Subject: Re: Comparison Zeiss 710 vs Nikon A1 ... full comparison requires matching resolution and scan speed
I would expect that in the photon counting (PC) mode all the different
confocals beahve as they should, i.e., longer dwell time = more counts.
Since PC is generally less sensitive to detector voltage fluctuations, it is
a good idea to use this mode for quantitative measurements anyway.
Now, the Olympus FV1000 uses "Hybrid" photon counting. Can somebody please
comment what exactly it is?
Thanks.
Stan
On Mon, 10 May 2010 14:11:51 -0700, Glen MacDonald
<[hidden email]> wrote:
>The FV-1000 also allows automatic scaling of the HV to maintain PMT
sensitivity when switching from fast scanning at 2 µs to full resolution
scanning at any longer dwell time. I wonder if the Zeiss has something
similar that alters the detectors over all scan rates, but which is less
obvious and has a less direct user control than the Fluoview's "Auto HV"
button that sits next to the dwell time selector.
>
>Glen
>On May 10, 2010, at 1:45 PM, Andreas Bruckbauer wrote:
>
>> On the Olympus FV1000 the signal varies (intentionally) with the scan
speed, longer dwell time = more photons. Zeiss seems to give out something
like counting rates which are then the same for different dwell times. It
would be good if the manufacturers give more details.
>>
>> best wishes
>>
>> Andreas
>>
>>
>>
>> -----Original Message-----
>> From: George McNamara <[hidden email]>
>> To: [hidden email]
>> Sent: Sun, 9 May 2010 14:58
>> Subject: Re: Comparison Zeiss 710 vs Nikon A1 ... full comparison
requires matching resolution and scan speed
>>
>> Hi listserv,
>>
>> Repeating what I posted in a message yesterday:
>>
>> I've noticed on both the Leica SP5, Zeiss LSM510 and Zeiss LSM710 that
"faster is brighter" for many fluorophores. That is, a 0.4 us dwell time
results in a brighter image than a 4 us which is brighter than a 40 us (then
repeated 0.4 us, which resulted in the same brightness as the first 0.4 us
dwell time - so not photobleaching -- not all fluorophores in the specimen
[Desmid slide from Carolina] changed brightness). This implies that <0.4 us
might be even brighter - i.e. resonant scanner mode. A couple of possible
explanations (not mutually exclusive):
>> a) photophysics (possibly caused by fluorophore-O2 photochemistry) - re
TRex (PubMed 19337661), papers by Sanden/Spielmann/Widergren et al (PubMed
20375039, 20196585, 19007245, 17385841).
>> b) calibration method(s) by Zeiss and/or Leica that try to (but do not
always) match output at all settings.
>> c) other??? (suggestions/comments welcome!).
>> See also
>>
>> high speed scanning has the potential to increase fluorescence yield and
to reduce photobleaching. Borlinghaus RT. Microsc Res Tech. 2006
Sep;69(9):689-92.PMID: 16878313
>>
>> Since I can obtain brighter or dimmer fluorescence on the same field of
view on the same microscope system by simply changing scan speed, I submit
that between microscope comparisons are just anecdotes unless every variable
is controlled and reported. Even then there may be variables under the hood
that the confocal manufacturer has not mentioned, such as the possibility
that the system is "calibrated" to all scan speeds output more or less the
same intensity.
>>
>>
>>
>> Comments with respect to the post below...
>>
>>
>> At 03:43 PM 4/11/2010, you wrote:
>>
>>> Hi Pedro,
>>>
>>> Seems you have short listed A1 and 710. Both the systems are released
almost the same time. We have little experience in handling the 710 system
but sufficiently exposed with A1. I have mentioned my opinion and
observation about these two systems earlier that is given below.
>>>
>>> A1 gives 4X4K image size, whereas 710 gives 6X6K. But do we really need
6KX6K? When we tried taking 6K images, we have experienced photo bleaching
in some attempts.
>>
>> GM: With a 100x1.4+ NA objective lens at the LSM710's 0.6x zoom (and
perfect specimen, i.e. refractive index matching and/or imaging right at the
coverglass ... and coverglass 170.0 um)
>>
>> XY resolution = 0.6 * 405 nm (the laser line) / 1.4 = 173 nm (assuming
pinhole 1.0 Airy unit, infinitely bright specimen could put sqrt(2) in the
denominator).
>> Nyquist sampling for a 2D image; 173 nm / 3 = 57 nm.
>>
>> (I'm not at the LSM710 so don't have the actual image field of view, but
this should be close):
>>
>> For 100x lens and 0.6x zoom, field of view should be about 160 um, so
>>
>> 160 um / 6000 pixels = 0.026 um = 26 nm.
>>
>> So, 6kx6k is oversampling by about a factor of two. For a very stable
specimen (photostability and vibration isolation), perfect imaging
conditions, lots of time (since probably will need Z-series for best
deconvolution) and optimized confocal deconvolution algorithm, the 6kx6k
setting might be useful. How much more useful than 4kx4k ... please do the
experiment, publish and send me the reference.
>>
>> GM Most confocal microscope software have "optimize" XY resolution
buttons - when in doubt, click on it. Incidentally, Paul Goodwin, Applied
Precision, told me Biotium CF405 fluorophore is very good for
immunofluorescence (context: OMX nanoscope). Might be a good starting point
for an experiment such as above. DAPI or Hoechst 33342 or 33358 in a perfect
mounting medium could also be a good choice for the above.
>>
>>> A1 has continuously variable zoom up to 1000X but LSM 710 is limited to
50X. However, we were not given proper training about how to use variable
zoom up to 1000X. In my view, the 1000X variable zoom is not that important
factor.
>>>
>>> Fortunately, in 710 there is a master pinhole unlike the earlier 510
that takes time for alignment. However, 710 uses conventional rectangular
pinhole but A1 has unique hexagonal pinhole resulting better images. We have
checked the same sample in both the systems.
>>
>> GM: Managing both a 4-pinhole LSM510 and a LSM710, I teach users the same
thing on both for multi-color fluorescence (ex. "colocalization"): for the
longest wavelength emission channel, select 1 Airy button, then match the
pinhole for the other channels. This way the (nominal) optical section
thickness is matched. In multitrack mode the pinhole setting can be changed
between tracks.
>>
>>
>>> The spectral detector of LSM 710 has 34 channels with simultaneously
acquisition unlike the previous model 510 that was with 4x8=32. However, the
spectral step size is 3nm (3x34=102nm wavelength resolution) and it is not
flexible like A1 that has 2.5nm, 6nm and 32nm (resulting 2.5x32=80nm,
6x32=192nm and 10x32= 320nm wavelength resolution). A1 has surprisingly
effective unmixing efficiency even in the close range.
>>
>> GM: LSM710/ZEN's 32-channel QUASAR detector default is ~10 nm, but is
adjustable (5 and 2.5 nm if I recall). When using the flanking PMTs the
spectral window for these can be adjusted in 1 nm steps - this is done with
two prism/slits - see the LSM710 internals schematic on the Zeiss web site.
Going to 1 nm might be useful to play around with the peaks vs off-peaks for
Europium or Terbium (which have long fluorescence lifetimes so scan speed
would need to be very slow, might be interesting if you have the time to do
a spectral scan overnight).
>>
>>> In LSM710 high wavelength range up to 1100 nm is possible for optimized
transmission. I am not sure if the A1 has the same capability. In our
(little) experience, 710 gives wonderful sensitivity. Their software Zen
2009 is compatible with Vista/Windows 7. Even anisotropy imaging is possible
with LSM 710 (do not know about A1). But you need to purchase this module
(extra cost).
>>>
>>> Nikon software is robust has almost all of the regular modules. However,
for Zen we need to pay extra for the add on modules. Some of the regular
functions are "optional" with Zen.
>>>
>>> For simultaneous photo activation and imaging, one need to incorporate
Duo system (two heads) into LSM 710 that adds up the cost. However, A1R scan
head serves this purpose without any cumbersome modifications and the speed
and performance is relatively incomparable. Though there are some annoying
"noise" is generated while we use the resonant scanner the speed and
performance is still impressive.
>>>
>>> If you are regularly going for live cell imaging and Ca++ imaging, I
feel A1 has so many features and it is not bleaching the dyes or induce
unexpected phototoxicity. Though, we found 710 has slightly better
sensitivity, we often face bleaching and laser induced toxicity problems for
the same set of experiments.
>>
>> GM: See top of message - scan speed, resolution, laser power at specimen,
need to be matched to make valid comparison.
>>
>>
>>> Eager to know the inputs of other users.
>>>
>>> No commercial interest.
>>>
>>> Roshma.
>>>
>>>
>>>
>>> On Sun, Apr 11, 2010 at 4:34 PM, Pedro J Camello <[hidden email]> wrote:
>>> Hi all,
>>>
>>> has anybody in the list compared Nikon A1 vs Zeiss 710. We´re purchasing a
>>> spectral micro with a TIRF module and motorization. Any input will be
>>> really wellcome (off list if you prefer)
>>>
>>> A second question, what is the most close to "live cell" sample to make
>>> real tests in confocal? I´m going to travel to test a couple of micros,
>>> and to carry or prepapre real living cells is rather complicated for us.
>>> We´re especially interested in ion (Ca2+) experiments.
>>>
>>> Thanks
>>>
>>>
>>> --
>>> Dr Pedro J Camello
>>> Dpt Physiology
>>> Faculty of Veterinary Sciences
>>> University of Extremadura
>>> 10071 Caceres
>>> Spain
>>> Ph: 927257000 Extension 51321/51290
>>> Fax:927257110
>>>
>>
>>
>>
>>
>>
>>
>> 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
>
>
>
>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]