%0 Journal Article %1 Huang2013a %A Huang, Fang %A Hartwich, Tobias M P %A Rivera-Molina, Felix E %A Lin, Yu %A Duim, Whitney C %A Long, Jane J %A Uchil, Pradeep D %A Myers, Jordan R %A Baird, Michelle A %A Mothes, Walther %A Davidson, Michael W %A Toomre, Derek %A Bewersdorf, Joerg %D 2013 %J Articles nAture methods %K cameras detectors imaging microscopy scmos smlm superresolution %N 653 %R 10.1038/Nmeth.2488 %T Video-rate nanoscopy using scmos camera-specific single-molecule localization algorithms %V 10 %X newly developed scientific complementary metal-oxide semiconductor (scmos) cameras have the potential to dramatically accelerate data acquisition, enlarge the field of view and increase the effective quantum efficiency in single-molecule switching nanoscopy. however, scmos-intrinsic pixel-dependent readout noise substantially lowers the localization precision and introduces localization artifacts. We present algorithms that overcome these limitations and that provide unbiased, precise localization of single molecules at the theoretical limit. using these in combination with a multi-emitter fitting algorithm, we demonstrate single-molecule localization super-resolution imaging at rates of up to 32 reconstructed images per second in fixed and living cells. Single-molecule switching nanoscopy (SMSN) techniques localize single molecules with precisions of \~10 nm by stochastically switching molecules on and off 1-3. Thousands (or even tens of thousands) of camera frames of blinking subsets of molecules are typically recorded to obtain a single image at about 25-to 40-nm resolution. The temporal and spatial resolutions are limited by several factors: the number of photons emitted by a single molecule per frame and the sensitivity (quantum efficiency) and readout speed of the camera. Back-illuminated electron-multiplying charge-coupled devices (EMCCDs) are commonly used for SMSN because of their low effective readout noise. However, the noise introduced by the amplification process results in a lower signal-to-noise ratio by a factor of \~2 1/2 and effectively halves the high quantum efficiency (\~95\%) of these sensors to \textless48\% (ref. 4) unless each pixel detects generally less than one photon on average 5. Furthermore, the readout speed of 512 × 512 pixel-EMCCD cameras is currently limited to 70 full frames per second (f.p.s.), and typical acquisition times for SMSN range from minutes to an hour. Since the first demonstration of live-cell imaging with these techniques 6 , an important goal has been to improve this speed. Limiting imaging to smaller regions of interest (ROIs) of the chip or using cameras with fewer pixels can increase the frame rate to several hundred f.p.s. and allow live-cell SMSN of photoswitchable fluorescent proteins with a temporal resolution of several seconds, down to 500 ms with organic fluorophores 7,8. However, the accessible field of view (FOV) is usually severely compromised for temporal resolutions of \~1 s. This trade-off between imaging speed and FOV has substantially limited the capability of researchers using SMSN to address practical biological questions: many biological phenomena are so rare that imaging only small FOVs is highly inefficient, or the phenomena are best interpreted in the larger context of the cell (requiring imaging of large FOVs). This problem is of even greater concern when we consider the application of SMSN to high-throughput screening approaches. The trend toward systematic and quantitative analysis of cellular systems has not been expanded to nanoscopy, possibly because high recording speeds of large FOVs have been lacking, thus impeding the use of SMSN for systematic studies of complex cell biological processes. Recently, sCMOS cameras have been introduced that feature an effective quantum efficiency of up to 73\% at a wavelength of 600 nm, a large FOV (larger than 2,000 × 2,000 pixels) and much faster readout speeds than those of EMCCD cameras-characteristics that make sCMOS devices attractive alternatives to EMCCD cameras. Unfortunately, sCMOS camera architecture results in every pixel having a unique noise characteristic, and the noise variances of individual pixels can range from several to thousands of analog-to-digital units squared (Fig. 1a and Supplementary Fig. 1). The feasibility of SMSN using sCMOS cameras has been recently demonstrated 9,10 , but the non-negligible, pixel-dependent noise of sCMOS cameras makes the single-molecule localiza-tion algorithms originally designed for the Poisson-distributed and pixel-independent noise in EMCCD cameras 4,11,12 unable to provide reliable position estimates.

@article{Huang2013a, abstract = {newly developed scientific complementary metal-oxide semiconductor (scmos) cameras have the potential to dramatically accelerate data acquisition, enlarge the field of view and increase the effective quantum efficiency in single-molecule switching nanoscopy. however, scmos-intrinsic pixel-dependent readout noise substantially lowers the localization precision and introduces localization artifacts. We present algorithms that overcome these limitations and that provide unbiased, precise localization of single molecules at the theoretical limit. using these in combination with a multi-emitter fitting algorithm, we demonstrate single-molecule localization super-resolution imaging at rates of up to 32 reconstructed images per second in fixed and living cells. Single-molecule switching nanoscopy (SMSN) techniques localize single molecules with precisions of {\~{}}10 nm by stochastically switching molecules on and off 1-3. Thousands (or even tens of thousands) of camera frames of blinking subsets of molecules are typically recorded to obtain a single image at about 25-to 40-nm resolution. The temporal and spatial resolutions are limited by several factors: the number of photons emitted by a single molecule per frame and the sensitivity (quantum efficiency) and readout speed of the camera. Back-illuminated electron-multiplying charge-coupled devices (EMCCDs) are commonly used for SMSN because of their low effective readout noise. However, the noise introduced by the amplification process results in a lower signal-to-noise ratio by a factor of {\~{}}2 1/2 and effectively halves the high quantum efficiency ({\~{}}95{\%}) of these sensors to {\textless}48{\%} (ref. 4) unless each pixel detects generally less than one photon on average 5. Furthermore, the readout speed of 512 × 512 pixel-EMCCD cameras is currently limited to 70 full frames per second (f.p.s.), and typical acquisition times for SMSN range from minutes to an hour. Since the first demonstration of live-cell imaging with these techniques 6 , an important goal has been to improve this speed. Limiting imaging to smaller regions of interest (ROIs) of the chip or using cameras with fewer pixels can increase the frame rate to several hundred f.p.s. and allow live-cell SMSN of photoswitchable fluorescent proteins with a temporal resolution of several seconds, down to 500 ms with organic fluorophores 7,8. However, the accessible field of view (FOV) is usually severely compromised for temporal resolutions of {\~{}}1 s. This trade-off between imaging speed and FOV has substantially limited the capability of researchers using SMSN to address practical biological questions: many biological phenomena are so rare that imaging only small FOVs is highly inefficient, or the phenomena are best interpreted in the larger context of the cell (requiring imaging of large FOVs). This problem is of even greater concern when we consider the application of SMSN to high-throughput screening approaches. The trend toward systematic and quantitative analysis of cellular systems has not been expanded to nanoscopy, possibly because high recording speeds of large FOVs have been lacking, thus impeding the use of SMSN for systematic studies of complex cell biological processes. Recently, sCMOS cameras have been introduced that feature an effective quantum efficiency of up to 73{\%} at a wavelength of 600 nm, a large FOV (larger than 2,000 × 2,000 pixels) and much faster readout speeds than those of EMCCD cameras-characteristics that make sCMOS devices attractive alternatives to EMCCD cameras. Unfortunately, sCMOS camera architecture results in every pixel having a unique noise characteristic, and the noise variances of individual pixels can range from several to thousands of analog-to-digital units squared (Fig. 1a and Supplementary Fig. 1). The feasibility of SMSN using sCMOS cameras has been recently demonstrated 9,10 , but the non-negligible, pixel-dependent noise of sCMOS cameras makes the single-molecule localiza-tion algorithms originally designed for the Poisson-distributed and pixel-independent noise in EMCCD cameras 4,11,12 unable to provide reliable position estimates.}, added-at = {2020-04-06T13:11:22.000+0200}, author = {Huang, Fang and Hartwich, Tobias M P and Rivera-Molina, Felix E and Lin, Yu and Duim, Whitney C and Long, Jane J and Uchil, Pradeep D and Myers, Jordan R and Baird, Michelle A and Mothes, Walther and Davidson, Michael W and Toomre, Derek and Bewersdorf, Joerg}, biburl = {https://www.bibsonomy.org/bibtex/28385407054d7155ff4c4cb7e2e9e5cf4/kfriedl}, doi = {10.1038/Nmeth.2488}, file = {::}, interhash = {f507a3c8604952fc7baa1555cf15318c}, intrahash = {8385407054d7155ff4c4cb7e2e9e5cf4}, journal = {Articles nAture methods}, keywords = {cameras detectors imaging microscopy scmos smlm superresolution}, number = 653, timestamp = {2020-04-06T14:37:30.000+0200}, title = {{Video-rate nanoscopy using scmos camera-specific single-molecule localization algorithms}}, volume = 10, year = 2013 }