High Speed Stage Scanning, Part IV

I’ve continued to test our high speed stage stitching (I, II, IIIa, IIIb). This time I’ve imaged the same sample as before, now using a 20x/0.75 NA objective. This requires shortening the strobe duration to 20 μs and requires 594 images to cover the entire tissue.  This objective also has a substantially smaller depth of field, and it appears that the Nikon Perfect Focus System and the mechanical Z-drive on the Ti cannot move fast enough to keep the sample in perfect focus while scanning at full speed. Leveling the stage insert so that the focus difference from one edge of the specimen to the other was only a few microns seems to have fixed this.

While testing this, I also compared the timing for this fast scanning acquisition to two other acquisition modes. In both of these modes, the stage stops at each coordinate where the images are acquired. The first mode uses hardware triggering and the ARRAY functions of the ASI controller as described by Austin, the second just generates a position list in Micro-Manager and acquires it normally through software.  The acquisition time differences are pretty pronounced:

  • Fast scanning: 90.7 s
  • ASI Triggered: 288 s
  • Micro-Manager: 565 s

Continue reading

Weekly Paper Roundup: Week of May 13

As relatively few microscopy papers are published each week, I’ll be only publishing these paper updates every month from now on.

  • A new light driven dimerization system has been published. This one uses UVR8, which is a dimer that dissociates on exposure to ~300 nm light.  The authors have used it to generate a light-activated protein secretion system [1].


  1. D. Chen, E.S. Gibson, and M.J. Kennedy, "A light-triggered protein secretion system", Journal of Cell Biology, vol. 201, pp. 631-640, 2013. http://dx.doi.org/10.1083/jcb.201210119

High Speed Stage Scanning, Part III

In my last post I promised to explain how I implemented high speed slide scanning in Micro-Manager. At the bottom of this post is the Micro-Manager code; I’ll walk you through how it works shortly. But first a preview of the results:


A thumbnail of the resulting stitched image. This is an H&E stained spleen section, approximately 2 cm x 1 cm, imaged with a 10x / 0.45 NA objective. This was stitched together from 176 images, acquired in 50 seconds. I’m only showing the thumbnail because I haven’t figured out a good way to post a 200 megapixel, 500 MB image.

Continue reading

Weekly Paper Roundup: Week of May 5

  • Sunney Xie’s lab has a new paper using a light-sheet approach for imaging single molecules. They use an AFM canteliever as a mirror to generate a light sheet by reflecting a beam from the transmitted direction. [1]
  • Brainbow 3 – a new version of Brainbow mice using improved fluorescent proteins.  It comes in a number of variants, and the supplementary material includes assessment of the performance of 15 fluorescent proteins in neurons in transgenic mice. [2]


  1. J.C.M. Gebhardt, D.M. Suter, R. Roy, Z.W. Zhao, A.R. Chapman, S. Basu, T. Maniatis, and X.S. Xie, "Single-molecule imaging of transcription factor binding to DNA in live mammalian cells", Nature Methods, vol. 10, pp. 421-426, 2013. http://dx.doi.org/10.1038/nmeth.2411
  2. D. Cai, K.B. Cohen, T. Luo, J.W. Lichtman, and J.R. Sanes, "Improved tools for the Brainbow toolbox", Nature Methods, vol. 10, pp. 540-547, 2013. http://dx.doi.org/10.1038/nmeth.2450

Choosing a PC for High Speed Imaging, Part III

I’ve posted previously about our attempts to build a PC capable of the full 1.1 GB/sec data rate that the Andor Zyla is capable of. Yesterday, we got our PC back from Colfax International, who generously agreed to swap out the Intel SSDs we originally had for Samsung 840 Pro SSDs. With four of these SSDs in RAID 0, we can write data at 2.1 GB/sec.  So there’s no problem keeping up with the full data rate of the Andor Zyla.


Sequential read and write speed of our 1TB RAID 0 array, using 4GB writes of random data.

Of course, at 1.1 GB/sec, we’ll fill the entire 1TB array in about 15 minutes, but that’s a problem for another day.  My next task is to actually do some microscopy with this system.


Paper Roundup, week of April 29th

  • A set of cell-permeable dyes that become fluorescent on exposure to H2S have been developed for following H2S production in vivo. This is a niche application, but a cool, novel reporter dye. [1]
  • A quantum dot conjugation procedure for attaching quantum dots to antibodies via protein A, and a protocol for doing multiple rounds of staining, imaging, stripping, and restaining for multiplex imaging.  They demonstrate the possibility of five round of five color staining [2]
  • A method of barcoding cells by expressing zinc finger constructs on their surface and then binding fluorescently labeled DNA to these. [3]


  1. V.S. Lin, A.R. Lippert, and C.J. Chang, "Cell-trappable fluorescent probes for endogenous hydrogen sulfide signaling and imaging H2O2-dependent H2S production", Proceedings of the National Academy of Sciences, vol. 110, pp. 7131-7135, 2013. http://dx.doi.org/10.1073/pnas.1302193110
  2. P. Zrazhevskiy, and X. Gao, "Quantum dot imaging platform for single-cell molecular profiling", Nature Communications, vol. 4, 2013. http://dx.doi.org/10.1038/ncomms2635
  3. P. Mali, J. Aach, J. Lee, D. Levner, L. Nip, and G.M. Church, "Barcoding cells using cell-surface programmable DNA-binding domains", Nature Methods, vol. 10, pp. 403-406, 2013. http://dx.doi.org/10.1038/nmeth.2407

Basic Ray Optics for Microscope Design

As mentioned in the previous post, I’ve been working on designing a microscope to be built on an optical rail. As part of the design, I’ve needed to calculate a bunch of distances and sizes – for instance, the size of the back focal plane – that are not usually provided by the objective manufacturer, but that are easy to calculate. So that you won’t have to hunt down all the necessary formulas (most are in chapter 9 of the Handbook of Biological Confocal Microscopy), I thought I would reproduce them here. Continue reading