Thanks so much for these tutorial vids! Absolutely fantastic explanations, very practical and easy to follow! One question though: in the case where 2 or more fluorophores have overlapping emission spectra, but those emissions are being detected in different optical trains (different lasers, each with its own PMT array), would you expect spectral overlap? My thinking is: in very complex polychromatic panels with, say, 15 colors, it would be a bit time-saving to ignore spectral overlap of fluorophores being detected in different optical trains.
Hi, great question! If you are using 2 spatially separated lasers and 2 different PMT arrays, colors that have the exact same emissions but are excited exclusively by different lasers would not have overlap. Examples of this are dyes like PECy7 and APCCy7. PECy7 is excited by blue or yellow lasers (not red) and APCCy7 is excited by a red laser (not blue or yellow). So on our instruments those 2 dyes, while having the same emission profile, require very little to no compensation. This is an ideal case, but most dyes are not so ideal in that they usually are a bit excited by multiple lasers.
@@ucmercedscifstemcellinstru711 Thanks for the super clear explanation! And yes, I've noticed that some fluorophores, such as PerCP, have very wide excitation spectra, and thus you can get some spectral overlap but due to "excitation overlap". For example, if using PerCP-Cy5.5 and Super Bright 702, even though PerCP is most efficiently excited with a 488nm blue laser, the 405nm violet laser will still cause some excitation of PerCP-Cy5.5 and that will bleed over to the Super Bright 702 channel (excited with the blue laser), since both Cy5.5 and SB702 have similar peak emissions.
Thank you for the video. Very helpful. I have a question: I ran a cancer patient tumor infiltrating lymphocytes sample with NK cells BV786 for single color control of my compensation panel, the positive and negative signals did not separate very well. How to define the positive peak for this kind of situation? Thanks!
Hi there, This video is great, really helpful, thanks for it. I was wondering if you would like to make a video explaining voltration and maybe Ab titration?
Thanks a lot for the video, I am trying to comp for 14 colors on LSRII and we usually adjust the voltages during the auto compensation process. For example, if FITC is my first color during compensation I would make a FSA and FITC plot and so on for all the other colors. I would make FITC peak is not overlapping with other peaks and I usually adjust the voltage for colors after FITC if necessary so that they don't overlap and I keep doing this for all the 14 colors. And Finally I hot calculate compensation and some times it fails saying few colors overlap a lot. It is really a pain to compensate for that many colors. Please let me know your thoughts and any suggestions you have. Thanks for your help.
You're starting to get into challenging territory with 14 colors. Much of your success will rely on good experimental design. For example, putting the high density antigens and good antibodies on the dim fluors and the low density antigens on the bright fluors. And try to avoid using co-expressed markers on colors that have a lot of spillover. Fluorofinder is a good resource for building your panel to minimize compensation. If you already have your panel set, I'd say it's best to go through each fluor one by one and make sure that it's brightest in the channel of interest by at least a log if possible. Adjust your voltages to make that happen. For example, if you're using your FITC single stain, make sure it's showing up brightest in the FITC channel. If not, you'll get greater than 100% compensation values and autocomp may give you issues. So you'll need to do 13 plots where you're looking at FITC versus everything else and make sure your signal is brightest for FITC. Then repeat for every other color, adjusting voltages as necessary. This will take some time, especially the first time you set it up. There are some tricks to set up the plots in the global worksheet to make your life easier during this process. I can explain further if that would be helpful.
@@ucmercedscifstemcellinstru711 Hi, thanks very much for this video, it is very helpful! I would also like to learn more about these "tricks" to set up the plots in the global worksheet. Could you please share these with me? Thank you!
@@ucmercedscifstemcellinstru711 thank you so much i am also facing a same problem. In my exoeriment i am using 10 flourophore and right now I am struggling with voltage setting and compensation so I need your more guidance if you possible. It will really help me to proceed further in my research where now i am stuck. My email id: jadhavsuchi@gmail.com
Thank you, this helped me a lot! But I don't get the P2 gate. How is the compensation matrix made? How does the P2 gate information contributes to it? I read that the P2 gate reveals the level of spillover of the fluorophore into other detectors, but we only measure the fluorescence of each fluorophore in one detector, eg FITC in FITC detector? I don't get what information the P2 gate brings..
You can always record all your single stain tubes and samples and then compensate later, even in 3rd party software like FlowJo. I always tend to compensate while I'm running single stains and prior to running samples so I can see my compensated data while I collect it. Then I can make sense of the data while it's running on the cytometer and make sure the staining panel looks correct and as expected.
Did you perform automatic compensation before that or is it possible to set up a panel directly like that ? I suppose that automatic compensation is needed when you also need to titrate antibodies and decide voltages. By the way the a automatic process is the only way to tell the program about negative controls right ?
Awesome video. At 11:10 when compensation is adjusted for APC-eFluor780 vs PE-Cy7 I see that the APC vs PE-Cy7 plot moved too. Could anyone explain why that was happening?
Very informative Thank you so much. I am a green hand of FCM. When I try to do auto compensation, the machine always turns me down with: PMT voltage is not consistent among compensation tubes. Could you tell me the reason for that and how to fix it up?
Why did you leave the 'include separate unstained control' box checked? This would violate the 2nd rule of compensation as the background fluorescence of negative and positive should be the same. Avoid using universal negative is key I believe?
Very informative! I would like to ask a question. How do you distinguish autofluorescence and true positive cells? Thanks so much! And also would you mind provide your email address thus I can ask some questions regarding flow in the future. Thanks so much!
Fun video and demonstration. However, you would be crucified at CYTO for eyeballing compensation on a digital instrument instead of using MFI. Probably for running manual compensation for 7 color would get you a whole lot of hate. I liked that you pointed it out to use the brightest population for (auto) comp, but being mindful that it might be an erroneous auto fluorescent population. I would also recommend increasing the default 5000 events to something higher.
I think you're right about getting crucified by the cytometry purists! But frankly, visually compensating is a practical solution that works well and is important in getting the concepts across to users, rather than a black box solution.
Great Video - few questions: 1) Can you manually compensate single stain controls as you show in the video using beads? 2) Once you manually compensate using single stain controls, does this automatically transfer to flowjo or do you have to recompensate when you upload the files? 3) If you choose to autocompensate using beads, can you recompensate in flowjo to compare matrices or is the compensation locked into how it was done on DIVA software?
1) yes, just make sure you have a positive and negative bead and the fluorochrome is the same as what you stained your sample with. Here's a bead product that binds to antibodies and is often used if cells are not available or you don't have a great positive control stain: www.thermofisher.com/order/catalog/product/01-2222-412) 2) The compensation matrix is saved in the fcs file and flowjo will automatically apply the compensation values you set on the instrument. You can always modify or remove the compensations in flowjo as well. To do this you can double click on the little grid symbol that appears next to your sample name in the flowjo workspace. 3) Yes, in the flowjo compensation editor you can modify the compensations no matter how they were performed on the instrument. Thank you for your questions.
Hi, using comp beads, what voltages would you setup? would you run unstained cells, setup voltages, then run comp beads without touching to voltages even if neg beads are higher than unstained cells? Am I correct here? Thanks
I'm not a fan of setting up voltages based on a negative control. The reason is without a positive control you don't know how bright your positive will be. Thus, you may set it up based on the unstained cells and then you put your sample on and find that the positive signal is off scale (voltage set too high) or does not give optimal separation (voltage set too low). So I always try to set up with single stained cells that are stained with whatever is in the actual panel. Of course, you don't always have that luxury (rare positive events, very little sample to work with, etc). When I have to use beads, I would use the positive beads to set the voltage so the beads are bright and you get good separation between positive and negative beads. Then I would check with unstained cells to see that the background is acceptable with that voltage setting. You're right that beads often have higher background than unstained cells. Also, it's important to know your instrument. For example, I know if I'm setting any PMT voltage outside of the 400-650 range on my LSR II that something is wrong.
many thanks for your reply. I always thought that unstained cells could be used to set approximately the voltages, then of course checking if beads were ok and adjust voltages if needed and re-check unstained cells are not out of scale. How do you check the range of the instrument? CST beads? voltage titrations? Thanks
When a baseline is performed with CST beads (only administrators can do this) the software will determine what it thinks are optimal PMT voltages for each channel based on where the channel is getting good linearity and low CVs. It does a voltage titration and some calculations to determine this. This is the default voltage value that you see for each channel when you make a new experiment. In general that's a good starting point for where to set voltages, but more often than not I find it's on the high end. I know the ranges of my channels more based on a lot of experience with adjusting voltages and seeing what's best with the fluorophores we use most.
![Adjusting compensation: how to handle a "bad" matrix [FlowJo advanced]](http://i.ytimg.com/vi/NEtDeck_V3I/mqdefault.jpg)
![Multicolor flow cytometry - Understanding the basics [WEBINAR]](http://i.ytimg.com/vi/8KQmiQksubY/mqdefault.jpg)
![ARGENTINA vs. CHILE [3-0] | RESUMEN | ELIMINATORIAS SUDAMERICANAS | FECHA 7](http://i.ytimg.com/vi/De3AWmXzhuA/mqdefault.jpg)
Probably the best video online to explain a practical way to carry out compensation in DIVA. Thank you!
That was amazing! more such videos please...
very useful! it helps the understanding of how to apply compensation to the plots. Thank you very much.
This is a great video for anyone new to compensation. Thanks a lot!
With autocompensation, you create a compensation matrix for all your colors. But it can be altered in a flow analysis software like FLowJo.
One of UC Merced's strengths besides energy materials is stem cell research. Their facility is top notch!
Thank you for the video ! This is very informative!!
Very helpful! Thanks so much for sharing.
Man, you are the best! That helped me quite a lot!
I own you a beer... Cheers from Brazil!
Thank you. I've learned a lot from your video.
Wonderful Video, thank you so much for sharing.
Would be good if you could please show us how to set voltages (and the rules you need to follow) when you have a multicolour experiment
He does in this video: ruclips.net/video/NlZbMKPWWMs/видео.html
Please🤝
Thanks so much for these tutorial vids! Absolutely fantastic explanations, very practical and easy to follow! One question though: in the case where 2 or more fluorophores have overlapping emission spectra, but those emissions are being detected in different optical trains (different lasers, each with its own PMT array), would you expect spectral overlap? My thinking is: in very complex polychromatic panels with, say, 15 colors, it would be a bit time-saving to ignore spectral overlap of fluorophores being detected in different optical trains.
Hi, great question! If you are using 2 spatially separated lasers and 2 different PMT arrays, colors that have the exact same emissions but are excited exclusively by different lasers would not have overlap. Examples of this are dyes like PECy7 and APCCy7. PECy7 is excited by blue or yellow lasers (not red) and APCCy7 is excited by a red laser (not blue or yellow). So on our instruments those 2 dyes, while having the same emission profile, require very little to no compensation. This is an ideal case, but most dyes are not so ideal in that they usually are a bit excited by multiple lasers.
@@ucmercedscifstemcellinstru711 Thanks for the super clear explanation! And yes, I've noticed that some fluorophores, such as PerCP, have very wide excitation spectra, and thus you can get some spectral overlap but due to "excitation overlap". For example, if using PerCP-Cy5.5 and Super Bright 702, even though PerCP is most efficiently excited with a 488nm blue laser, the 405nm violet laser will still cause some excitation of PerCP-Cy5.5 and that will bleed over to the Super Bright 702 channel (excited with the blue laser), since both Cy5.5 and SB702 have similar peak emissions.
Thank you for the video. Very helpful. I have a question: I ran a cancer patient tumor infiltrating lymphocytes sample with NK cells BV786 for single color control of my compensation panel, the positive and negative signals did not separate very well. How to define the positive peak for this kind of situation? Thanks!
Helpful to me
Hi there,
This video is great, really helpful, thanks for it. I was wondering if you would like to make a video explaining voltration and maybe Ab titration?
Can you please send the link for the basic LSR set up that you mentioned? I couldn’t find it. Thank you!
Thanks a lot for the video, I am trying to comp for 14 colors on LSRII and we usually adjust the voltages during the auto compensation process. For example, if FITC is my first color during compensation I would make a FSA and FITC plot and so on for all the other colors. I would make FITC peak is not overlapping with other peaks and I usually adjust the voltage for colors after FITC if necessary so that they don't overlap and I keep doing this for all the 14 colors. And Finally I hot calculate compensation and some times it fails saying few colors overlap a lot. It is really a pain to compensate for that many colors. Please let me know your thoughts and any suggestions you have. Thanks for your help.
You're starting to get into challenging territory with 14 colors. Much of your success will rely on good experimental design. For example, putting the high density antigens and good antibodies on the dim fluors and the low density antigens on the bright fluors. And try to avoid using co-expressed markers on colors that have a lot of spillover. Fluorofinder is a good resource for building your panel to minimize compensation. If you already have your panel set, I'd say it's best to go through each fluor one by one and make sure that it's brightest in the channel of interest by at least a log if possible. Adjust your voltages to make that happen. For example, if you're using your FITC single stain, make sure it's showing up brightest in the FITC channel. If not, you'll get greater than 100% compensation values and autocomp may give you issues. So you'll need to do 13 plots where you're looking at FITC versus everything else and make sure your signal is brightest for FITC. Then repeat for every other color, adjusting voltages as necessary. This will take some time, especially the first time you set it up. There are some tricks to set up the plots in the global worksheet to make your life easier during this process. I can explain further if that would be helpful.
Great!, Can you please elaborate on that?, we can also connect via email Praveen98481@gmail.com
@@ucmercedscifstemcellinstru711 Hi, thanks very much for this video, it is very helpful! I would also like to learn more about these "tricks" to set up the plots in the global worksheet. Could you please share these with me? Thank you!
@@ucmercedscifstemcellinstru711 thank you so much i am also facing a same problem. In my exoeriment i am using 10 flourophore and right now I am struggling with voltage setting and compensation so I need your more guidance if you possible. It will really help me to proceed further in my research where now i am stuck. My email id: jadhavsuchi@gmail.com
Thank you, this helped me a lot!
But I don't get the P2 gate. How is the compensation matrix made? How does the P2 gate information contributes to it? I read that the P2 gate reveals the level of spillover of the fluorophore into other detectors, but we only measure the fluorescence of each fluorophore in one detector, eg FITC in FITC detector? I don't get what information the P2 gate brings..
Do you have to do the compensation while running the single stainin tube, or you can do it after ? Thanks a lot for your video.
You can always record all your single stain tubes and samples and then compensate later, even in 3rd party software like FlowJo. I always tend to compensate while I'm running single stains and prior to running samples so I can see my compensated data while I collect it. Then I can make sense of the data while it's running on the cytometer and make sure the staining panel looks correct and as expected.
Did you perform automatic compensation before that or is it possible to set up a panel directly like that ? I suppose that automatic compensation is needed when you also need to titrate antibodies and decide voltages. By the way the a automatic process is the only way to tell the program about negative controls right ?
Awesome video. At 11:10 when compensation is adjusted for APC-eFluor780 vs PE-Cy7 I see that the APC vs PE-Cy7 plot moved too. Could anyone explain why that was happening?
Very informative Thank you so much. I am a green hand of FCM. When I try to do auto compensation, the machine always turns me down with: PMT voltage is not consistent among compensation tubes. Could you tell me the reason for that and how to fix it up?
Why did you leave the 'include separate unstained control' box checked? This would violate the 2nd rule of compensation as the background fluorescence of negative and positive should be the same. Avoid using universal negative is key I believe?
Thanks!!!!!
Thank you. Do you have any virtual training courses about flow cytometry?
Very informative! I would like to ask a question. How do you distinguish autofluorescence and true positive cells? Thanks so much! And also would you mind provide your email address thus I can ask some questions regarding flow in the future. Thanks so much!
Fun video and demonstration. However, you would be crucified at CYTO for eyeballing compensation on a digital instrument instead of using MFI. Probably for running manual compensation for 7 color would get you a whole lot of hate. I liked that you pointed it out to use the brightest population for (auto) comp, but being mindful that it might be an erroneous auto fluorescent population. I would also recommend increasing the default 5000 events to something higher.
I think you're right about getting crucified by the cytometry purists! But frankly, visually compensating is a practical solution that works well and is important in getting the concepts across to users, rather than a black box solution.
Great Video - few questions:
1) Can you manually compensate single stain controls as you show in the video using beads?
2) Once you manually compensate using single stain controls, does this automatically transfer to flowjo or do you have to recompensate when you upload the files?
3) If you choose to autocompensate using beads, can you recompensate in flowjo to compare matrices or is the compensation locked into how it was done on DIVA software?
1) yes, just make sure you have a positive and negative bead and the fluorochrome is the same as what you stained your sample with. Here's a bead product that binds to antibodies and is often used if cells are not available or you don't have a great positive control stain: www.thermofisher.com/order/catalog/product/01-2222-412) 2) The compensation matrix is saved in the fcs file and flowjo will automatically apply the compensation values you set on the instrument. You can always modify or remove the compensations in flowjo as well. To do this you can double click on the little grid symbol that appears next to your sample name in the flowjo workspace. 3) Yes, in the flowjo compensation editor you can modify the compensations no matter how they were performed on the instrument. Thank you for your questions.
Hi, using comp beads, what voltages would you setup? would you run unstained cells, setup voltages, then run comp beads without touching to voltages even if neg beads are higher than unstained cells? Am I correct here? Thanks
I'm not a fan of setting up voltages based on a negative control. The reason is without a positive control you don't know how bright your positive will be. Thus, you may set it up based on the unstained cells and then you put your sample on and find that the positive signal is off scale (voltage set too high) or does not give optimal separation (voltage set too low). So I always try to set up with single stained cells that are stained with whatever is in the actual panel. Of course, you don't always have that luxury (rare positive events, very little sample to work with, etc). When I have to use beads, I would use the positive beads to set the voltage so the beads are bright and you get good separation between positive and negative beads. Then I would check with unstained cells to see that the background is acceptable with that voltage setting. You're right that beads often have higher background than unstained cells. Also, it's important to know your instrument. For example, I know if I'm setting any PMT voltage outside of the 400-650 range on my LSR II that something is wrong.
many thanks for your reply. I always thought that unstained cells could be used to set approximately the voltages, then of course checking if beads were ok and adjust voltages if needed and re-check unstained cells are not out of scale. How do you check the range of the instrument? CST beads? voltage titrations? Thanks
When a baseline is performed with CST beads (only administrators can do this) the software will determine what it thinks are optimal PMT voltages for each channel based on where the channel is getting good linearity and low CVs. It does a voltage titration and some calculations to determine this. This is the default voltage value that you see for each channel when you make a new experiment. In general that's a good starting point for where to set voltages, but more often than not I find it's on the high end. I know the ranges of my channels more based on a lot of experience with adjusting voltages and seeing what's best with the fluorophores we use most.
六六六！