Sensitivity and AOBS
Locked
|
Dear all,
We are in the process of purchasing a confocal microscope to serve a broad range of biological applications. We are considering the Leica SP5 and the Zeiss 710. Both companies are claiming “our confocal provides the highest sensitivity”, but they are unable to provide some proofs of that. Could you provide me some practical feedback? Particularly we are a bit confused about the AOBS. Leica people claim more than 95% transmittance, while Zeiss people are saying that the transmittance is much worse than any conventional beam splitter. Again nobody is providing any kind of evidence of such statements. Do you have any experience/information on this subject? Thanks. Dr. Ken. G. Pluomis |
 
|
CONFOCAL TL Andresen |
Re: Sensitivity and AOBS
|
|
Hi Ken, I have just been through
exactly the same process. I did not succeed in finding out actual proof of who
was right but ended up buying the SP5, which was mainly due to a better deal. We tested the microscopes
severely but could not find clear differences. One of the things I would
have liked to know is that Zeiss claims that their new dual beamsplitter suppresses
the laser light to a much higher degree than the AOBS (known as the OD number).
I asked Leica for data on the AOBS but they have never gotten back to me on
this even though I asked many times. They also say it doesn’t matter
because the laser light is removed later by the spectral cut off. But this is
what Zeiss claims to be the biggest advantage, and I could not find out if it
makes sense or not. I would be very happy to
know if you find actual differences. Best regards, Thomas
From:
Dear all, We are in the process of
purchasing a confocal microscope to serve a broad range of biological
applications. We are considering the Leica SP5 and the Zeiss 710. Both
companies are claiming “our confocal provides the highest
sensitivity”, but they are unable to provide some proofs of that. Could
you provide me some practical feedback? Particularly we are a bit
confused about the AOBS. Leica people claim more than 95% transmittance, while
Zeiss people are saying that the transmittance is much worse than any
conventional beam splitter. Again nobody is providing any kind of evidence of
such statements. Do you have any experience/information on this subject? Thanks. Dr. Ken. G. Pluomis |
|
|
Dear List,
I would greatly appreciate if you could refer to a supplier of
a high precision horizontal table (in a box-like configuration) for sample/slide
preparation and/or storage.
It is a kind of an obvious thing in the NEW ERA of "PALM and
STORM along thinking" or simply surface imaging (TIR microscopy
...).
Thank you very much in advance,
Vitaly
NCI-Frederick,
301-846-6575
|
|
|
In reply to this post by CONFOCAL TL Andresen
The key thing I would look for is a simple optical path and what light detector they are using. Photomultiplier tubes come in many different flavors with lots of different spectral ranges, sensitivities, and quantum efficiencies. Depending on what wavelength you are looking for your detector can have a large impact.
In terms of filtering out the laser light from the fluorescence signal; there are quite good band pass and long pass filters out there and they are commonly available. Rather than worrying about 10% plus or minus in the filtering I'd be more concerned about the actual detectors used and how the light is coupled into them. Craig On Thu, Oct 2, 2008 at 1:15 PM, <[hidden email]> wrote:
|
|
|
In reply to this post by Ken Pluomis
=
Hello Ken, I have struggled with this type of question in the past, and I have to admit from the start I don't have any answer. While getting some definite answer from someone will be valuable, I now feel this is probably a non-issue (at least for me): There are so may factors affecting the final sensitivity of a confocal system, that trying to find out if one particular component is better in this or that system is probably meaningless.... chances are they are both performing just fine. To test the relative performance of the two systems, the only way I can see would be to have a suitable sample (preferably fine structures of low-medium staining intensity), try to get an acceptable image on both systems (at some level of image quality based on resolution and signal to noise), and see how the images on both systems degrade over multiple frames when you run a time lapse. This of course would assume that both systems are perfectly adjusted and optimized, and are performing to the best of their abilities, which in itself is probably not a trivial thing to achieve or determine. In my opinion, both systems will be just fine for 95% of your applications. Then, there may be a few applications here or there for which one system may have an edge, although it is hard for me to imagine that you would come across any experiment that you could do on one system but not on the other. If you are anticipating really demanding experiments, then the best would be to try them on either system and see which one works best. I had to make a similar choice a few years ago between the Leica SP2 AOBS and the Zeiss LSM 510. We went with the 510 (for various reasons, not necessarily linked to performance). In retrospect, this is what I thought were the main issues we encountered: The 510 is filter-based. This means that for fast imaging, you need to have a multi-band dichroic and set your detection channels in such a way that you do not move any hardware between channels. By doing this, you sacrifice sensitivity (because of narrower detection bands). If you want to maximize sensitivity, you need to switch mirrors and filters, and you loose speed. Since you can't predict which combinations of dyes you will be using, there are times when you won't have the choice and will need to use the slower scan mode (with filters and mirrors switching). With a filterless system, you won't have to make those choices. This has not been a critical issue for us, though... by that I mean we have been able to live with those limitations and compromises. However, a more important issue has been that since we are in a multi-user facility, we are sometimes worried that users will have a configuration that requires mirror/filter switching, and there is a chance they try to run some fast scans (line scans, for instance), which will put incredible stress on the hardware. You won't run into this type of issues with a filterless system. Other than that, the 510 has performed fine for us, even though it has been a bit temperamental (hardware and software-wise). We don't own an SP2 (or SP5), so I don't know how happy we would have been with that, and I haven't looked carefully into the improvements the 710 brings. Other practical differences may be the speed at which you can run wide-band spectral scans on the two systems, so if you are considering doing fast spectral scans, maybe check how both systems perform in that respect. Another issue of practical importance will be the support you get from both companies in your area. Once you have purchased your instrument, it's unlikely you'll ever worry about the sensitivity of the AOBS again... my $ 0.02 (which are worth about $ 0.012 as we speak, and still going down...) Julio. -- Julio Vazquez Fred Hutchinson Cancer Research Center Seattle, WA 98109-1024 On Oct 2, 2008, at 11:55 AM, Ken Pluomis wrote:
|
|
|
In reply to this post by Craig Brideau
In addition to the filters and the PMTs, one issue is the light scatter within the confocal head. The best is a light path that is clean and there is no defracted light reaching the PMT. In previous Zeiss and Leica instruments this has been a problem. In the 710 confocals, Zeiss has greatly reduced this problem, but I have no experience with the Leica.
Dick Burry ----- Original Message ----- From: Craig Brideau <[hidden email]> Date: Thursday, October 2, 2008 4:27 pm Subject: Re: Sensitivity and AOBS To: [hidden email] > The key thing I would look for is a simple optical path and what light detector they are using. Photomultiplier tubes come in many different flavors with lots of different spectral ranges, sensitivities, and quantum efficiencies. Depending on what wavelength you are looking for your detector can have a large impact.
> In terms of filtering out the laser light from the fluorescence signal; there are quite good band pass and long pass filters out there and they are commonly available. Rather than worrying about 10% plus or minus in the filtering I'd be more concerned about the actual detectors used and how the light is coupled into them. > Craig > On Thu, Oct 2, 2008 at 1:15 PM, <<A href="javascript:main.compose('new','t=thomas.andresen@risoe.dk')">thomas.andresen@...> wrote:
> Spam > Not spam > Forget previous vote Richard W. Burry, Ph.D. Department of Neuroscience, College of Medicine Campus Microscopy and Imaging Facility, Director The Ohio State University Associate Editor, Journal of Histochemistry and Cytochemistry 277 Biomedical Research Tower 460 West Twelfth Avenue Columbus, Ohio 43210 Voice 614.292.2814 Cell 614.638.3345 Fax 614.247.8849 |
|
|
In reply to this post by Ken Pluomis
A very basic question is
what fraction of the emitted photons are detected ? An answer covers most of the critical microscope components. On the whole confocal microscopes are not short of excitation light, its detected photons that are in short supply. If you get answer from you suppliers you might share the information. Jeremy Adler Cell Biology The Wenner-Gren Inst. Arrhenius Laboratories F5 Stockholm University Stockholm 106 91 Sweden ________________________________ From: Confocal Microscopy List on behalf of Ken Pluomis Sent: Thu 02-Oct-08 20:55 To: [hidden email] Subject: Sensitivity and AOBS Dear all, We are in the process of purchasing a confocal microscope to serve a broad range of biological applications. We are considering the Leica SP5 and the Zeiss 710. Both companies are claiming "our confocal provides the highest sensitivity", but they are unable to provide some proofs of that. Could you provide me some practical feedback? Particularly we are a bit confused about the AOBS. Leica people claim more than 95% transmittance, while Zeiss people are saying that the transmittance is much worse than any conventional beam splitter. Again nobody is providing any kind of evidence of such statements. Do you have any experience/information on this subject? Thanks. Dr. Ken. G. Pluomis |
|
|
In reply to this post by Ken Pluomis
Hi, Ken,
A couple of years ago we had demos of the SP5 and 510 Meta at the same time. We could look at a single sample on one instrument and then compare
the images acquired by the second instrument. The SP5 blew the 510 away as far as sensitivity. There were samples that the 510 could not detect at all for which the SP5 produced lovely images. So the SP5 with AOBS is definitely more sensitive than the filter based 510. I don't know about the 710, though. If you can get Leica and Zeiss to demo their instruments side by side you can get a really good comparison of the performances of both machines. The Leica rep used to love these side-by-side comparisons when he was up against the 510. I don't know if they're as anxious to go up against the 710. Good Luck,
George
Dear all,
We are in the process of purchasing a confocal microscope to serve a broad range of biological applications. We are considering the Leica SP5 and the Zeiss 710. Both companies are claiming our confocal provides the highest sensitivity, but they are unable to provide some proofs of that. Could you provide me some practical feedback? Particularly we are a bit confused about the AOBS. Leica people claim more than 95% transmittance, while Zeiss people are saying that the transmittance is much worse than any conventional beam splitter. Again nobody is providing any kind of evidence of such statements. Do you have any experience/information on this subject? Thanks. Dr. Ken. G. Pluomis |
|
|
In reply to this post by Ken Pluomis
Dear Dr. Sudipta Maiti
I would really appreciate you could send me this paper as well as the program.
Many thanks.
Dr. Ken G. Pluomis
----- Original Message ---- From: Sudipta Maiti <[hidden email]> To: [hidden email] Sent: Friday, October 3, 2008 5:35:50 AM Subject: Re: Sensitivity and AOBS One of the best ways I know to do this is described in our paper: Quantitative measurement of the resolution and sensitivity of confocal microscopes using line-scanning fluorescence correlation spectroscopy It may sound a bit daunting to the uninitiated, but it is pretty simple, really. All it takes is to record an image of a bead solution for a few seconds. The data can be analyzed with a program that I can send off the list. Sudipta On Thu, 2 Oct 2008 12:02:49 -0700, Ken Pluomis wrote |
|
|
Goodhouse, Joseph G. |
Re: Sensitivity and AOBS
|
|
In reply to this post by George Ring
One
thing you need to measure for these comparisons is the amount of
excitation power that is coming to your specimen. This can vary due to the
different optics but they should be set to the same level on each system.
You need a tunable sensitive light meter that can read in the uWatt range placed
above the objective. Scan in a spot or line mode rapidly to prevent meter
blanking.
Joe Goodhouse Visit us at http://www.molbio1.princeton.edu/facility/confocal/ From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of George Ring Sent: Friday, October 03, 2008 8:37 AM To: [hidden email] Subject: Re: Sensitivity and AOBS Hi, Ken,
A couple of years ago we had demos of the SP5 and 510 Meta at the
same time. We could look at a single sample on one instrument and then
compare
the images acquired by the second instrument. The SP5 blew the 510 away as far as sensitivity. There were samples that the 510 could not detect at all for which the SP5 produced lovely images. So the SP5 with AOBS is definitely more sensitive than the filter based 510. I don't know about the 710, though. If you can get Leica and Zeiss to demo their instruments side by side you can get a really good comparison of the performances of both machines. The Leica rep used to love these side-by-side comparisons when he was up against the 510. I don't know if they're as anxious to go up against the 710. Good Luck,
George
Dear all,
We are in the process of
purchasing a confocal microscope to serve a broad range of biological
applications. We are considering the Leica SP5 and the Zeiss 710. Both companies
are claiming “our confocal provides the highest sensitivity”, but they are
unable to provide some proofs of that. Could you provide me some practical
feedback? Particularly we are a bit
confused about the AOBS. Leica people claim more than 95% transmittance, while
Zeiss people are saying that the transmittance is much worse than any
conventional beam splitter. Again nobody is providing any kind of evidence of
such statements. Do you have any experience/information on this
subject? Thanks. Dr. Ken. G.
Pluomis |
|
|
Hi All,
Joe's suggestion is an important part of determining the overall
quantum detection efficiency of any confocal or wide field microscope,
the main question then being how many fluorescence photons are
captured and detected per the number of photons that reach the
excitation volume in the sample. About 10+ years ago I went to the
trouble of performing this measurement with a BioRad 1024 equipped
with an inverted Nikon using a 1.4 NA objective.
As a reference sample I used a either a 1 or? 10 uM fluorescein
(pH 8.0, 50% glycerol: 50% PBS) sample in freshly prepared
diphenylenediamine antifade and used a focus position 10 um from
coverslip into the solution. I had previously found that intensity
measurements using fluorescein solutions ranging from 0.1 uM up to 100
uM gave a decent straight line. Of course, different total dwell times
were used and the data normalized accordingly.
To make this story shorter, taking into account the Q.E.
and e of fluorescein at pH 8.0, the laser
power delivered at 488 nm, the measured fluorescence signal, pinhole
size set to realize optimal x,y,z resolution, I measured an
overall Q.D.E. of ~0.0004, which was similar to the results given in
an article in the 2nd Edition of J.P.'s "....Handbook of Confocal
Microscopy..." for an earlier version of the BioRad confocal
(R.I.P.). The 1024 model detected only about 1 photon in about 2500
emitted photons. The Nikon objective did its job meaning that with an
NA of 1.4 it managed to captured about 30% of emitted photons; the
rest of the losses, a factor of 0.0013, resulted from the other
optical and detection components of the system. Clearly, there was a
very large potential for improvement in sensitivity.
Finally, for present day microscopes, a crude working comparison
between instruments should not be too difficult. Just make reference
sample(s) that emit at a couple of wavelengths that are reasonably
stable such as Alexa 488 nm and a couple of others at longer
wavelengths for which the Q.E.'s and extinction coefficients are
known, or which you can measure (there are a lot of options for ref.
samples). Make sure that you use objectives that are nominally rated
for the same NA (I always measure the PSF for any objective I buy,
especially at the high end) and make sure that the settings for
pinholes are properly configured so you are working at the Rayleigh
resolution and that you normalize for differences in dwell time.
Always check the statistical noise of the output signal. Even when the
signals are strong the noise profile can tell you whether you are
truly getting the dynamic range suggested by the peak signal
intensities. You can have outputs that appear nearly identical in
intensities but one system may give you 32 levels of statistically
useful values whereas another may give you 64 or better.
Although there are many highly qualified and expert microscope
vendor representatives, there are many sales people that lack the
expertise to explain the technical differences that pertain to their
instruments. It's caveat emptor folks.
Final comment: In this instance, since I have been a third party
consultant for various pharmaceutical companies purchasing new
microscopy systems, I can't claim to be a disinterested party.
Mario
One thing you need to measure for these comparisons is the amount of excitation power that is coming to your specimen. This can vary due to the different optics but they should be set to the same level on each system. You need a tunable sensitive light meter that can read in the uWatt range placed above the objective. Scan in a spot or line mode rapidly to prevent meter blanking. Joe Goodhouse Visit us at http://www.molbio1.princeton.edu/facility/confocal/ -- ________________________________________________________________________________
Mario M. Moronne, Ph.D. ph (510) 528-8076
|
| Free forum by Nabble | Edit this page |