How are published spectra measured?

[Shalin Mehta]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Dear All,

 Is it safe to assume that spectra published on vendor websites of different dyes and FPs are acquired with 'flat-response' measurement systems ?
I wish to get estimates of spectrum that is presented to detector for some alexa dyes and GFP variants. Invitrogen folks please inform us how alexa dye spectra are measured.
If I apply the spectral curves of objectives, dichroics and filters (again measured with 'flat-response' meter which I think is safe to assume), will the resultant spectrum be represntative of what is seen by detector in reality? If anyone has tried this earlier, please share what you saw.


Cheers,
Shalin

--
~~~~~~~~~~~~~~~~~~~~~~~~~
Shalin Mehta
Graduate Student in Bioengineering, NUS
mobile: [phone]
blog: shalin.wordpress.com
~~~~~~~~~~~~~~~~~~~~~~~~~~

[George McNamara]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Hi Shalin,

There is no such thing as a 'flat response' system, only uncalibrated and calibrated detectors.

Also, bad idea to assume anything about anything. This is what the methods & materials sections of peer reviewed publications are for. Details on several of the Alexa Fluor dyes were published in 2000 in JHC (Panchuk-Voloshina et al):

Absorbance measurements were performed with a Hitachi U-2000 spectrophotometer (Hitachi Instruments; Boulder, CO). Fluorescence measurements were performed using an SLM-Aminco fluorometer SLM Instrument (Urbana, IL).


The spectra that MPI/Invitrogen acquire on their spectrofluorimeters are much higher resolution (typically 1 nm step size, often higher) than what you can acquire on a microscope based system unless (a) you attach a spectrofluorimeter (like the old SLM) or spectrometer to your microscope (yes, it has been done) or (b) acquire images with www.lightforminc.com's PARISS imaging spectrometer (typically 0.8 nm step size, high dynamic range, and essentially no crosstalk between wavelengths). You can ask Guy Cox and others how well the inexpensive Ocean Optics spectrometers work for acquiring microscopy spectra.

See articles by Zucker and Lerner in Cytometry and elsewhere for many of the things that go wrong on confocal "spectral imagers". Some of the most obvious problems of commercial spectral imagers are (1) the Zeiss META detector has movable laser line blocking pins in front of the 32-channel PMT -- if the pins are moved, you get different spectra (and you can't assume they are always where you want them or the instrument says they are), the META detector spectra are only as good as the last calibration (which is done by matching intensities in adjacent channels with the transmitted lamp on), (3) one (or more?) Leica SP1 spectrometer leaf locked up after several uses. Zucker and Lerner also published on something called "WSI" (start wavelength), which varied between scanheads on the various META's, SP1's and SP2's they tested. Read their articles for details.


Chroma, Omega, Semrock etc have hundreds of filter spectra data posted online. Most of the dye/fluorescent protein data are already in PubSpectra ( http://home.earthlink.net/~pubspectra/) and can also be graphically viewed, and some efficiency calculations done online, at the Fluorescent Dye Spectra site at UArizona, http://www.mcb.arizona.edu/ipc/fret/index.html

Zeiss ("shop zeiss.com") and Olympus have many objective lens spectra posted online as graphs. If you or anyone else want to digitize them, I would love to have the data to add to PubSpectra.

George



At 01:46 AM 11/26/2007, you wrote:

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Dear All,

 Is it safe to assume that spectra published on vendor websites of different dyes and FPs are acquired with 'flat-response' measurement systems ?
I wish to get estimates of spectrum that is presented to detector for some alexa dyes and GFP variants. Invitrogen folks please inform us how alexa dye spectra are measured.
If I apply the spectral curves of objectives, dichroics and filters (again measured with 'flat-response' meter which I think is safe to assume), will the resultant spectrum be represntative of what is seen by detector in reality? If anyone has tried this earlier, please share what you saw.


Cheers,
Shalin

--
~~~~~~~~~~~~~~~~~~~~~~~~~
Shalin Mehta
Graduate Student in Bioengineering, NUS
mobile: [phone]
blog: shalin.wordpress.com
~~~~~~~~~~~~~~~~~~~~~~~~~~

[Ignatius, Mike]

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Dear Shalin,

 

All the data published on our website are measured on a spectrophotometer (absorption) or a spectrofluorometer (fluorescence excitation and emission).  The spectrofluorometer is corrected for wavelength-dependent variations of excitation source intensity, monochromator/lens/mirror transmission/reflectivity and photomultiplier efficiency.  Thus the curves are, as far as possible, instrument independent and can be reproduced on any instrument that has been likewise corrected.   The samples used for these fluorescence measurements are optically dilute (OD<0.1 at excitation wavelength) and the pathlength is 1 cm.  

 

Although the spectral bias of the optical response elements on a microscope can be corrected in the same way as on a spectrofluorometer, this seems to be rarely done in practice.  Nevertheless spectrofluometer-acquired reference spectra and spectra acquired on a microscope equipped with a spectrally-resolved detector often show surprisingly good correspondence.  We did a poster comparing LSM 510 Meta data to the web reference spectra at the Biophysics meeting in 2002. It is posted here: http://probes.invitrogen.com/media/publications/276.pdf.

 

Kindly,

 

The crew at Molecular Probes/Invitrogen (Iain Johnson to be precise.)

 

---

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Shalin Mehta
Sent: Sunday, November 25, 2007 10:46 PM
To: [hidden email]
Subject: How are published spectra measured?

Search the CONFOCAL archive at http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal Dear All,

 Is it safe to assume that spectra published on vendor websites of different dyes and FPs are acquired with 'flat-response' measurement systems ?
I wish to get estimates of spectrum that is presented to detector for some alexa dyes and GFP variants. Invitrogen folks please inform us how alexa dye spectra are measured.
If I apply the spectral curves of objectives, dichroics and filters (again measured with 'flat-response' meter which I think is safe to assume), will the resultant spectrum be represntative of what is seen by detector in reality? If anyone has tried this earlier, please share what you saw.


Cheers,
Shalin

--
~~~~~~~~~~~~~~~~~~~~~~~~~
Shalin Mehta
Graduate Student in Bioengineering, NUS
mobile: [phone]
blog: shalin.wordpress.com
~~~~~~~~~~~~~~~~~~~~~~~~~~