[quickpalm] PALM and STORM hardware - cylindrical lens for 3D STORM and PALM with astigmatism

[Ricardo Henriques]

Dear Colleagues,

After some requests I have uploaded the characteristics of the
cylindrical lens we use on the QuickPALM paper to obtain the 3D super-
resolution. If you are interest there is also a small tutorial online
(http://code.google.com/p/quickpalm/wiki/Tutorial_Astigmatism) on how
we set the lens on our system.

Personally I feel that the lens absorbs quite a bit of light and this
might be problematic as the resolution in PALM/STORM depends on the
SNR. We have had great results using dyes that emit a large amount of
photons such as Cy5 (around 6000 photons allowing 40 nm 3D resolution)
but I'm afraid that it might be slightly problematic while using
genetically encoded fluorophores that emit in the range of 500
photons.

Let me know if you find a better alternative to this lens.

Best regards,
Ricardo Henriques

[Ricardo Henriques]

Hello everyone,

First of all, in case you have missed it, here is the link for the cylindrical lens characteristics: http://quickpalm.googlegroups.com/web/Cylindrical+lens+reference.pdf?gda=nv1HZFEAAAC9KNNxQAtQSvJYlhpa2Pnu53IX7veOli7sDDDpRpc7nOGh-X324QMbS4ipb7FxAzaXP_5gCo_PzxCP_EwLygrUUwk_6Qi3BU8HCN0q6OYwM5VxXgp_nHWJXhfr7YhqVgA

A second note on this subject. I have been asked how problematic it is to constantly slide in and out the slider with the lens and if there is a need to constantly recalibrate the system every time you slide the lens in.

I did some tests on my system (Nikon TI), I like the way the TI works because the slider makes a "click" when it gets to the correct position, maybe the other microscope bodies do the same. So, the tests are simple, we measure the width and height of imaged beads over depth by moving the piezo at 10nm steps, then we take out the slider with the lens, reinsert it and remake the measurements. They actually don't show any meaningful changes and I think it is safe to say that we can reinsert the slider as much as we want without major disruption to the alignment. This said, it does not hurt to make new calibration tables from time to time. See attached some picture that might help or check out this tutorial: http://code.google.com/p/quickpalm/wiki/Tutorial_Astigmatism.

[Jan]

Hi there,
I'm currently trying to incorporate astigmatism into my microscopy system to use it for particle tracking and have to settle on the right lens now. Could you explain how you came up with the focal length and the position of the cylindrical lens in your setup? Of course the focal length has to be high and judging from your papers, it seems to make sense to place it in the infinity path of the objective. However, I've also seen cases where even bigger focal lengths were chosen and the lens was placed right before the camera in the path of the imaging lens (see Kubitschek et al: http://www.opticsinfobase.org/oe/abstract.cfm?uri=oe-20-18-19697 ). Could you elaborate on that?

Best regards,
Jan

[Ricardo Henriques]

Hi Jan,

We are putting the lens just after the emission filter in the the filter cube (non-infinity path). We use a 10m FL, in comparison the group of X. Zhuang is using a 1m FL cylindrical lens (not fully sure, but you can double check their papers) just before the camera. 

We based our FL decision on the experience of another group doing astigmatism (Ignacio Izeddin at ENS - Paris), they calculated the optimal FL based on the objective magnification, camera pixel size and the objective-camera distance. The choice of a long FL means a week astigmatism, our PSF does not critically increase in size when the particles are in the focal plane. However I cannot pinpoint precisely their calculations, if you feel this is something important then perhaps you can contact them directly.

Here's the variations of our PSFs over the z-axis: http://youtu.be/tH84fO0u9v8

Regards,

-Ricardo

--
Ricardo Henriques, PhD
Institut Pasteur (Paris, France).
For contact information see: https://sites.google.com/site/paxcalpt/

[Jan]

Thanks for the help, I'll look into it!