Spectral datasets for Rainbow Beads and Fluorophores
16 views
Skip to first unread message
Peter McLean
Sep 30, 2016, 5:37:55 PM9/30/16
to SBSC Flow Cytometry Working Group
Hi everyone,
In preparing to work through some analyses of the data we have coming in from our mini interlab, I was left unsure about which information we should be using for our measurement calibration. To make our calibration a reality, we need:
1) the optical configuration of each instrument (we're settling for the reported band-pass wavelengths, correct?)
2) a consensus excitation and emission spectrum for the fluorophores we're using (both the green and red proteins) — @Nicholas: what version of GFP are we looking at? @General: who has the best spectral data for these fluors?
3) a consensus excitation and emission spectrum for the rainbow beads we're using to calibrate (are we still in agreement that the Spherotech-provided data is the best we currently have? do we have ANY excitation information about the beads?) — @General: could someone who knows where that data is located copy it to the interlab directory?
The spectral data we use for this study (both the fluors and the beads) should be placed in a reference folder inside the interlab directory so it is forever associated with this data and the results. Light-path/optical configurations of instruments could be placed in a README with each dataset.
If we are in need of better (or any) spectral data on either the fluors or the beads, let me know and I'll email some NIST folk who might be able to help us.
Cheers,
Peter
In preparing to work through some analyses of the data we have coming in from our mini interlab, I was left unsure about which information we should be using for our measurement calibration. To make our calibration a reality, we need:
1) the optical configuration of each instrument (we're settling for the reported band-pass wavelengths, correct?)
2) a consensus excitation and emission spectrum for the fluorophores we're using (both the green and red proteins) — @Nicholas: what version of GFP are we looking at? @General: who has the best spectral data for these fluors?
3) a consensus excitation and emission spectrum for the rainbow beads we're using to calibrate (are we still in agreement that the Spherotech-provided data is the best we currently have? do we have ANY excitation information about the beads?) — @General: could someone who knows where that data is located copy it to the interlab directory?
The spectral data we use for this study (both the fluors and the beads) should be placed in a reference folder inside the interlab directory so it is forever associated with this data and the results. Light-path/optical configurations of instruments could be placed in a README with each dataset.
If we are in need of better (or any) spectral data on either the fluors or the beads, let me know and I'll email some NIST folk who might be able to help us.
Cheers,
Peter
Jake Beal
Sep 30, 2016, 6:02:21 PM9/30/16
to sbsc-flow...@googlegroups.com
Hi, Peter:
1) the optical configuration of each instrument (we're settling for the reported band-pass wavelengths, correct?)
Yes, that's correct.
The information recorded is in cytometers.gsheet in the mini-interlab directory.
2) a consensus excitation and emission spectrum for the fluorophores we're using (both the green and red proteins) — @Nicholas: what version of GFP are we looking at? @General: who has the best spectral data for these fluors?
I've made a folder called "spectra" in the mini-interlab folder and put in spectra obtained from one of the online services.
3) a consensus excitation and emission spectrum for the rainbow beads we're using to calibrate (are we still in agreement that the Spherotech-provided data is the best we currently have? do we have ANY excitation information about the beads?) — @General: could someone who knows where that data is located copy it to the interlab directory?
I've put the excitation/emission table that we've got from SpheroTech into the spectra folder. It's only got emission information for excitation with a few of the most common lasers, but that will hopefully be sufficient for this test.
Thanks,
-Jake
Peter McLean
Oct 12, 2016, 4:32:34 PM10/12/16
to SBSC Flow Cytometry Working Group
So for a formal followup, the 2016 interlab folder has been reorganized a bit, but has all of the data we have collected to date. I have duplicated the spherotech spectra csv and added it as an xlsx sheet with some manually scaled plots.
In followup to our conference call—and some more careful thinking about the data we have and what we're trying to do—I'd like to post the following corrections/hypotheses:
- We can use the spectra from the beads and the fluors—for any available excitation wavelength—to convert measured a.u. signal values to "bead units" for all band-passes for that excitation line. This would put all measurements for a given laser onto the same scale, albeit it still arbitrary and unit-less.
- Because the beads are excitable at all wavelengths, and emit across 'all' wavelengths—and also because the emission profile does not appear to change significantly as excitation wavelength increases—we can(?) assign MEFL units to the measured bead peaks in all channels, for all laser lines.
- With this, it's a simple transformation to take all channels for all lasers and place measured fluors onto the rainbow bead MEFL scale.
I *currently think* this works for the same reason that spillover is calculated as a linear contribution to the target channel... but I'm not too confident in this idea yet. My current big question is whether we can ignore the differences in "excitability" for the reference particles and the fluors for all excitation wavelengths.
Peter
Why? Well, glad you asked. I believe that these emission spectra allow us to establish the relationship between the fluors and the beads through different band-passes.
In followup to our conference call—and some more careful thinking about the data we have and what we're trying to do—I'd like to post the following corrections/hypotheses:
- We can use the spectra from the beads and the fluors—for any available excitation wavelength—to convert measured a.u. signal values to "bead units" for all band-passes for that excitation line. This would put all measurements for a given laser onto the same scale, albeit it still arbitrary and unit-less.
- Because the beads are excitable at all wavelengths, and emit across 'all' wavelengths—and also because the emission profile does not appear to change significantly as excitation wavelength increases—we can(?) assign MEFL units to the measured bead peaks in all channels, for all laser lines.
- With this, it's a simple transformation to take all channels for all lasers and place measured fluors onto the rainbow bead MEFL scale.
I *currently think* this works for the same reason that spillover is calculated as a linear contribution to the target channel... but I'm not too confident in this idea yet. My current big question is whether we can ignore the differences in "excitability" for the reference particles and the fluors for all excitation wavelengths.
Peter
Why? Well, glad you asked. I believe that these emission spectra allow us to establish the relationship between the fluors and the beads through different band-passes.
Reply all
Reply to author
Forward
0 new messages