Fixed Pattern Noise in Localization Microscopy

Authors

  • Dr. Patrick Fox-Roberts,

    1. Randall Division of Cell and Molecular Biophysics, New Hunt's House, Guy's Campus, King's College London, SE1 1UL (United Kingdom)
    Search for more papers by this author
  • Tianqi Wen,

    1. Randall Division of Cell and Molecular Biophysics, New Hunt's House, Guy's Campus, King's College London, SE1 1UL (United Kingdom)
    2. Department of Electronic Engineering, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong SAR (China)
    Search for more papers by this author
  • Dr. Klaus Suhling,

    1. Department of Physics, King's College London, Strand, London, WC2R 2LS (United Kingdom)
    Search for more papers by this author
  • Dr. Susan Cox

    Corresponding author
    1. Randall Division of Cell and Molecular Biophysics, New Hunt's House, Guy's Campus, King's College London, SE1 1UL (United Kingdom)
    • Randall Division of Cell and Molecular Biophysics, New Hunt's House, Guy's Campus, King's College London, SE1 1UL (United Kingdom)

    Search for more papers by this author

Abstract

Localization microscopy vastly improves the resolution achieved by fluorescence microscopy by fitting the positions of individual fluorophores. We examine the reconstructions produced by different fitting algorithms for instances of fixed pattern noise—systematic tendencies to alter estimated emitter positions according to their subpixel location in a way that does not reflect the ground truth structure. We show that while not readily visible at standard empirical signal strengths, fixed pattern noise can occur when performing sub-pixel fitting, and that its degree varies according to the algorithm used and the relative size of the pixels compared to the point spread function. For pixel sizes in the range 80–170 nm, this results in variations in accuracy of the order of 2–4 nm—comparatively small for many applications, but non-negligible in scenarios where very high accuracy is sought.

Ancillary