Two Q's: CCD's and mEOS2

[Feinstein, Timothy]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hello all,

I have two questions, bundled into one email to save space.  Any help on or off-list is much appreciated.

1. Not precisely confocal, but related enough.  As the new core manager I am bringing quantitative widefield, TIRF and in particular FRET to what was previously an exclusive confocal core.  This will call for a good CCD but at the moment we have a Hamamatsu Orca 2.8 SCMOS, which is nice for speed but lacking in sensitivity and un-binnable.  As I recall EM-CCDs have the best sensitivity but a problem with non-Gaussian noise.  Is this still true?  If you had to choose one model for the best combination of sensitivity and quantitative response, what would you choose?  Thanks in advance.

2. While helping several users with photoconversion of mEOS2, we have noticed an interesting relationship between laser intensity and ROI scan speed.  In our hands it seems like turning the 405nm laser up mostly increases bleaching, whereas effective photoconversion depends on finding a long enough pixel dwell time.  Maybe I am just showing my lack of experience with photoconversion, but it seemed like a counterintuitive result.

Also, I just heard about mEOS4 on this list for the first time.  Does its increased stability mean anything for live imaging, or does it mostly help with post-fixative applications?

Thanks and all the best,


TF


Timothy Feinstein, Ph.D. | Confocal Manager, Van Andel Research Institute
333 Bostwick Ave., N.E., Grand Rapids, Michigan 49503
Phone: [phone] | Email: [hidden email]<mailto:[hidden email]>

[Craig Brideau]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Re Q2: Does your system auto-adjust exposure time when you 'zoom' or use an
ROI? (sorry, had to ask just in case)  What are the timescales on which
photoconversion of the molecule takes place vs. bleaching?


On Fri, Nov 8, 2013 at 12:38 PM, Feinstein, Timothy <
[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hello all,
>
> I have two questions, bundled into one email to save space.  Any help on
> or off-list is much appreciated.
>
> 1. Not precisely confocal, but related enough.  As the new core manager I
> am bringing quantitative widefield, TIRF and in particular FRET to what was
> previously an exclusive confocal core.  This will call for a good CCD but
> at the moment we have a Hamamatsu Orca 2.8 SCMOS, which is nice for speed
> but lacking in sensitivity and un-binnable.  As I recall EM-CCDs have the
> best sensitivity but a problem with non-Gaussian noise.  Is this still
> true?  If you had to choose one model for the best combination of
> sensitivity and quantitative response, what would you choose?  Thanks in
> advance.
>
> 2. While helping several users with photoconversion of mEOS2, we have
> noticed an interesting relationship between laser intensity and ROI scan
> speed.  In our hands it seems like turning the 405nm laser up mostly
> increases bleaching, whereas effective photoconversion depends on finding a
> long enough pixel dwell time.  Maybe I am just showing my lack of
> experience with photoconversion, but it seemed like a counterintuitive
> result.
>
> Also, I just heard about mEOS4 on this list for the first time.  Does its
> increased stability mean anything for live imaging, or does it mostly help
> with post-fixative applications?
>
> Thanks and all the best,
>
>
> TF
>
>
> Timothy Feinstein, Ph.D. | Confocal Manager, Van Andel Research Institute
> 333 Bostwick Ave., N.E., Grand Rapids, Michigan 49503
> Phone: [phone] | Email: [hidden email]<mailto:
> [hidden email]>
>

[George McNamara]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi Tim,

I disagree with your statement "lacking in sensitivity". The quantum
efficiency of your ORCA-2.8 is better than the ORCA4.0 at less than 500
nm, and better than "Gen 1" sCMOS at less than ~570 nm. See Spectral
Response graph at
http://www.hamamatsu.com/jp/en/community/life_science_camera/product/search/C11440-22CU/index.html
and SNR curve next to it.

The Huang et al method for full calibration of every pixel should apply
to all sCMOS cameras and deal with the pixel specific gains (which is
not really noise).

Video-rate nanoscopy using sCMOS camera-specific single-molecule
localization algorithms. <http://www.ncbi.nlm.nih.gov/pubmed/23708387>

Huang F, Hartwich TM, Rivera-Molina FE, Lin Y, Duim WC, Long JJ, Uchil
PD, Myers JR, Baird MA, Mothes W, Davidson MW, Toomre D, *Bewersdorf J*.

Nat Methods . 2013 Jul;10(7):653-8. doi: 10.1038/nmeth.2488.  PMID: 23708387


You can achieve the optical equivalent of binning by not using the
highest magnification objective lens.

There are specific applications where back illuminated EMCCD vs sCMOS,
one will win. For example, UAIM is EMCCD specific:

Ultrahigh accuracy imaging modality for super-localization microscopy.
<http://www.ncbi.nlm.nih.gov/pubmed/23455923>

Chao J, Ram S, Ward ES, Ober RJ.

Nat Methods. 2013 Apr;10(4):335-8. doi: 10.1038/nmeth.2396.  PMID: 23455923


I recommend you put the FLASH2.8 on the microscope that makes the most
sense and if your customer base needs another camera system, select
whatever camera type, and /manufacturer/model fits that need.


mEos4 - still unpublished, except for a poster I saw in Loren Looger's
lab (and maybe some online abstracts that Google has not found). Try
contacting Maria at
https://www.linkedin.com/pub/maria-gabriela-paez-segala/61/978/b55 or
Loren at HHMI Janelia Farm for more details.


Sincerely,


George


On 11/8/2013 1:38 PM, Feinstein, Timothy wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hello all,
>
> I have two questions, bundled into one email to save space.  Any help on or off-list is much appreciated.
>
> 1. Not precisely confocal, but related enough.  As the new core manager I am bringing quantitative widefield, TIRF and in particular FRET to what was previously an exclusive confocal core.  This will call for a good CCD but at the moment we have a Hamamatsu Orca 2.8 SCMOS, which is nice for speed but lacking in sensitivity and un-binnable.  As I recall EM-CCDs have the best sensitivity but a problem with non-Gaussian noise.  Is this still true?  If you had to choose one model for the best combination of sensitivity and quantitative response, what would you choose?  Thanks in advance.
>
> 2. While helping several users with photoconversion of mEOS2, we have noticed an interesting relationship between laser intensity and ROI scan speed.  In our hands it seems like turning the 405nm laser up mostly increases bleaching, whereas effective photoconversion depends on finding a long enough pixel dwell time.  Maybe I am just showing my lack of experience with photoconversion, but it seemed like a counterintuitive result.
>
> Also, I just heard about mEOS4 on this list for the first time.  Does its increased stability mean anything for live imaging, or does it mostly help with post-fixative applications?
>
> Thanks and all the best,
>
>
> TF
>
>
> Timothy Feinstein, Ph.D. | Confocal Manager, Van Andel Research Institute
> 333 Bostwick Ave., N.E., Grand Rapids, Michigan 49503
> Phone: [phone] | Email: [hidden email]<mailto:[hidden email]>
>
>    

--



George McNamara, Ph.D.
Single Cells Analyst
L.J.N. Cooper Lab
University of Texas M.D. Anderson Cancer Center
Houston, TX 77054
Tattletales http://works.bepress.com/gmcnamara/26/

[Orla Hanrahan]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
*****

Hi Tim,

If you need to choose one camera with the best sensitivity and best quantitative response you can choose either EMCCD or sCMOS.  Both cameras are sensitive and fast and have good quantum efficiency.  

EMCCD's are extremely sensitive with <1e- read noise, QE above 90 % and the ability to use the camera as an EM camera or a conventional CCD.  Andor have a range of EMCCD cameras available and our most popular camera is the iXon Ultra 897. This has a 512 x 512 sensor format, 16 um pixels and readout speeds of 56 fps full FOV. A new mode just added, the Optically centered Crop Mode, allows one to readout at 569 fps at 128 x 128 ROI. This is an ideal mode for Live Cell Super Resolution.

There is a noise associated with EMCCD cameras and this is called multiplicative noise which stems from the EM gain. The addition of this noise which is equivalent to sqrt (2) effectively reduces the QE of the camera.

sCMOS on the other hand offers a large FOV, excellent resolution with 6.5 um pixels, a 5.5 MP or 4.2 MP sensor format and a QE of 72 %.  As well as this the read noise is ~1e- without EM gain and sCMOS sensors can run at 100 fps with the full FOV or much faster at smaller ROIs.  

At the end of the day the camera you choose depends on the applications which will be studied.  EMCCD technology will outperform sCMOS when it comes to single molecule detection experiments or any low light modalities such as spinning disk confocal microscopy.

If you need any more information about these technologies please visit our website (www.andor.com) or alternatively you can contact me directly.

Best wishes,

Orla Hanrahan
-------------------------------------------------------------------------------
Orla Hanrahan, PhD
Application Specialist, Life Science Imaging, Andor Technology

[hidden email]






On Fri, Nov 8, 2013 at 12:38 PM, Feinstein, Timothy < [hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> *****
>
> Hello all,
>
> I have two questions, bundled into one email to save space.  Any help
> on or off-list is much appreciated.
>
> 1. Not precisely confocal, but related enough.  As the new core
> manager I am bringing quantitative widefield, TIRF and in particular
> FRET to what was previously an exclusive confocal core.  This will
> call for a good CCD but at the moment we have a Hamamatsu Orca 2.8
> SCMOS, which is nice for speed but lacking in sensitivity and
> un-binnable.  As I recall EM-CCDs have the best sensitivity but a
> problem with non-Gaussian noise.  Is this still true?  If you had to
> choose one model for the best combination of sensitivity and
> quantitative response, what would you choose?  Thanks in advance.
>
> 2. While helping several users with photoconversion of mEOS2, we have
> noticed an interesting relationship between laser intensity and ROI
> scan speed.  In our hands it seems like turning the 405nm laser up
> mostly increases bleaching, whereas effective photoconversion depends
> on finding a long enough pixel dwell time.  Maybe I am just showing my
> lack of experience with photoconversion, but it seemed like a
> counterintuitive result.
>
> Also, I just heard about mEOS4 on this list for the first time.  Does
> its increased stability mean anything for live imaging, or does it
> mostly help with post-fixative applications?
>
> Thanks and all the best,
>
>
> TF
>
>
> Timothy Feinstein, Ph.D. | Confocal Manager, Van Andel Research
> Institute
> 333 Bostwick Ave., N.E., Grand Rapids, Michigan 49503
> Phone: [phone] | Email: [hidden email]<mailto:
> [hidden email]>
>