Super-resolution fight club: A broad assessment of 2D &amp; 3D single-molecule localization microscopy software

Sage, Daniel; Pham, Thanh-an; Babcock, Hazen P.; Lozano-Pérez, Tomás; Pengo, Thomas; Chao, Jerry; Velmurugan, Ramraj; Herbert, Alex; Agrawal, Anurag; Colabrese, Silvia; Wheeler, Ann; Archetti, Anna; Rieger, Bernd; Ober, Raimund J.; Gm, Hagen; Sibarita, Jean‐Baptiste; Ries, Jonas; Henriques, Ricardo; Unser, Michaël; Holden, Séamus

doi:10.1101/362517

Cited by 19 publications

(27 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another observation concerns the fact that the Double-Helix can reach a better axial than lateral RMSE. This fact, which was also observed in the recent SMLM challenge [74], can be explained by the large lateral support of the Double-Helix PSF as well as its good axial discrimination.…”

Section: Discussionsupporting

confidence: 69%

“…In particular, off-the-grid sparse regularized methods have shown their efficiency for high density settings [55,10]. For a complete review and comparisons of existing methods, we refer the reader to the two recent SMLM challenges [73,74].…”

Section: Single Molecule Localization Microscopymentioning

confidence: 99%

“…In order to assess the quality of the reconstructed volumes, we consider standard metrics which reflect both the detection rate and the localization error [74,73]. Given a recovered frame and a tolerance radius r > 0, we pair estimated molecules and ground truth (GT) molecules when the distance between them is lower than r. Paired estimated molecules are then referred as true positive (TP) while unpaired ones as false positive (FP).…”

Section: Metrics For Evaluationmentioning

confidence: 99%

See 2 more Smart Citations

The sliding Frank–Wolfe algorithm and its application to super-resolution microscopy

Denoyelle¹,

Duval²,

Peyré³

et al. 2019

Inverse Problems

167

View full text Add to dashboard Cite

This paper showcases the theoretical and numerical performance of the Sliding Frank-Wolfe, which is a novel optimization algorithm to solve the BLASSO sparse spikes super-resolution problem.The BLASSO is a continuous (i.e. off-the-grid or grid-less) counterpart to the well-known 1 sparse regularisation method (also known as LASSO or Basis Pursuit). Our algorithm is a variation on the classical Frank-Wolfe (also known as conditional gradient) which follows a recent trend of interleaving convex optimization updates (corresponding to adding new spikes) with non-convex optimization steps (corresponding to moving the spikes). Our main theoretical result is that this algorithm terminates in a finite number of steps under a mild non-degeneracy hypothesis. We then target applications of this method to several instances of single molecule fluorescence imaging modalities, among which certain approaches rely heavily on the inversion of a Laplace transform. Our second theoretical contribution is the proof of the exact support recovery property of the BLASSO to invert the 1-D Laplace transform in the case of positive spikes. On the numerical side, we conclude this paper with an extensive study of the practical performance of the Sliding Frank-Wolfe on different instantiations of single molecule fluorescence imaging, including convolutive and non-convolutive (Laplace-like) operators. This shows the versatility and superiority of this method with respect to alternative sparse recovery technics.

show abstract

Section: Discussionsupporting

confidence: 69%

Section: Single Molecule Localization Microscopymentioning

confidence: 99%

Section: Metrics For Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

The sliding Frank–Wolfe algorithm and its application to super-resolution microscopy

Denoyelle¹,

Duval²,

Peyré³

et al. 2019

Inverse Problems

167

View full text Add to dashboard Cite

show abstract

“…This limitation can be overcome by suitable reference samples, which allow a quantitative characterization and systematic optimization of a superresolution microscopy workflow. This reference-based approach has been used successfully to benchmark various superresolution analysis software packages, where simulated synthetic images served as reference standards 9 . In the past, both artificial and cellular structures have been used as physical references.…”

Section: Introductionmentioning

confidence: 99%

Nuclear pores as versatile reference standards for quantitative superresolution microscopy

Thevathasan¹,

Kahnwald²,

Cieslinski³

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

Quantitative fluorescence and superresolution microscopy are often limited by insufficient data quality or artifacts. In this context, it is essential to have biologically relevant control samples to benchmark and optimize the quality of microscopes, labels and imaging conditions.Here we exploit the stereotypic arrangement of proteins in the nuclear pore complex as in situ reference structures to characterize the performance of a variety of microscopy modalities. We created four genome edited cell lines in which we endogenously labeled the nucleoporin Nup96 with mEGFP, SNAP-tag or HaloTag or the photoconvertible fluorescent protein mMaple. We demonstrate their use a) as 3D resolution standards for calibration and quality control, b) to quantify absolute labeling efficiencies and c) as precise reference standards for molecular counting.These cell lines will enable the broad community to assess the quality of their microscopes and labels, and to perform quantitative, absolute measurements.

show abstract

“…In this context, NanoJ has been designed for high-performance image analysis, using GPU computing, ensuring the quick processing of large data volumes (44)(45)(46)(47)(48). Further, NanoJ's modular and open-source nature and its integration within ImageJ allow it to be used in conjunction with other analysis software packages (49)(50)(51)(52). We are continuously supporting, adapting and expanding the framework to include new approaches, such as 3D imaging.…”

Section: Nanoj-fluidicsmentioning

confidence: 99%

NanoJ: a high-performance open-source super-resolution microscopy toolbox

Laine

Tosheva

Gustafsson

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

Super-resolution microscopy has become essential for the study of nanoscale biological processes. This type of imaging often requires the use of specialised image analysis tools to process a large volume of recorded data and extract quantitative information. In recent years, our team has built an open-source image analysis framework for super-resolution microscopy designed to combine high performance and ease of use. We named it NanoJ - a reference to the popular ImageJ software it was de-veloped for. In this paper, we highlight the current capabilities of NanoJ for several essential processing steps: spatio-temporal alignment of raw data (NanoJ-Core), super-resolution image re-construction (NanoJ-SRRF), image quality assessment (NanoJ-SQUIRREL), structural modelling (NanoJ-VirusMapper) and control of the sample environment (NanoJ-Fluidics). We expect to expand NanoJ in the future through the development of new tools designed to improve quantitative data analysis and measure the reliability of fluorescent microscopy studies.

show abstract

Super-resolution fight club: A broad assessment of 2D & 3D single-molecule localization microscopy software

Cited by 19 publications

References 48 publications

The sliding Frank–Wolfe algorithm and its application to super-resolution microscopy

The sliding Frank–Wolfe algorithm and its application to super-resolution microscopy

Nuclear pores as versatile reference standards for quantitative superresolution microscopy

NanoJ: a high-performance open-source super-resolution microscopy toolbox

Contact Info

Product

Resources

About