T-cells.”


In this case, since the researchers are not concerned if the treatment causes a decrease in the CD4+ cell, a one-tailed T-test will be performed, and can be written as: 


μA ≥ μB


2. Determine the alternate hypothesis.


“In patients with Irumodic Syndrome, treatment by Byphodine increases the percentage of CD4+ T-cells.” 


3. Establish the threshold.


By convention, the α is typically set to 0.05. This comes from work by R.A. Fisher who stated in his work Statistical Methods for Research Workers (13th Edition): 


The value for which P=0.05, or 1 in 20, is 1.96 or nearly 2; it is convenient to take this point as a limit in judging whether a deviation ought to be considered significant or not. Deviations exceeding twice the standard deviation are thus formally regarded as significant.


There are cases where the threshold can be changed. Increasing the α makes it easier to show significance at the expense of committing a Type I statistical error (false positive). Decreasing the α makes it hard to show significance, and increases the chance of committing a Type II statistical error (false negative). Care must be taken, however, to ensure that the reason for the change is well-documented and spelled out.


For this example, we will set the α to 0.05. 


4. Collect the flow cytometry data.


Following all best practices, with a well-controlled instrument, all appropriate gating and reference controls used to generate the data table below. 


Table of %CD4+ PBMCs

















































Pre-treatment Post-treatment
18.5 26.7
20.1 22.2
25.2 34.5
16.5 23.6
23.3 29.6
22.6 29.1
18.0 40.1
19.3 35.3
17.4 39.5
19.9 31.4

Once this data is entered into our statistical analysis package of choice (we personally use Graphpad Prism), we can generate an appropriate graph: 


t-test statistical analysis of flow cytometry data | Expert Cytometry | t-test formula for data analysis


In the above case, the data is plotted, with the mean and standard deviation plotted.


When the one-way T-Test is calculated, the P-value is 0.0003, which is lower than the threshold. Therefore, the null hypothesis is rejected, and the alternate hypothesis is accepted. As a result, this data supports the conclusion.


The use of the T-Test makes the assumption that the data follows a normal distribution.  If this is not the case, there are non-parametric tests that will allow for the statistical analysis similar to the T-Test. These include the Wilcoxon test and the Mann-Whitney test. In non-parametric tests, the data is ranked according to the value (from lowest to highest), regardless of where the data comes from.


Non-parametric tests test the null hypothesis that the data is distributed at random, with the alternate hypothesis being that the data is not randomly distributed, but one population has larger values than the other.


The Student’s T-Test is an essential tool in the researcher’s toolkit to confirm that the data generated in the course of the investigation supports the hypothesis driving the research. Proper application of the T-Test (and related non-parametric tests) to determine statistical significance in the data will improve confidence in the conclusions of any published work. Following the steps outlined above will allow the researcher to correctly apply the proper statistical tool for their data.


To learn more about getting your flow cytometry data published and to get access to all of our advanced materials including 20 training videos, presentations, workbooks, and private group membership, get on the Flow Cytometry Mastery Class wait list.


Flow Cytometry Mastery Class wait list | Expert Cytometry | Flow Cytometry Training