How to design high-parameter experiments to achieve more information from your spectral flow cytometry

In a webinar now available on demand, Dr. Greg Veltri, Director of Biological Sciences and Business Development at Sony Biotechnology, reviews the basic principles of spectral flow cytometry and explains how users of all levels can easily design panels for high-parameter experiments and achieve a greater depth of information.

This webinar highlights how spectrally adjacent fluorochromes, fluorescent proteins, and fluorochromes excited by multiple lasers can be easily incorporated into a panel when using Sony spectral systems. Veltri also introduces the uses and applications of the ID7000 spectral analyzer, the newest addition to the Sony portfolio, and answers questions a range of interesting questions from scientists about the technology and its application.

Read on for highlights from the Q&A discussion at the end of the live webinar or register to watch the full webinar on demand:

Q: The first question is, if I'm using colors from the library, can I adjust voltages and will the library still work?

GV: Yes, you can adjust the voltages. This is different from using a cytometer that's not spectral. What the single positive control is looking at is the full curve of the spectrum. So, it doesn't look at the intensity, the algorithm is unmixing the curve.

Q: In traditional flow cytometry, compensation controls must be at least as bright as the samples. Is that also true of single stain controls for mixing? And if so, does that limit the utility of the saved library of spectra?

GV: This question is related to the first question. It doesn't work like traditional flow cytometry in that you have to worry about the brightness of the single positive control from mixing. Again, the algorithm is just looking at the curve and then it's separating the curve. This separation enables you to identify the colors, not where the negative is sitting.

Q: How many autofluorescence populations can be used for subtracting background from multicolor samples?

GV: The system is set up to do approximately about eight autofluorescent populations. Now, the way that we do autofluorescence on this instrument is that subtraction is really defining it as a color. So, just think of it as when we do an autofluorescence, we're defining a color for each of the different types of reflective and refractive light that we're getting off the cytometer.

Q: Is there still a need to use FMOs (fluorescence minus one controls)?

GV: We're not saying that you shouldn't use FMOs. But what the system allows you to do using a spectral curve is see what the fluorochrome is of that population based on the gate that you're using. If you need to know the staining capability or the interaction that's occurring on the cell surface or intracellular, then you still need to use an FMO.

Q: How well does the sample mix in 384-well plates?

GV: We've designed it so that it does a good job. If you're worried about the coefficient of variation (CV) of the mixing, what we've seen is that the movement of the population is minimal in terms of CV. We see a constant mixing occurring while firing the sample.

Q: If you use fluorochromes with very similar properties, do you advise to ensure they identify markers that are not co-expressed?

GV: It depends on how similar they are. If you have very similar fluorochromes, for example, GFP and FITC or Alexa Fluor 488 and FITC, unmixing could get a little trickier. This would apply to any spectral flow cytometer. So, yes, I would probably advise to not do a co-express.

Q: What is the rule of thumb I can use for selecting two fluorochromes that closely space spectra?

GV: The rule of thumb would be that you want to keep your fluorochromes at least about 10 nanometers in wavelength from each other. We have a highly overdetermined system on here, so we're able to see fluorochromes that are very close to each other. However, once it gets any closer than that it starts to become a little bit more difficult. It is also worth bearing in mind the least square method or weighted least square method and the unmixing algorithms. You do have to worry if you have a single peak fluorochrome that you're not overshadowing one population versus the other when it's that close.

Q: Can the system/library learn new spectra not in the library?

GV: Yes, in the new system you're able to define new names, and where they're excited and where they're being admitted. On our system, when you define it, you're also able to place a virtual filter on there. This means that if you wanted to see it based on a bandpass filter that you use on a normal cytometer, then you would able to define it there also. Just be aware that the unmixing doesn't use the virtual filters.

Q: What is the minimal cell particle size detected by the system?

GV: We have done the Spirotech small particle self-test, which goes to about 0.2 microns or 200 nanometers and we have no problem resolving the noise from that. We haven’t done anything lower than that yet.

Q: Are there additional laser options for the ID7000 besides the seven lasers you showed during the presentation?

GV: No, the seven lasers that we have on the system are the only ones we offer. The reason being, we've optimized the optical deck for those seven lasers so that you're able to see dim and bright populations with no problem.

Q: Apart from the number of lasers and channels, what's the technical difference between the ID7000 and the previous spectral analyzer from Sony?

GV: The technical difference is that on the ID7000, we went to a grading system. On the SP and the SA, we use a prism system. The reason we changed over is that when we first developed the technology back in 2010 or 2008, grading systems weren't as refined and sensitive as they are now. So, as we developed this system, we just improved upon the optics as the technology developed. This is the only key difference between the two systems.

Q: A follow-up question on the ID7000. Are the detectors PMTs or APDs or something else?

GV: They are PMTs, but they are customized PMTs. We have a special relationship with our vendor so we've worked it out so that they're sensitive across the whole spectrum.

Q: This question on performing a mixing: Do fluorochromes show different spectra depending on what they bind? This is referring to beads or cells performing the same in the system.

GV: That's an excellent question. I know on a normal cytometer that's not doing spectral that you actually do see a difference. Now, what we've done is we've looked at what the spectral curve looks like on a bead and a cell and we haven't seen any differences between the two. I think it'd probably be interesting to go further with different types of buffers, and fixatives and things like that, to look at it more. But our preliminary work has shown that we don't see any difference in the spectral curve itself occurring.

Watch this webinar on demand >>

SelectScience runs 3-4 webinars a month across various scientific topics, discover more of our upcoming webinars>>