quantitative confocal microscopy
Locked
 
|
POUVREAU SANDRINE |
quantitative confocal microscopy
|
Hello.
I had this discussion with several colleagues (biologist like me), and did some research on my own, but I figurate the best would be to submit the question to this list. Here’s the point: we are doing quantitative measurement with confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There are 3 parameters of the PMT that can be configured: detector gain, amplifier offset, amplifier gain. The only parameter I adjust to improve the signal is the detector gain. I keep the amplifier gain at 1 as I read in several papers that increasing it will not improve the signal over noise ratio (they also say that it is bad for several reasons that I can not summarize here). Is that correct? I put the offset usually at zero. I saw that a change in offset can affect the calcium signal. In any case, I keep the same offset in a series of records that I whish to compare. Any comment on this? My guess is that the same should be done with the amplifier gain. Ah, a last point: the offset should be tuned so no pixel in the background has the value of zero right? Does anybody have any clarification for these questions? Thanks for your help. Regards Sandrine |
|
|
Julian Smith III |
Re: quantitative confocal microscopy
|
|
I'm running an Olympus Fluoview, but I think the answers are the same:
Yes--detector gain to adjust max signal--no saturated pixels where you plan to measure. Amp gain at 1 Offset at a few to no zero pixels in background (again, not where you plan to measure). Don't change the settings across specimens, if you plan to compare them to each other. Julian Sandrine Pouvreau wrote: > Hello. > I had this discussion with several colleagues (biologist like me), and did some > research on my own, but I figurate the best would be to submit the question > to this list. Here’s the point: we are doing quantitative measurement with > confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There > are 3 parameters of the PMT that can be configured: detector gain, amplifier > offset, amplifier gain. The only parameter I adjust to improve the signal is the > detector gain. I keep the amplifier gain at 1 as I read in several papers that > increasing it will not improve the signal over noise ratio (they also say that it > is bad for several reasons that I can not summarize here). Is that correct? I > put the offset usually at zero. I saw that a change in offset can affect the > calcium signal. In any case, I keep the same offset in a series of records that > I whish to compare. Any comment on this? My guess is that the same should > be done with the amplifier gain. > Ah, a last point: the offset should be tuned so no pixel in the background has > the value of zero right? > Does anybody have any clarification for these questions? Thanks for your help. > > Regards > Sandrine > > -- Julian P.S. Smith III Director, Winthrop Microscopy Facility Dept. of Biology Winthrop University 520 Cherry Rd. Rock Hill, SC 29733 803-323-2111 x6427 (vox) 803-323-3448 (fax) 803-524-2347 (cell) Research Website www.birdnest.org/smithj Personal Website www.rociada-east.net |
|
|
John Oreopoulos |
Re: quantitative confocal microscopy
|
|
In reply to this post by POUVREAU SANDRINE
Hi Sandrine, your question is sure to draw answers as it has been asked several times before. Quantitative fluorescence microscopy is not trivial, but can be somewhat reliable if you try to carefully control as many experimental parameters as possible. Other people on this list server can give you more details, but as a quick reference, try reading Jim Pawley's "39 steps":
http://labs.pbrc.edu/cellbiology/documents/39steps.pdf As for detector gain and offset settings, Alan Hibbs' book has a nice section that outlines a good method for determining the best settings: http://www.amazon.com/Confocal-Microscopy-Biologists-Alan-Hibbs/dp/0306484684 All the best, John Oreopoulos On 2010-07-28, at 9:05 AM, Sandrine Pouvreau wrote: > Hello. > I had this discussion with several colleagues (biologist like me), and did some > research on my own, but I figurate the best would be to submit the question > to this list. Here’s the point: we are doing quantitative measurement with > confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There > are 3 parameters of the PMT that can be configured: detector gain, amplifier > offset, amplifier gain. The only parameter I adjust to improve the signal is the > detector gain. I keep the amplifier gain at 1 as I read in several papers that > increasing it will not improve the signal over noise ratio (they also say that it > is bad for several reasons that I can not summarize here). Is that correct? I > put the offset usually at zero. I saw that a change in offset can affect the > calcium signal. In any case, I keep the same offset in a series of records that > I whish to compare. Any comment on this? My guess is that the same should > be done with the amplifier gain. > Ah, a last point: the offset should be tuned so no pixel in the background has > the value of zero right? > Does anybody have any clarification for these questions? Thanks for your help. > > Regards > Sandrine |
|
|
Glen MacDonald-2 |
Re: quantitative confocal microscopy
|
|
This may be of interest to the community and I apologize anyone is offended by the slightly commercial nature.
Alan Hibbs had shipped copies of his book to me in anticipation of a teaching engagement in North America. Unfortunately, he unexpectedly passed away before he could visit. I've still got a dozen copies for US$125, shipping by US mail is about $12. Proceeds are sent to his family. If interested, please reply directly to my home email, [hidden email]. Regards, Glen Glen MacDonald Core for Communication Research Virginia Merrill Bloedel Hearing Research Center Box 357923 University of Washington Seattle, WA 98195-7923 USA (206) 616-4156 [hidden email] On Jul 28, 2010, at 6:25 AM, John Oreopoulos wrote: > Hi Sandrine, your question is sure to draw answers as it has been asked several times before. Quantitative fluorescence microscopy is not trivial, but can be somewhat reliable if you try to carefully control as many experimental parameters as possible. Other people on this list server can give you more details, but as a quick reference, try reading Jim Pawley's "39 steps": > http://labs.pbrc.edu/cellbiology/documents/39steps.pdf > > As for detector gain and offset settings, Alan Hibbs' book has a nice section that outlines a good method for determining the best settings: > http://www.amazon.com/Confocal-Microscopy-Biologists-Alan-Hibbs/dp/0306484684 > > All the best, > > John Oreopoulos > > > > On 2010-07-28, at 9:05 AM, Sandrine Pouvreau wrote: > >> Hello. >> I had this discussion with several colleagues (biologist like me), and did some >> research on my own, but I figurate the best would be to submit the question >> to this list. Here’s the point: we are doing quantitative measurement with >> confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There >> are 3 parameters of the PMT that can be configured: detector gain, amplifier >> offset, amplifier gain. The only parameter I adjust to improve the signal is the >> detector gain. I keep the amplifier gain at 1 as I read in several papers that >> increasing it will not improve the signal over noise ratio (they also say that it >> is bad for several reasons that I can not summarize here). Is that correct? I >> put the offset usually at zero. I saw that a change in offset can affect the >> calcium signal. In any case, I keep the same offset in a series of records that >> I whish to compare. Any comment on this? My guess is that the same should >> be done with the amplifier gain. >> Ah, a last point: the offset should be tuned so no pixel in the background has >> the value of zero right? >> Does anybody have any clarification for these questions? Thanks for your help. >> >> Regards >> Sandrine |
|
|
Boswell, Carl A - (cboswell) |
Re: quantitative confocal microscopy
|
|
Do you take Visa?
C Carl A. Boswell, Ph.D. Molecular and Cellular Biology University of Arizona 520-954-7053 FAX 520-621-3709 ----- Original Message ----- From: "Glen MacDonald" <[hidden email]> To: <[hidden email]> Sent: Wednesday, July 28, 2010 9:42 AM Subject: Re: quantitative confocal microscopy This may be of interest to the community and I apologize anyone is offended by the slightly commercial nature. Alan Hibbs had shipped copies of his book to me in anticipation of a teaching engagement in North America. Unfortunately, he unexpectedly passed away before he could visit. I've still got a dozen copies for US$125, shipping by US mail is about $12. Proceeds are sent to his family. If interested, please reply directly to my home email, [hidden email]. Regards, Glen Glen MacDonald Core for Communication Research Virginia Merrill Bloedel Hearing Research Center Box 357923 University of Washington Seattle, WA 98195-7923 USA (206) 616-4156 [hidden email] On Jul 28, 2010, at 6:25 AM, John Oreopoulos wrote: > Hi Sandrine, your question is sure to draw answers as it has been asked > several times before. Quantitative fluorescence microscopy is not trivial, > but can be somewhat reliable if you try to carefully control as many > experimental parameters as possible. Other people on this list server can > give you more details, but as a quick reference, try reading Jim Pawley's > "39 steps": > http://labs.pbrc.edu/cellbiology/documents/39steps.pdf > > As for detector gain and offset settings, Alan Hibbs' book has a nice > section that outlines a good method for determining the best settings: > http://www.amazon.com/Confocal-Microscopy-Biologists-Alan-Hibbs/dp/0306484684 > > All the best, > > John Oreopoulos > > > > On 2010-07-28, at 9:05 AM, Sandrine Pouvreau wrote: > >> Hello. >> I had this discussion with several colleagues (biologist like me), and >> did some >> research on my own, but I figurate the best would be to submit the >> question >> to this list. Here’s the point: we are doing quantitative measurement >> with >> confocal microscopy (calcium measurement) using a Zeiss LSM exciter. >> There >> are 3 parameters of the PMT that can be configured: detector gain, >> amplifier >> offset, amplifier gain. The only parameter I adjust to improve the signal >> is the >> detector gain. I keep the amplifier gain at 1 as I read in several papers >> that >> increasing it will not improve the signal over noise ratio (they also say >> that it >> is bad for several reasons that I can not summarize here). Is that >> correct? I >> put the offset usually at zero. I saw that a change in offset can affect >> the >> calcium signal. In any case, I keep the same offset in a series of >> records that >> I whish to compare. Any comment on this? My guess is that the same should >> be done with the amplifier gain. >> Ah, a last point: the offset should be tuned so no pixel in the >> background has >> the value of zero right? >> Does anybody have any clarification for these questions? Thanks for your >> help. >> >> Regards >> Sandrine |
|
|
leoncio vergara |
Re: quantitative confocal microscopy
|
|
In reply to this post by Glen MacDonald-2
I just sent you a reply to the e-mail listed. I am interested
-----Original Message----- From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Glen MacDonald Sent: Wednesday, July 28, 2010 11:43 AM To: [hidden email] Subject: Re: quantitative confocal microscopy This may be of interest to the community and I apologize anyone is offended by the slightly commercial nature. Alan Hibbs had shipped copies of his book to me in anticipation of a teaching engagement in North America. Unfortunately, he unexpectedly passed away before he could visit. I've still got a dozen copies for US$125, shipping by US mail is about $12. Proceeds are sent to his family. If interested, please reply directly to my home email, [hidden email]. Regards, Glen Glen MacDonald Core for Communication Research Virginia Merrill Bloedel Hearing Research Center Box 357923 University of Washington Seattle, WA 98195-7923 USA (206) 616-4156 [hidden email] On Jul 28, 2010, at 6:25 AM, John Oreopoulos wrote: > Hi Sandrine, your question is sure to draw answers as it has been asked several times before. Quantitative fluorescence microscopy is not trivial, but can be somewhat reliable if you try to carefully control as many experimental parameters as possible. Other people on this list server can give you more details, but as a quick reference, try reading Jim Pawley's "39 steps": > http://labs.pbrc.edu/cellbiology/documents/39steps.pdf > > As for detector gain and offset settings, Alan Hibbs' book has a nice section that outlines a good method for determining the best settings: > http://www.amazon.com/Confocal-Microscopy-Biologists-Alan-Hibbs/dp/030 > 6484684 > > All the best, > > John Oreopoulos > > > > On 2010-07-28, at 9:05 AM, Sandrine Pouvreau wrote: > >> Hello. >> I had this discussion with several colleagues (biologist like me), >> and did some research on my own, but I figurate the best would be to >> submit the question to this list. Here's the point: we are doing >> quantitative measurement with confocal microscopy (calcium >> measurement) using a Zeiss LSM exciter. There are 3 parameters of the >> PMT that can be configured: detector gain, amplifier offset, >> amplifier gain. The only parameter I adjust to improve the signal is >> the detector gain. I keep the amplifier gain at 1 as I read in >> several papers that increasing it will not improve the signal over >> noise ratio (they also say that it is bad for several reasons that I >> can not summarize here). Is that correct? I put the offset usually at >> zero. I saw that a change in offset can affect the calcium signal. In >> any case, I keep the same offset in a series of records that I whish to compare. Any comment on this? My guess is that the same should be done with the amplifier gain. >> Ah, a last point: the offset should be tuned so no pixel in the >> background has the value of zero right? >> Does anybody have any clarification for these questions? Thanks for your help. >> >> Regards >> Sandrine |
|
|
Glen MacDonald-2 |
Re: quantitative confocal microscopy
|
|
Thank you Leonico.
Sorry Carl, no VISA. Don't feed the banks, it only encourages them. Glen On Jul 28, 2010, at 9:51 AM, Vergara, Leoncio A. wrote: > I just sent you a reply to the e-mail listed. I am interested > > -----Original Message----- > From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Glen MacDonald > Sent: Wednesday, July 28, 2010 11:43 AM > To: [hidden email] > Subject: Re: quantitative confocal microscopy > > This may be of interest to the community and I apologize anyone is offended by the slightly commercial nature. > > Alan Hibbs had shipped copies of his book to me in anticipation of a teaching engagement in North America. Unfortunately, he unexpectedly passed away before he could visit. I've still got a dozen copies for US$125, shipping by US mail is about $12. Proceeds are sent to his family. If interested, please reply directly to my home email, [hidden email]. > > > Regards, > Glen > > Glen MacDonald > Core for Communication Research > Virginia Merrill Bloedel Hearing Research Center Box 357923 University of Washington Seattle, WA 98195-7923 USA > (206) 616-4156 > [hidden email] > > > > > > > > > On Jul 28, 2010, at 6:25 AM, John Oreopoulos wrote: > >> Hi Sandrine, your question is sure to draw answers as it has been asked several times before. Quantitative fluorescence microscopy is not trivial, but can be somewhat reliable if you try to carefully control as many experimental parameters as possible. Other people on this list server can give you more details, but as a quick reference, try reading Jim Pawley's "39 steps": >> http://labs.pbrc.edu/cellbiology/documents/39steps.pdf >> >> As for detector gain and offset settings, Alan Hibbs' book has a nice section that outlines a good method for determining the best settings: >> http://www.amazon.com/Confocal-Microscopy-Biologists-Alan-Hibbs/dp/030 >> 6484684 >> >> All the best, >> >> John Oreopoulos >> >> >> >> On 2010-07-28, at 9:05 AM, Sandrine Pouvreau wrote: >> >>> Hello. >>> I had this discussion with several colleagues (biologist like me), >>> and did some research on my own, but I figurate the best would be to >>> submit the question to this list. Here's the point: we are doing >>> quantitative measurement with confocal microscopy (calcium >>> measurement) using a Zeiss LSM exciter. There are 3 parameters of the >>> PMT that can be configured: detector gain, amplifier offset, >>> amplifier gain. The only parameter I adjust to improve the signal is >>> the detector gain. I keep the amplifier gain at 1 as I read in >>> several papers that increasing it will not improve the signal over >>> noise ratio (they also say that it is bad for several reasons that I >>> can not summarize here). Is that correct? I put the offset usually at >>> zero. I saw that a change in offset can affect the calcium signal. In >>> any case, I keep the same offset in a series of records that I whish to compare. Any comment on this? My guess is that the same should be done with the amplifier gain. >>> Ah, a last point: the offset should be tuned so no pixel in the >>> background has the value of zero right? >>> Does anybody have any clarification for these questions? Thanks for your help. >>> >>> Regards >>> Sandrine |
|
|
Martin Wessendorf-2 |
Re: quantitative confocal microscopy--photon-counting
|
|
In reply to this post by POUVREAU SANDRINE
This is a bit of a digression for this thread, but I wonder if anyone
can tell me why photon-counting isn't available on more confocals. It seems as if it would be the method of choice for quantitative experiments--assuming that you're avoiding pile-ups, the counts for a given wavelength will be directly proportion to the fluorescence. But perhaps I'm missing something. Thanks-- Martin -- Martin Wessendorf, Ph.D. office: (612) 626-0145 Assoc Prof, Dept Neuroscience lab: (612) 624-2991 University of Minnesota Preferred FAX: (612) 624-8118 6-145 Jackson Hall, 321 Church St. SE Dept Fax: (612) 626-5009 Minneapolis, MN 55455 e-mail: [hidden email] |
|
|
Mark Cannell |
Re: quantitative confocal microscopy--photon-counting
|
|
Hi Martin
The primary problem is speed (IMHO). The counters start to show pile up at ~1MHz. To get (say) 10 photons per pixel this implies a pixel scan rate <100 KHz which for a 1MHz image means >10s /frame. Even at this low rate the image is still very poor (10 phot in brightest pixel). Basically the low speed, loss of linearity at high photon count rates and extra complexity doesn't make it worthwhile. On the other hand, analogue is fast and linear and is only slightly lower signal to noise. Cheers Mark > This is a bit of a digression for this thread, but I wonder if anyone > can tell me why photon-counting isn't available on more confocals. It > seems as if it would be the method of choice for quantitative > experiments--assuming that you're avoiding pile-ups, the counts for a > given wavelength will be directly proportion to the fluorescence. But > perhaps I'm missing something. > > Thanks-- > > Martin |
|
|
George McNamara |
Re: quantitative confocal microscopy
|
|
In reply to this post by POUVREAU SANDRINE
Hi Sandrine,
Detector gain (PMT voltage) - ideally would like to be a modest value, such as 600 or 700. Amplifier offset - adjust so that no laser light results in all values being above zero. Note that if you change scan speed and/or detector gain, the offset value(s) may change. When using 12-bit data mode, I typically aim for an offset of ~100, but am happy with anything less than 200 (at moderate gain). Amplifier gain - you are correct. Do not change from 1.0. It is just does post acquisition arithmetic. Other items: You did not mention what fluorophore(s) or fluorescent protein you are using. Or whether you are using the best filter or other wavelength setting(s) for these. Depending on the cell's autofluorescence, this may need optimization. You did not mention laser power: * Make sure the laser output is stable! Both within a frame and over experiment time. A starting point is to turn on the transmitted light detector (which as a bonus gets you the cell morphology) and see that you are getting stable intensity. I recommend no polarizing elements in the light path. If you keep your confocal system off at night, may take many hours for the laser output to stabilize (I have never been sure whether it is the lasers or the AOTF intensity controller as being the guilty party or parties). * If you are using an argon visible laser on an LSM510, run the laser at 6.1 Amps and leave the laser power dialog open (it might fluctuate on its own between 5.9 and 6.3 Amps - that is ok). On our system I've seen the power change to crazy values, like 7.9 Amps, apparently on its own (also on hitting re-use, even though I know the saved setting was not that value). On the LSM710 Argon laser's remote, turn the knob so it is in the middle of the green light range. * Laser power to the specimen ... lower is better! For example, cardiomyocytes die a spectacular death after a few seconds of 488 nm excitation when loaded with Fluo-4 and imaged at high power. If you are using an -AM conjugate, more is not better. Besides the possibility of turning your Calcium ion reporter into a chelator, the -AM leaving group is formaldehyde. Could be worse: GFP maturation produces one molecule of H2O2 (so two for CaMeleon), and red fluorescent protein maturation may produce two H2O2's. No pixel in the background at zero ... correct, this is ideal. However, at high gain (say 900 volts) and an offset such that no laser light results in average intensity value around 100 (12-bit mode), when you are imaging with the laser and specimen, if you acquire enough images, eventually statistics dictate that some outlier will reach zero. So what! When you use the "no light" image (or an average of a whole bunch of them) in background subtraction, lots of values outside the cells will hit zero. A more important item is getting the detector gain and laser power optimized so you can use most of the dynamic range of the confocal to get the data you need. This requires guesswork or using the first specimen to get maximum intensity (typically, saturated Calcium ion with ionophore). Best wishes, George At 09:05 AM 7/28/2010, you wrote: >Hello. >I had this discussion with several colleagues (biologist like me), >and did some >research on my own, but I figurate the best would be to submit the question >to this list. Here's the point: we are doing quantitative measurement with >confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There >are 3 parameters of the PMT that can be configured: detector gain, amplifier >offset, amplifier gain. The only parameter I adjust to improve the >signal is the >detector gain. I keep the amplifier gain at 1 as I read in several >papers that >increasing it will not improve the signal over noise ratio (they >also say that it >is bad for several reasons that I can not summarize here). Is that correct? I >put the offset usually at zero. I saw that a change in offset can affect the >calcium signal. In any case, I keep the same offset in a series of >records that >I whish to compare. Any comment on this? My guess is that the same should >be done with the amplifier gain. >Ah, a last point: the offset should be tuned so no pixel in the >background has >the value of zero right? >Does anybody have any clarification for these questions? Thanks for your help. > >Regards >Sandrine George McNamara, Ph.D. Image Core Manager Analytical Imaging Core Facility University of Miami, Miller School of Medicine Miami, FL 33136 [hidden email] [hidden email] 305-243-8436 office http://www.sylvester.org/AICF (Analytical Imaging Core Facility) http://www.sylvester.org/AICF/pubspectra.zip (the entire 2000+ spectra .xlsx file is in the zip file) http://home.earthlink.net/~geomcnamara |
|
|
Nuno Moreno |
Re: quantitative confocal microscopy
|
|
Just one comment related to this. When running in light mode, the laser would adjust current in order to have constant No. of photons output. This means that a few % variation in current might not be translated into equivalent light variation.
All the best, Nuno Moreno, PhD Head of Equipment Management Instituto Gulbenkian de Ciência Fundação Calouste Gulbenkian Tel . +351 4464538 Fax. +351 4407970 On Jul 29, 2010, at 2:55 AM, George McNamara wrote: > Hi Sandrine, > > Detector gain (PMT voltage) - ideally would like to be a modest value, such as 600 or 700. > > Amplifier offset - adjust so that no laser light results in all values being above zero. Note that if you change scan speed and/or detector gain, the offset value(s) may change. When using 12-bit data mode, I typically aim for an offset of ~100, but am happy with anything less than 200 (at moderate gain). > > Amplifier gain - you are correct. Do not change from 1.0. It is just does post acquisition arithmetic. > > Other items: > > You did not mention what fluorophore(s) or fluorescent protein you are using. Or whether you are using the best filter or other wavelength setting(s) for these. Depending on the cell's autofluorescence, this may need optimization. > > You did not mention laser power: > > * Make sure the laser output is stable! Both within a frame and over experiment time. A starting point is to turn on the transmitted light detector (which as a bonus gets you the cell morphology) and see that you are getting stable intensity. I recommend no polarizing elements in the light path. If you keep your confocal system off at night, may take many hours for the laser output to stabilize (I have never been sure whether it is the lasers or the AOTF intensity controller as being the guilty party or parties). > > * If you are using an argon visible laser on an LSM510, run the laser at 6.1 Amps and leave the laser power dialog open (it might fluctuate on its own between 5.9 and 6.3 Amps - that is ok). On our system I've seen the power change to crazy values, like 7.9 Amps, apparently on its own (also on hitting re-use, even though I know the saved setting was not that value). On the LSM710 Argon laser's remote, turn the knob so it is in the middle of the green light range. > > * Laser power to the specimen ... lower is better! For example, cardiomyocytes die a spectacular death after a few seconds of 488 nm excitation when loaded with Fluo-4 and imaged at high power. > > If you are using an -AM conjugate, more is not better. Besides the possibility of turning your Calcium ion reporter into a chelator, the -AM leaving group is formaldehyde. Could be worse: GFP maturation produces one molecule of H2O2 (so two for CaMeleon), and red fluorescent protein maturation may produce two H2O2's. > > > No pixel in the background at zero ... correct, this is ideal. However, at high gain (say 900 volts) and an offset such that no laser light results in average intensity value around 100 (12-bit mode), when you are imaging with the laser and specimen, if you acquire enough images, eventually statistics dictate that some outlier will reach zero. So what! When you use the "no light" image (or an average of a whole bunch of them) in background subtraction, lots of values outside the cells will hit zero. > > A more important item is getting the detector gain and laser power optimized so you can use most of the dynamic range of the confocal to get the data you need. This requires guesswork or using the first specimen to get maximum intensity (typically, saturated Calcium ion with ionophore). > > Best wishes, > > George > > > At 09:05 AM 7/28/2010, you wrote: >> Hello. >> I had this discussion with several colleagues (biologist like me), and did some >> research on my own, but I figurate the best would be to submit the question >> to this list. Here's the point: we are doing quantitative measurement with >> confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There >> are 3 parameters of the PMT that can be configured: detector gain, amplifier >> offset, amplifier gain. The only parameter I adjust to improve the signal is the >> detector gain. I keep the amplifier gain at 1 as I read in several papers that >> increasing it will not improve the signal over noise ratio (they also say that it >> is bad for several reasons that I can not summarize here). Is that correct? I >> put the offset usually at zero. I saw that a change in offset can affect the >> calcium signal. In any case, I keep the same offset in a series of records that >> I whish to compare. Any comment on this? My guess is that the same should >> be done with the amplifier gain. >> Ah, a last point: the offset should be tuned so no pixel in the background has >> the value of zero right? >> Does anybody have any clarification for these questions? Thanks for your help. >> >> Regards >> Sandrine > > > > > > > > George McNamara, Ph.D. > Image Core Manager > Analytical Imaging Core Facility > University of Miami, Miller School of Medicine > Miami, FL 33136 > [hidden email] > [hidden email] > 305-243-8436 office > http://www.sylvester.org/AICF (Analytical Imaging Core Facility) > http://www.sylvester.org/AICF/pubspectra.zip (the entire 2000+ spectra .xlsx file is in the zip file) > http://home.earthlink.net/~geomcnamara |
|
|
Andreas Bruckbauer |
Re: quantitative confocal microscopy
|
|
As i understand from discussions with Zeiss, the LSM 510 Argon laser runs in current control mode. The light output might change over longer times but it should be fairly constant for the duration of the experiment as long as the laser is switched on for a while and temperature in the room is stable. When the tube current fluctuates strongly I guess that something is wrong with the control electronics or power supply.
best wishes Andreas -----Original Message-----
From: Nuno Moreno <[hidden email]> To: [hidden email] Sent: Thu, 29 Jul 2010 9:57 Subject: Re: quantitative confocal microscopy Just one comment related to this. When running in light mode, the laser would |
|
|
Vitaly Boyko |
Re: quantitative confocal microscopy
|
|
In reply to this post by George McNamara
Dear Confocalists,
I have a feeling that Confocalists have a strong wish to picture confocal imaging in very bright tones.
Accurate quantitative imaging on a standard (not custom built) confocal scope is a very tough job. And I have very serious doubts that biologists on their own can handle it professionally.
Sandrine, I suggest you to contact a large BioImaging Facility where mathematicians, physicists with some basic knowledge of biology work together with biologists, who are also part of the staff at the BioImaging Facility. Unfortunately, it might not be an easy task to find such a Facility.
MPI in Dresden? Janelia Farm???? any more?
Good luck,
Vitaly
From: George McNamara <[hidden email]> To: [hidden email] Sent: Wed, July 28, 2010 9:55:54 PM Subject: Re: quantitative confocal microscopy Hi Sandrine, Detector gain (PMT voltage) - ideally would like to be a modest value, such as 600 or 700. Amplifier offset - adjust so that no laser light results in all values being above zero. Note that if you change scan speed and/or detector gain, the offset value(s) may change. When using 12-bit data mode, I typically aim for an offset of ~100, but am happy with anything less than 200 (at moderate gain). Amplifier gain - you are correct. Do not change from 1.0. It is just does post acquisition arithmetic. Other items: You did not mention what fluorophore(s) or fluorescent protein you are using. Or whether you are using the best filter or other wavelength setting(s) for these. Depending on the cell's autofluorescence, this may need optimization. You did not mention laser power: * Make sure the laser output is stable! Both within a frame and over experiment time. A starting point is to turn on the transmitted light detector (which as a bonus gets you the cell morphology) and see that you are getting stable intensity. I recommend no polarizing elements in the light path. If you keep your confocal system off at night, may take many hours for the laser output to stabilize (I have never been sure whether it is the lasers or the AOTF intensity controller as being the guilty party or parties). * If you are using an argon visible laser on an LSM510, run the laser at 6.1 Amps and leave the laser power dialog open (it might fluctuate on its own between 5.9 and 6.3 Amps - that is ok). On our system I've seen the power change to crazy values, like 7.9 Amps, apparently on its own (also on hitting re-use, even though I know the saved setting was not that value). On the LSM710 Argon laser's remote, turn the knob so it is in the middle of the green light range. * Laser power to the specimen ... lower is better! For example, cardiomyocytes die a spectacular death after a few seconds of 488 nm excitation when loaded with Fluo-4 and imaged at high power. If you are using an -AM conjugate, more is not better. Besides the possibility of turning your Calcium ion reporter into a chelator, the -AM leaving group is formaldehyde. Could be worse: GFP maturation produces one molecule of H2O2 (so two for CaMeleon), and red fluorescent protein maturation may produce two H2O2's. No pixel in the background at zero ... correct, this is ideal. However, at high gain (say 900 volts) and an offset such that no laser light results in average intensity value around 100 (12-bit mode), when you are imaging with the laser and specimen, if you acquire enough images, eventually statistics dictate that some outlier will reach zero. So what! When you use the "no light" image (or an average of a whole bunch of them) in background subtraction, lots of values outside the cells will hit zero. A more important item is getting the detector gain and laser power optimized so you can use most of the dynamic range of the confocal to get the data you need. This requires guesswork or using the first specimen to get maximum intensity (typically, saturated Calcium ion with ionophore). Best wishes, George At 09:05 AM 7/28/2010, you wrote: > Hello. > I had this discussion with several colleagues (biologist like me), and did some > research on my own, but I figurate the best would be to submit the question > to this list. Here's the point: we are doing quantitative measurement with > confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There > are 3 parameters of the PMT that can be configured: detector gain, amplifier > offset, amplifier gain. The only parameter I adjust to improve the signal is the > detector gain. I keep the amplifier gain at 1 as I read in several papers that > increasing it will not improve the signal over noise ratio (they also say that it > is bad for several reasons that I can not summarize here). Is that correct? I > put the offset usually at zero. I saw that a change in offset can affect the > calcium signal. In any case, I keep the same offset in a series of records that > I whish to compare. Any comment on this? My guess is that the same should > be done with the amplifier gain. > Ah, a last point: the offset should be tuned so no pixel in the background has > the value of zero right? > Does anybody have any clarification for these questions? Thanks for your help. > > Regards > Sandrine George McNamara, Ph.D. Image Core Manager Analytical Imaging Core Facility University of Miami, Miller School of Medicine Miami, FL 33136 [hidden email] [hidden email] 305-243-8436 office http://www.sylvester.org/AICF (Analytical Imaging Core Facility) http://www.sylvester.org/AICF/pubspectra.zip (the entire 2000+ spectra .xlsx file is in the zip file) http://home.earthlink.net/~geomcnamara |
|
|
Nuno Moreno |
Re: quantitative confocal microscopy
|
|
There are 2 different types of quantification. Relative or Absolute. Relative quantification can be made in a regular imaging facility. Ion (ratio-metric) imaging brings some coarse absolute quantification from a relative/comparative ones and can therefore be made almost everywhere, assuming that you know what you are doing. The problem is that most of the times this last request in not achieved.
Regarding absolute quantification you will need to go for single/few molecule imaging, fcs, etc and then, you are right, it is not so common to have expertise on this field as a service. We are running a project to get coarse absolute quantification from a relative one using math and a calibrated light source but still not sure if we will have the time/expertise for doing this. For now it uses loci to get spectral info and works for SP5 and 510 but only relative (benchmarking). Have a look on www.cirklo.org -> mezuro Yours, Nuno Moreno, PhD Head of Equipment Management Instituto Gulbenkian de Ciência Fundação Calouste Gulbenkian Tel . +351 4464538 Fax. +351 4407970 On Jul 29, 2010, at 3:28 PM, Vitaly Boyko wrote: > Dear Confocalists, > > I have a feeling that Confocalists have a strong wish to picture confocal imaging in very bright tones. > > Accurate quantitative imaging on a standard (not custom built) confocal scope is a very tough job. And I have very serious doubts that biologists on their own can handle it professionally. > > Sandrine, I suggest you to contact a large BioImaging Facility where mathematicians, physicists with some basic knowledge of biology work together with biologists, who are also part of the staff at the BioImaging Facility. Unfortunately, it might not be an easy task to find such a Facility. > > MPI in Dresden? Janelia Farm???? any more? > > Good luck, > > Vitaly > > > From: George McNamara <[hidden email]> > To: [hidden email] > Sent: Wed, July 28, 2010 9:55:54 PM > Subject: Re: quantitative confocal microscopy > > Hi Sandrine, > > Detector gain (PMT voltage) - ideally would like to be a modest value, such as 600 or 700. > > Amplifier offset - adjust so that no laser light results in all values being above zero. Note that if you change scan speed and/or detector gain, the offset value(s) may change. When using 12-bit data mode, I typically aim for an offset of ~100, but am happy with anything less than 200 (at moderate gain). > > Amplifier gain - you are correct. Do not change from 1.0. It is just does post acquisition arithmetic. > > Other items: > > You did not mention what fluorophore(s) or fluorescent protein you are using. Or whether you are using the best filter or other wavelength setting(s) for these. Depending on the cell's autofluorescence, this may need optimization. > > You did not mention laser power: > > * Make sure the laser output is stable! Both within a frame and over experiment time. A starting point is to turn on the transmitted light detector (which as a bonus gets you the cell morphology) and see that you are getting stable intensity. I recommend no polarizing elements in the light path. If you keep your confocal system off at night, may take many hours for the laser output to stabilize (I have never been sure whether it is the lasers or the AOTF intensity controller as being the guilty party or parties). > > * If you are using an argon visible laser on an LSM510, run the laser at 6.1 Amps and leave the laser power dialog open (it might fluctuate on its own between 5.9 and 6.3 Amps - that is ok). On our system I've seen the power change to crazy values, like 7.9 Amps, apparently on its own (also on hitting re-use, even though I know the saved setting was not that value). On the LSM710 Argon laser's remote, turn the knob so it is in the middle of the green light range. > > * Laser power to the specimen ... lower is better! For example, cardiomyocytes die a spectacular death after a few seconds of 488 nm excitation when loaded with Fluo-4 and imaged at high power. > > If you are using an -AM conjugate, more is not better. Besides the possibility of turning your Calcium ion reporter into a chelator, the -AM leaving group is formaldehyde. Could be worse: GFP maturation produces one molecule of H2O2 (so two for CaMeleon), and red fluorescent protein maturation may produce two H2O2's. > > > No pixel in the background at zero ... correct, this is ideal. However, at high gain (say 900 volts) and an offset such that no laser light results in average intensity value around 100 (12-bit mode), when you are imaging with the laser and specimen, if you acquire enough images, eventually statistics dictate that some outlier will reach zero. So what! When you use the "no light" image (or an average of a whole bunch of them) in background subtraction, lots of values outside the cells will hit zero. > > A more important item is getting the detector gain and laser power optimized so you can use most of the dynamic range of the confocal to get the data you need. This requires guesswork or using the first specimen to get maximum intensity (typically, saturated Calcium ion with ionophore). > > Best wishes, > > George > > > At 09:05 AM 7/28/2010, you wrote: > > Hello. > > I had this discussion with several colleagues (biologist like me), and did some > > research on my own, but I figurate the best would be to submit the question > > to this list. Here's the point: we are doing quantitative measurement with > > confocal microscopy (calcium measurement) using a Zeiss LSM exciter. There > > are 3 parameters of the PMT that can be configured: detector gain, amplifier > > offset, amplifier gain. The only parameter I adjust to improve the signal is the > > detector gain. I keep the amplifier gain at 1 as I read in several papers that > > increasing it will not improve the signal over noise ratio (they also say that it > > is bad for several reasons that I can not summarize here). Is that correct? I > > put the offset usually at zero. I saw that a change in offset can affect the > > calcium signal. In any case, I keep the same offset in a series of records that > > I whish to compare. Any comment on this? My guess is that the same should > > be done with the amplifier gain. > > Ah, a last point: the offset should be tuned so no pixel in the background has > > the value of zero right? > > Does anybody have any clarification for these questions? Thanks for your help. > > > > Regards > > Sandrine > > > > > > > > George McNamara, Ph.D. > Image Core Manager > Analytical Imaging Core Facility > University of Miami, Miller School of Medicine > Miami, FL 33136 > [hidden email] > [hidden email] > 305-243-8436 office > http://www.sylvester.org/AICF (Analytical Imaging Core Facility) > http://www.sylvester.org/AICF/pubspectra.zip (the entire 2000+ spectra .xlsx file is in the zip file) > http://home.earthlink.net/~geomcnamara > > |
«
Return to Confocal Microscopy List
|
1 view|%1 views
| Free forum by Nabble | Edit this page |