Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Correspondence
  • Published:

Reply to "Polarization modulation adds little additional information to super-resolution fluorescence microscopy"

Nature Methods volume 13, pages 8–9 (2016)Cite this article

Subjects

Walla et al. reply:

We are grateful to Keller et al. for the opportunity to clarify the role of the SPEED (sparsity penalty–enhanced estimation by demodulation) algorithm and the impact of the polarization information in SPoD (super-resolution by polarization demodulation). We concur with Keller et al. that for densely labeled and spatially sparse samples, the information gain from polarization is negligible in their reanalysis of our data, as they observed that nearly identical resolution can also be achieved by sparse deconvolution alone using the SPEED algorithm. However, this does not mean that no gain in resolution is possible with the modulation information alone.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

Figure 1: Raw modulation data as a function of label polarization and fiber distance in actin filaments.

References

  1. Hafi, N. et al. Nat. Methods 11, 579–584 (2014).

    Article  CAS  Google Scholar 

Download references

Acknowledgements

We thank A. Albrecht for help with generating some figures. Parts of this work were supported by the German Science Foundation (CRC 803).

Author information

Authors and Affiliations

  1. Department of Biophysical Chemistry, Institute for Physical and Theoretical Chemistry, University of Braunschweig, Braunschweig, Germany

    Nour Hafi, Laura S van den Heuvel & Peter J Walla

  2. Biomolecular Spectroscopy and Single-Molecule Detection Group, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany

    Matthias Grunwald & Peter J Walla

  3. Felix Bernstein Institute for Mathematical Statistics in Bioscience, Göttingen, Germany

    Timo Aspelmeier & Axel Munk

  4. Institute for Mathematical Stochastics, University of Göttingen, Göttingen, Germany

    Timo Aspelmeier & Axel Munk

  5. Statistical Inverse Problems in Biophysics Group, Max Planck Institute for Biophysical Chemistry, Göttingen, Germany

    Timo Aspelmeier & Axel Munk

  6. Institute of Organic and Biomolecular Chemistry, University of Göttingen, Göttingen, Germany

    Claudia Steinem

  7. Division of Cellular Neurobiology, Zoological Institute, University of Braunschweig, Braunschweig, Germany

    Martin Korte

Authors
  1. Nour Hafi
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Matthias Grunwald
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Laura S van den Heuvel
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Timo Aspelmeier
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Claudia Steinem
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Martin Korte
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Axel Munk
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Peter J Walla
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Peter J Walla.

Ethics declarations

Competing interests

The University of Braunschweig, with which N.H., L.S.v.d.H., M.K. and P.J.W. are affiliated, has filed patents (US patent 14/362,346 and German patent 102011087770.3) for parts of this work.

Supplementary information

Supplementary Text and Figures

Supplementary Figure 1, Supplementary Methods and Supplementary Note (PDF 3236 kb)

Modulation data from various samples.

Diffraction limited images, movies of the modulated part of the emission, images of the constant part of the emission, movies of the total signal and color coded phase information from fluorescence modulation for various samples. (a) Single fluorescence molecules on coverslip. (b) Liposomes in liquid environment. (c) Membrane-labelled neurons. (d) Immunolabeled microtubules. (e) Molecule pair simulations of varying distances and relative orientation differences. (f) Labeled actin filaments on cover slide. The diffraction limited images are the average of the total signals in the raw data movies. The raw data movies are composed of a modulated part and a constant part. The modulated part is the time-dependent, modulated fraction in the raw data (constant offset in each pixel subtracted). The remaining non-modulated part is the constant, time-independent offset in the total signal. The phase information is derived from Fourier analysis in each pixel and is color coded. As described in Hafi et al. the relative modulation amplitude depends approximately on 1/√Nmol and is thus smaller in the total signal for more densely labelled samples. The modulated raw data are averages from (a) 70, (b) 10, (c) 90, 80, (d) 70, (e) 1 and (f) 20 periods and are repeated ten times in the Video. Set-up dependent periodical deviations in the overall excitation intensity (typically ~5%) were corrected for. (MOV 7664 kb)

Raw modulation data from various samples.

(a-d) Diffraction limited images, (e-h) modulated parts and (i-l) SPEED images as well as profile plots of single molecule pairs (a,e,i), membrane labeled neurons (b,f,j) and immunolabeled microtubules (c,d,g,h,k,l). The profile plots for the modulation movies (e-h) show the modulation amplitudes (black). Colored profile plots: individual time frames from the modulated data with varying contributions from the spine apparatus and the closest membrane (f) and from the two fibers (g). The modulated raw data are averages from 6, 90 and 70 periods for the single molecule, neuron and microtubule data, respectively, and are repeated ten times in the Video. A drift correction was applied for samples that moved more than one pixel during that time. (MOV 11866 kb)

Raw modulation data as a function of label polarization and fiber distance in actin filaments.

(a) Total signal modulation movie. (b) Corresponding movie of the modulated part of the emission (constant offset in each pixel subtracted). (c) Total signal modulation movie from inset i. (d) Corresponding color coded phase image. The phase was derived from Fourier analysis in each pixel and is a measure for the label polarization. (e) Corresponding movie of the modulated part of the emission. (f,g) Two different image frames from raw modulating data from inset ii. Blue lines: linear fits of the fiber positions. (h) Total signal modulation movie from inset ii differing by 40° in label polarization along with profile plot lines used for the data shown in i-k. (i) Profile plots from the total signal modulation movie shown in h. Blue vertical lines indicate the true positions of the fibers. (j) Corresponding profile plots series of the modulated part of the emission. (k) Corresponding profile plots of the raw modulating data in two selected image frames. (l-n) Results of fitting two modulating Gaussian functions to the data shown in i, j and k, respectively. For both Gaussians the position, the modulation amplitude, the modulation phase and a constant offset were optimized. Blue vertical lines indicate the true positions of the fibers, black vertical lines indicate fitted positions. The fitted positions are also indicated by white dots in (h). To improve fit quality, the agreement with the pure raw modulating data (compare j and m) was weighed twice as high as the agreement with the total signal (compare i and l). (o) Corresponding fit to time-averaged data (no modulation). Blue vertical lines indicate the true positions of the fibers, red vertical lines indicate fitted positions. (p) Deviation of the fitted fiber positions from the true fiber positions for the 40° data. Black dots denote difference between fitted and true positions in the modulated data (l). Red dots denote the difference between fitted and true positions in the unmodulated data (o). (q) Data observed for two fibers differing by 70° in label polarization (inset iii). (r) Data observed for two fibers differing by 5° in label polarization (inset iv). The modulated data are averages from 20 periods and are repeated ten times in the Video. (MOV 7864 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hafi, N., Grunwald, M., van den Heuvel, L. et al. Reply to "Polarization modulation adds little additional information to super-resolution fluorescence microscopy". Nat Methods 13, 8–9 (2016). https://doi.org/10.1038/nmeth.3721

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/nmeth.3721

This article is cited by

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing