Single-Molecule Localization Microscopy  •  Software Benchmarking

Collection of reference datasets

The benchmarking of SMLM software package mainly relies on the usage of common reference datasets, in particular synthetic datasets with known ground-truth.

Conditions of use These reference datasets are designed to be largely used by the developpers to validate software and by the users to check a software. Reference: Sage et al. Quantitative evaluation of software packages for SMLM, Nature Methods 12, 2015.

banner

Real Experiments

The experimental datasets consist of a sequence of frames from real stained biologicial samples. The standard acquisition parameters are also provided.

picture1
Tubulin ConjAL647

Experimental sequence of 27'529 frames (128x128 pixels) with the parameters of acquisition.

Reference: Suliana Manley, Julia Gunzenhäuser and Nicolas Olivier, A starter kit for point-localization super-resolution imaging, Current Opinion in Chemical Biology 15, 2011.

Tubulins • Long Sequence

Experimental sequence of 1'500 frames with the parameters of acquisition
High density of fluorophores per frame.

Courtesy of Nicolas Olivier and Debora Keller, LEB, EPFL

picture1
Tubulins • High Density

Experimental sequence of 500 frames with the parameters of acquisition
High density of fluorophores per frame.

Courtesy of Nicolas Olivier and Debora Keller, LEB, EPFL

picture1
Tubulin AF647

The dataset represents a fixed cell, stained with mouse anti-alpha-tubulin primary antibody and Alexa647 secondary antibody. The intermittent increase in signal is due reactivation with a 405 nm laser.
This dataset is an experimental sequence of 9990 frames of 128x128 pixels.

Courtesy of Nicolas Olivier and Suliana Manley, LEB, EPFL

Realistic Simulations • Challenge 2013

These datasets are simulated sequence of realistic frames. The bio-inspired samples are 3D continuous models that imitates biological structures. These datasets were used for the challenge 2013.

Tubulins_I 2D

Published February 1, 2013 as training dataset for the challenge 2013

banner

This sample consists to a realistic structure of 7 tubulins (constant diameter 25 nm) and 1 tubulin (constant diameter 40 nm).
The depth of the sample is from 0 to 300 nm.
100000 fluorophores are activated over 2400 frames.
Low level of read-out noise autofluorescence background.

Tubulins_II 2D

Published February 1, 2013 as training dataset for the challenge 2013

banner

This sample consists to a realistic structure of 7 tubulins (constant diameter 25 nm) and 1 tubulin (constant diameter 40 nm).
The depth of the sample is from 0 to 300 nm.
100000 fluorophores are activated over 2400 frames.
High level of read-out noise and autofluorescence background.

Bundled_Tubes_Long_Sequence 2D

Published February 2, 2013 as training dataset for the challenge 2013

banner

Bundles of 8 tubes of 30 nm diameter
Sparse density, 81049 molecules on 12000 frames

Bundled_Tubes_High_Density 2D

Published February 2, 2013 as training dataset for the challenge 2013

banner

Bundles of 8 tubes of 30 nm diameter
High-density, 81049 fluorophores on 168 frames

Contest Dataset 1 • Low Density (LS) 2D

Published February 2, 2013 as contest dataset for the challenge 2013

banner

Tubulins of various diameters
10'000 frames

Contest Dataset 1 • High Density (HD) 2D

Published February 2, 2013 as contest dataset for the challenge 2013

banner

Tubulins of various diameters
1'000 frames • high-density of fluorophores

Contest Dataset 2 • Low Density (LS) 2D

Published February 2, 2013 as contest dataset for the challenge 2013

banner

Network of tubulins
12000 frames

Contest Dataset 2 • High Density (HD) 2D

Published February 2, 2013 as contest dataset for the challenge 2013

banner

Network of tubulins
204 frames • high-density of fluorophores

Contest Dataset 3 • Low Density (HD) 2D

Published February 2, 2013 as contest dataset for the challenge 2013

banner

Helicoidal tubes, deep sample from 0 to 1μm
7000 frames

Contest Dataset 3 • High Density (HD) 2D

Published February 2, 2013 as contest dataset for the challenge 2013

banner

Helicoidal tubes, deep sample from 0 to 1μm
600 frames • high-density of fluorophores

Realistic Simulations • Challenge 2016

These datasets are simulated sequence of realistic frames. The bio-inspired samples are 3D continuous models that imitates biological structures. These datasets were used for the challenge 2016.

MT1.N1.LD AS • DH • BP

Published June 7, 2016 (DH updated October, 25 2016) as contest dataset for the challenge 2016

banner

Experimental conditions
  • Structure: MT1, 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 19'996 frames
  • Modality: 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N1: typical photon counts and background levels for Alexa647 labelled STORM sample
  • Molecule density: 0.2

MT2.N1.HD AS • DH • BP

Published June 7, 2016 (DH updated October, 25 2016) as contest dataset for the challenge 2016

banner

Experimental conditions
  • Structure: MT2, 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 3'125 frames
  • Modality: 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N1: typical photon counts and background levels for Alexa647 labelled STORM sample
  • Molecule density: 2

MT3.N2.LD 2D • AS • DH • BP

Published June 7, 2016 (DH updated October, 25 2016) as contest dataset for the challenge 2016

banner

Experimental conditions
  • Structure: MT3, 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 3'020 frames
  • Modality: 2D, 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N2: photoswitchable fluorescent protein labelled sample such as mEos2 or Dendra2
  • Molecule density: 0.2

MT4.N2.HD 2D • AS • DH • BP

Published June 7, 2016 (DH updated October, 25 2016) as contest dataset for the challenge 2016

banner

Experimental conditions
  • Structure: 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 20'000 frames
  • Modality: 2D, 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N2: photoswitchable fluorescent protein labelled sample such as mEos2 or Dendra2
  • Molecule density: 2

ER1.N3.LD 2D

Published June 7, 2016 (DH updated October, 25 2016) as contest dataset for the challenge 2016

banner

Experimental conditions
  • Structure: simulation of cellular organelle (endoplasmic reticulum/mitochondria) inspired structure in the field of view of 6.4 x 6.4 x 0.7 μm
  • Sequence: 19'620 frames
  • Modality: 2D
  • Noise N3: typical photon counts and level of background for site specific dye-labelled live cell STORM sample such as ER Tracker
  • Molecule density: 0.2

ER2.N3.HD 2D

Published June 7, 2016 (DH updated October, 25 2016) as contest dataset for the challenge 2016

banner

Experimental conditions
  • Structure: simulation of cellular organelle (endoplasmic reticulum/mitochondria) inspired structure in the field of view of 6.4 x 6.4 x 0.7 μm
  • Sequence: 3'020 frames
  • Modality: 2D
  • Noise N3: typical photon counts and level of background for site specific dye-labelled live cell STORM sample such as ER Tracker
  • Molecule density: 5

MT0.N1.LD 2D • AS • DH • BP

Published June 7, 2016 as training dataset for the challenge 2016

banner

Experimental conditions
  • Structure: MT0, 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 19'996 frames
  • Modality: 2D, 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N1: typical photon counts and background levels for Alexa647 labelled STORM sample
  • Molecule density: 0.2

MT0.N1.HD 2D • AS • DH • BP

Published June 7, 2016 as training dataset for the challenge 2016

banner

Experimental conditions
  • Structure: MT0, 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 2'500 frames
  • Modality: 2D, 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N1: typical photon counts and background levels for Alexa647 labelled STORM sample
  • Molecule density: 2

MT0.N2.LD 2D • AS • DH • BP

Published June 7, 2016 as training dataset for the challenge 2016

banner

Experimental conditions
  • Structure: MT0, 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 19'996 frames
  • Modality: 2D, 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N2: photoswitchable fluorescent protein labelled sample such as mEos2 or Dendra2
  • Molecule density: 2

MT0.N2.HD 2D • AS • DH • BP

Published June 7, 2016 as training dataset for the challenge 2016

banner

Experimental conditions
  • Structure: MT0, 3 microtubules in the field of view of 6.4 x 6.4 x 1.5 μm
  • Sequence: 2'500 frames
  • Modality: 2D, 3D-Astigmatism, 3D-Double-Helix, 3D-Biplane
  • Noise N2: photoswitchable fluorescent protein labelled sample such as mEos2 or Dendra2
  • Molecule density: 0.2

Artificial Constructions

These datasets represents artificial structure with some actived fluorophores at some specific postions. These datasets are mainly used to evaluated particular features of the localization software, like the ability to discriminate two neighboring fluorophores in the same frame.

picture1
Eye

This dataset consists to 4 non-overlapping tubes of 1nm of radius in a field of view 38.4 x 38.4 μm. Only 300 fluorophores are activated in 40 frames. It is an ideal dataset: no perturbation, no additional noise.

picture1
Snow

This dataset contains fluorophores are placed on a regular grid with various depth and various number of photons. The field of view is 50 μm x 50 μm

picture1
Seashell

The synthetic sample has been created articifically using a 3D artificial structure shape. This shape represents a seashell of six branches with various fluorophore densities. This sample is provided in two versions: flat sample (no Z) and thick sample (1 μm of depth).

© 2017 Biomedical Imaging Group, Ecole Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
Last update: 31 Mar 2017