Re: Fluoview 1000 and FCS

Posted by Jennifer Clarke on
URL: http://confocal-microscopy-list.275.s1.nabble.com/Fluoview-1000-and-FCS-tp4594644p4637254.html

Dear Stan

The preceeding scan size doesnt matter for point scanning FCS measurement, and
the measurement volume is determined by the confocal volume.

However yes the spatial resolution of the preceeding image would, in part,
affect your accuracy for positioning the xyz coordinate of the point FCS
measurement.   For us the preceeding image is usually 256x256 (ie 256 pixels
per line) or 512x512 with 60 or 63x lens and between 1 and 6x zoom for
visualisation of details of the cell (again zoom doesnt matter for the FCS
measurement, only for visualisation).

For live cells, a much bigger problem here however is movement of the cellular
components during the course of the FCS point measurement.
This is where RICS and scanning FCS and the SimFCS software become invaluable as
the bulk cell movement during the measurement can be taken into account in the
analysis.  

For solution measurements we would typically take at least 3 measurements and
average them.

For measurements on cells, we are trying to measure the movement of receptors on
the membrane and the movement of the respective ligands, which we add to the
media.  This is very difficult as the cell membrane is constantly in motion.
We try to position several measurement points along the membrane according to
an image.  Due to movement of the cell since the image was taken and during the
course of the measurements not all the measurement points end up actually
including the membrane.  Usually it is clear in the analysis which measurement
points were entirely intracellular and which were extracellular.  Our main
problem now is how to accomodate enough components into the FCS analysis (eg
for the ligand, diffusion components include 1 triplet state, 2 free diffusion
in extracellular component of measurement volume, 3 restricted movement in
vicinity of membrane, 4 receptor binding events, and potentially any
internalised ligand as well) and the fact that different membrane measurement
points will include different proportions of extracellular and intracellular
space).
So, on live cells it gets very complicated vey quickly!

Here we have a Leica SP5 with Avalance Photo Diode detectors and an FCS System
with ISS Vista software.  This is fine for FCS.  
Unfortunately there are problems using the APD image data aquired in the Leica
software (LASAF) for analysis with RICS and N&B as the LASAF data is somehow
modified and no longer exists as true raw data.  True raw data is required for
analysis with RICS and N&B as these techniques analyse the intensity
fluctuations, so the data cannot be in any way already smoothed or averaged.
With Enrico Grattons help, we are trying to find a way around this problem via
feeding the APD image data straight into ISS Vista, bypassing LASAF.

Unfortunately we do not regularly have access to an Olympus FV1000, we just used
one when we were fortunate to visit the Gratton lab last year.

Hope this helps

I am not an expert in this, so I would be interested in any other listers
comments too

Kind regards
Jennifer



Quoting Stanislav Vitha <[hidden email]>:

> Dear Jennifer,
> when you do point scanning FCS on the Fluoview 1000, what scan size do you
> typically have prior to selecting the point scan tool? That is, how many
> pixels per line was in the live view image? I am just curious about the
> total number of datapoints that one would need e.g., for solutions or cells
> (pixels per line x timepoints).
>
> Thanks!
>
> Stan Vitha.
> Microscopy and Imaging Center,
> Texas A&M University
>
> On Tue, 23 Feb 2010 11:19:43 +1030, Jennifer Clarke
> <[hidden email]> wrote:
>
> >Dear all
> >
> >I can confirm that the FV1000 can be used for FCS measurements, we also used
> it
> >with SimFCS software and it works very well.
> >
> > - the laser does not blink, it just scans continually at the designated
> point
> >(use crosshair icon and select desired point on image for measurement)
> >
> > - the max number of points on FV1000 in point scanning mode is 32766.
> This
> >number of points is sufficient for an accurate measurement.
> >
> >Acquisition in this way is that it is equivalent to capturing FCS data in
> "time
> >mode" in ISS Vista software (the other system we use, where we can choose
> either
> >"time mode" which counts photons/time-block, or "photon mode" which counts
> each
> >photon with its respective time).
> >
> >In ISS Vista, we always exclude the first data point in the analysis,
> because
> >it
> >has no preceeding data points to be correlated with for the correlation
> >analysis.
> >I cant recall excluding the first point when we have analysed FV1000 point
> scan
> >data in SimFCS, I'm not sure if there is an option to do so.  If the first
> data
> >point can be excluded in the analysis I would do so.  Do not simply delete
> the
> >first data point, as this would simply mean that now the second data point
> has
> >no first point to be correlated to so the problem still remains.
> >If the first point can not be excluded from the analysis, it is only one
> point
> >whose correlation will be of reduced accuracy, out of the 32766 points, so
> I
> >dont think this matters very much, and perhaps (? - maybe a mathematician
> can
> >answer) it is not as important in time mode as in photon mode.
> >
> >SimFCS is definately worth trying, yes 30day demo licence is free, and the
> full
> >licence is very economical anyway.  Go to the website for the Laboratory
> for
> >Fluorescence Dynamics; http://www.lfd.uci.edu/
> >
> >For point-scanning data collection on the FV1000, configure for optimal
> >resoluion, ie use a high NA objective etc, select photon counting mode, ie
> >"photon cnt", chose appropriate pixel time eg 8-12.5us/pixel for solution
> or
> >12.5-40us/pixel for measurement on cells, select crosshair icon and set
> desired
> >point in image, set number of frames to 32766 (if you type in an overly
> high
> >number eg 1000000 it will automatically reset to the maximum available)
> with
> >interval set to "freerun".  Make sure you export the data as the "raw"
> 16-bit
> >files.
> >
> >If you need a hand working through the analysis in the SimFCS software
> contact
> >me offline.
> >Note that the Wo (PSF waist) in SimFCS is defined using 1/e2 max intensity
> not
> >half max intensity.
> >
> >Its always a good idea to include a reference dye for which you already have
> a
> >good idea of expected or theoretical diffusion coefficient.
> >
> >Hope this helps
> >Kind regards
> >Jennifer
> >
> >--
> >Jennifer Clarke BSc (Hons) PhD
> >Research Associate, Anatomy and Histology
> >Centre for Neuroscience, School of Medicine
> >&
> >Facility Manager, Optical Microscopy Suite, Flinders Microscopy
> >
> >Flinders University
> >GPO Box 2100, Adelaide 5001
> >Phone: [phone]
> >Email: [hidden email]
> >
> >--------------------------------------------------------
> >
> >Quoting Stanislav Vitha <[hidden email]>:
> >
> >> Hi,
> >> I second Aseem's suggestion regarding SImFCS.
> >>
> >> A user of our core facility is using FV1000 and the SimFCS software from
> the
> >> Gratton lab, it appears to work quite well, as far as I know. The SimFCS
> >> software can be installed as a fully-functional 1-month demo, so you can
> >> test it yourself before making the final decision. One big advantage of
> the
> >> RICS approach that I see is the ability to subtract slow movements, such
> as
> >> sample or stage drift.
> >> A while ago I was told by Olympus that point scanning was not going to
> work
> >> for FCS on the FV1000 system, but I do not quite remember their technical
> >> explanation. From my tests, I remember there was an issue with the first
> or
> >> last datapoint being of incorrect value, due to the timing of the AOTF
> >> "shutter", or synchronization of the AOTF with data acquisition. You
> could
> >> try point-scanning of the fluorescent plastic slide from Chroma, the
> >> fluctuations that you see should indicate the instrument
> noise/instability,
> >> including the laser.
> >> The number of datapoints in point-scanning mode on FV1000 was limited to
> >> 4096, i.e., the max. permitted line width. I was told that before a scan
> a
> >> piece of code is sent to the control board(s), essentially reprogramming
> the
> >> firmware, and this somehow imposes a limit on the number of datapoints.
> >> You could do time-lapse point scan to acquire multiples of 4096 points,,
> but
> >> then you have an unknown amount of time between the time frames.
> >>
> >> Anyway, I think SimFCS is worth trying.
> >>
> >> Stan Vitha
> >> Microscopy and Imaging Center
> >> Texas A&M University
> >>
> >> On Sat, 20 Feb 2010 22:09:52 +0530, aseem mishra <[hidden email]>
> >> wrote:
> >>
> >> >Hi Jean,
> >> >
> >> >You might want to have a look at Enrico Gratton's work and specially
> >> >SimFCS wherein one can calculate diffusion coefficients from an image
> >> >collected in a point-scan mode. I do not know if a blinking laser
> >> >gives you the possibility of doing FCS (Do let me know if such a
> >> >possibility exists. I would love to try it at my place.) We rather use
> >> >a hardware correlator and a simple laser (diode/HeNe)
> >> >
> >> >Aseem
> >> >
> >> >
> >> >On Fri, Feb 19, 2010 at 3:27 AM, Jean-Pierre CLAMME
> >> ><[hidden email]> wrote:
> >> >>
> >> >> Hi all,
> >> >>
> >> >> I'm thinking about modifying our fluoview 1000 to do FCS. The fluoview
> >> has
> >> >> the possibility to do single point measurement.  However as we can
> still
> >> >> choose the pixel time in this mode, I was wondering if the laser beam
> >> always
> >> >> stays at the same point of interest and continuously illuminates the
> >> point
> >> >> or if it actually "blinks".
> >> >>
> >> >> If it doesn't "blink", I would think it could hinder the FCS
> measurement
> >> et
> >> >> thus I'm wondering if there is an option to change this in the soft. ?
> >> >>
> >> >> Thanks,
> >> >>
> >> >> JP
> >> >>
> >> >>
> >> >>
> >> >>
> >> >> - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
> >> >> Jean-Pierre CLAMME, PhD
> >> >> Senior Scientist
> >> >> Nitto Denko Technical
> >> >> 501 Via Del Monte
> >> >> Oceanside, CA 92058
> >> >> E-mail: [hidden email]
> >> >> Phone: [phone]
> >> >>
> >> >
> >> >
> >> >
> >> >--
> >> >Aseem Mishra
> >> >Senior Research Assistant,
> >> >Malaria Group,
> >> >International Centre for Genetic Engineering and Biotechnology,
> >> >New Delhi-110067
> >> >INDIA
> >>
> >
> >
> >--
> >Jennifer Clarke BSc (Hons) PhD
> >Research Associate, Anatomy and Histology
> >Centre for Neuroscience, School of Medicine
> >&
> >Facility Manager, Optical Microscopy Suite, Flinders Microscopy
> >(available for training and assistance on Mondays only)
> >
> >Flinders University
> >GPO Box 2100, Adelaide 5001
> >Phone: [phone]
> >Email: [hidden email]
>

--
Jennifer Clarke BSc (Hons) PhD
Research Associate, Anatomy and Histology
Centre for Neuroscience, School of Medicine
&
Facility Manager, Optical Microscopy Suite, Flinders Microscopy
(available for training and assistance on Mondays only)

Flinders University
GPO Box 2100, Adelaide 5001
Phone: [phone]
Email: [hidden email]