| A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z | AA | AB | AC | AD | AE | AF | AG | AH | AI | AJ | AK | AL | AM | AN | AO | AP | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | Microscopy | |||||||||||||||||||||||||||||||||||||||||
2 | ||||||||||||||||||||||||||||||||||||||||||
3 | Technique | resolution | number of colors | sensitivity | throughput | limitations | links | experts in 4DN | comments | |||||||||||||||||||||||||||||||||
4 | xy* (nm) | z* (nm) | time/frame | per image/cell (seconds) | ||||||||||||||||||||||||||||||||||||||
5 | Diffraction-limited fluorescence | User | Expert | Future | Comments | User | Expert | Future | Comments | User | Expert | Future | Comments | User | Expert | Future | Comments | User | Expert | Future | Comments | |||||||||||||||||||||
6 | Widefield | 250* | 100** | likely no substantial change | *Abbe limit **Deconvolution Superresolution techniques are better at going below these resolution limits | 1200* | 200** | likely no substantial change | *Abbe limit **Deconvolution 2π microscopes increase resolution in z | ms | commonly 4 | no theoretical limit | Spectral karyotyping and multiplex FISH techniques allow use of many colors | fuorophore dependent. Development of brighter, bleach resistant fluors will continue to increase sensitivity | bleaching limited - depends on initial intensity | Best to have at least 2X signal/noise | Deconvolution cannot remove noise. | Joan Ritland* (Groudine Lab, Fred Hutch, Seattle), Grunwald, Singer | Deconvolution has been used to increase resolution (remove blur) of widefield 3D images for over 30 years and requires no microscope modifications. | |||||||||||||||||||||||
7 | Widefield | 300 | <250 | Diffraction limit | Widefield imaging is diffraction limited, but often the actually achieved resolution is lower due to instrument limitations. Values are for GFP like dyes. | 900 | <900 | Deconvolution, simultaneous 3D image acquisition, aberration control | Widefield imaging is diffraction limited, but often the actually achieved resolution is lower due to instrument limitations. Values are for GFP like dyes. | 0.1 | 0.01 | Detector noise analysis, labeling, SNR | The major limit is phototoxicity as it limits the amount of light that can be applied. | 3 | 4-5 (primary) | unlimited | Today most commercial widefield systems have 3 channels, normally to be used sequentially. Commercial adaptors to run multiple channels simulatnously are available. If samples are fixed a re-labeling strategy (stain, image, remove label, stain image) can overcome this limit today | single molecule. Limits: dyes, non-specific labeling, phototoxicity. Another component is certainly the target availability and density. | 1-50 images per stack per cell | 1-50 images per stack per cell | 1 3D encoded image per cell/time point | Proof of concept papers show that imaging volumes can allow 0.1 s stacked resolution. | The limits for fixed cells are similar to stochastic imaging approaches, for living cells its phototoxicity and label stability. Limits often depend intimately on the application | David Grunwald*, Joerg Bewersdorf, Warren Zipfel, Rob Singer, Rob Coleman, Jonas Ries | A simple and versitail method frequently used for imaging fixed and life samples. Performance can be pushed substaintially by experts, however if not calibrated the method frequently performs below expectation. As in stochastical imaging localization of point like source is possible. As widefield is often applied to living cells the time resolution, excitation power limits and labeling density are to be considered. Localization is better than the resolution (Standard: xy <100, z<900, expert: xy: <25 expert<900). For localization in z the data structure and analysis methods become more important then in 2D. Future limits for localization: xy: SNR, analysis methods, life cell suitability, z: Deconvolution, simultaneous 3D image acquisition, aberration control, SNR, analysis methods, life cell suitability. | |||||||||||||||||
8 | Confocal | 750 | 30 ms | ~6 | Ritland / | |||||||||||||||||||||||||||||||||||||
9 | airy-scan | 170 | 120- 140 | This strongly depends on the modality: SR, RS,.. CO | 350 | 27 fps (fast mode at 480x480) | Strongly depends on your system to be imaged | 1-2 | 2-4 | 4 – 8× improvement of signal-to-noise (SNR) compared to conventional confocal setup. Should be suitable for FCS analysis (GAsP detectors). Thus probably few molecules | To achieve a good S/N ratio, the scanning speed needs to be adjusted. Thus, even if fast scanning is possible, such as stated by the company- 27fps(at 480 x 480 pixels – it does not assure your structures to be clearly resolved and detected in high quality, especially if diffraction limited and having low fluorophore copy numbers. | https://www.zeiss.com/microscopy/int/products/confocal-microscopes/lsm-880-with-airyscan-.html, https://www.nature.com/articles/nmeth.f.388 | Jan Ellenberg, Arina Rybina* (EMBL), Zipfel | For Zeiss Airy Scan LSM 880. Having one Array detector, one can image 1 color at time, however choosing line by line scanning you can achieve almost simultaneous imaging of 2 colors. Extension to 3-4 colors also possible, but will be accompanied by slower acquisition. But, “Airyscanning works for any dye. | ||||||||||||||||||||||||||||
10 | Spinning Disk | 250 | Diffraction limit | Depends on Objective NA, light wavelength | 600-1500 | Depends on Objective NA, light wavelength, imaging medium refractive index, pinhole size | 100 - 500 ms | 20 ms | Depends on signal intensity | 4 | 4 | Sequential and combinatorial labelling can expand the number of "channels" indefinitely | Back thinned sCMOS camera technology with high QE (95%) can increase the amount of signal detected. Fluorophore dependent. | up to ~ 300 cells/second in 2D, up to ~ 40 cells/second in 3D | Estimates based on a 40X objective and a 2550 * 2160 sCMOS camera, single plane acquisition, single channel | SDCM axial resolution is lower than laser scanning confocal microscopy. | Gianluca Pegoraro* (NCI/Misteli), Several others | SDCM is extremely well suited for rapid 3D live cell imaging and excels in high-throughput imaging applications. SDCM is now a mature technique and it is standard in a majority of microscopy cores | ||||||||||||||||||||||||
11 | Light-sheet | 300 | 1000 | 5 ms | ~3 | |||||||||||||||||||||||||||||||||||||
12 | Lattice light-sheet | 370* | 290* | 80 | Combination with superresolution approaches (e.g. non-linear SIM, RESOLFT) will push the resolution | 950* | 650* | 120 | 20 ms | 10 ms | 1 ms | depends on fluorescent probes | 4 | 5 | 15 | single fluorophores | 10 s / 3D stack | 5 s | 1 s | rather low NA (1.1) in the detection, complex instrument | http://science.sciencemag.org/content/346/6208/1257998 https://www.nature.com/articles/nature22369 http://science.sciencemag.org/content/360/6386/eaaq1392 | Robin Diekmann* (Ries Lab, EMBL) | *can be operated in SIM mode with double the resolution Low phototoxicity, low background, specifically tailored to live-cell imaging | |||||||||||||||||||
13 | Single-particle tracking | 50* 20* | 30* 10* | 30* 10* | Fluorescent proteins, organic dyes | -* 100* | 80* 40* | 60* 25* | Fluorescent proteins, organic dyes | 30-100 ms | 10 ms | 1 ms | Multi-timescale experiments may be required | 1 | 2 | 3+ | Single dyes, limited by unspecific background and labelled particle concentration | 10 min | minutes | minutes, but parallelized | Blinking of fluorophores leads to undesirable gaps in trajectories. Labelled particles need to be sparse to avoid ambiguous trajectory assignments, often prohibiting complete labelling of a protein population or requiring photoactivatable labels. Multiple labels per particle increase photon budget but may conflict with the sparsity criterion. Trajectory length limited by photobleaching and/or diffusion out of focal plane. | Review papers https://pubs.acs.org/doi/pdf/10.1021/acs.chemrev.6b00815 https://www.nature.com/articles/nmeth.2808 | Bewersdorf (Yale), Cisse (MIT), Jan-Hendrik Spille* (jhspille@mit.edu, Cisse lab), Darzacq (Berkeley), Grunwald (UMass) | Spatial resolution scales with 1/sqrt(number of detected photons per frame). Photon budget per fluorophore is distributed across multiple localizations per particle, effectively lowering the number of detected photons per localization. Thus, bright fluorophores (and low background) required. Information is extracted from thousands of trajectories comprising a few to dozens of localizations. *Multiply all spatial resolutions by sqrt(n), where n the number of time points in a trajectory. Typically, n = 3 – 50. | ||||||||||||||||||
14 | ||||||||||||||||||||||||||||||||||||||||||
15 | Superresolution | |||||||||||||||||||||||||||||||||||||||||
16 | SIM: Structured illumination | 120 | 100 | 60 | Abbe/2, unless non-linear effects are exploited | 700 | 300 | 150 | 1 s | 30 ms | 10 ms | 2 | 3 | 5 | requires high SNR to avoid reconstruction artifacts | requires good optics, data analysis can introduce artifacts | Singer / Zipfel | |||||||||||||||||||||||||
17 | STED | 25 | 25 | 5 | fixed cells. 35 for living cells | 80 | 80 | 5 | 25 nm (z-only), 80 x 80 x 80 isotropic in x,y,z | fast, ideal for live-cell. high local speed (i.e. the framerate scales inversely with the image size) | 1 | 3 | 5 | We routinely imaged 3 dyes, and 2 simultaneously (depends on # of depletion lasers) | single dyes. APD detection with highest detection sensitivity (>62% @ 680nm) | Minute | Seconds | Seconds | Good for live-cell imaging; as fast as a confocal microscope | STORM/PALM provide a very high localization precision, which translates into resolutions at slightly better than STED. | References on novel techniques to limit photo-bleaching/ photo-damage: Heine, Jörn, et al. Proceedings of the National Academy of Sciences (2017): 201708304. Staudt, Thorsten, et al. Optics express 19.6 (2011): 5644-5657. | Brian P. English*/ Robert H. Singer (Janelia/ Albert Einstein Medical College), Bewersdorf (Yale) | STED stands out from all other super-resolution techniques due to the fact that the optical hardware directly, without any further processing, produces a super-resolution image. Based on our own experience, including imaging of the diverse set of samples brought by other researchers at Janelia, NIH and elsewhere, we made the following observations: - Sample preparation and mounting is basically identical to standard confocal imaging. - Multiple fluorophores yield high quality results. - Wide coverage of imaging spectrum is possible due to the availability of several STED lasers. - Ability for long-time imaging of live specimens, operational modes can limit photobleaching. - Quick super-resolved image can be obtained, including in 3D. - Perfect registration of multiple channels is guaranteed by the STED beam structure. - Small file sizes result from the imaging sessions due to a super resolved final image. - No need for image post-processing, thus avoiding analysis artefacts. | |||||||||||||||||||
18 | interferometric | 25 | 25 | Bewersdorf (Yale) | ||||||||||||||||||||||||||||||||||||||
19 | live cell SMLM, SPT, imaging nascent transcription sites/TS and single RNA molecules | 20-40* 50* | 10-20* 30* | 6* 30* | organic dyes fluorescent proteins | 200-500* 300* | 50-100* 200* | 30* 150* | single plane imaging Z-stacks | sec-min* 50 ms** | ms-min* 35 ms** | ms | *SPT, **TS/RNA imaging. Future goal is to obtain new fluorophores and imaging modalities that allow fast time resolution over minutes to hours | 1 | 2-3 | 3-4 | SPT: Single dyes limited by background TS/RNA imaging: 48 fluorescent proteins bound to an RNA molecule | 30 min, 20 TC cells/5 hrs, 1 neuron /h | 20 min, 35 TC cells/5 hrs, 3 neuron /h | Depends on experiment, time resolution and length of movies At 2HZ frame rates 1 Z-stack/every minute | For organic fluorophores-bright non-blinking photostable dyes are limited particularly in near-infrared wavelengths, also limited to just Halo and SNAP tags for specific covalent attachment of dyes to proteins for multicolor imaging in same cell. We still have to budget photons for both long(tens of minutes-hours) and short (millisecond frame rates) term imaging | Links, references: Coleman RA, Liu Z, Darzacq X, Tjian R, Singer RH, Lionnet T. Imaging Transcription: Past, Present, and Future. Cold Spring Harb Symp Quant Biol. 2015;80:1-8. PubMed PMID: 26763984; PubMed Central PMCID: PMC4915995., Hansen AS, Woringer M, Grimm JB, Lavis LD, Tjian R, Darzacq X. Robust model-based analysis of single-particle tracking experiments with Spot-On. Elife. 2018 Jan 4;7. pii: e33125. doi:10.7554/eLife.33125. PMID:29300163. Vera M, Biswas J, Senecal A, Singer RH, Park HY. Single-Cell and Single-Molecule Analysis of Gene Expression Regulation. Annu Rev Genet. 2016 Nov 23;50:267-291. Review. PMID:27893965 | Robert Singer* (Albert Einstein College of Medicine,Robert.Singer@einstein.yu.edu), Robert Coleman* (Albert Einstein College of Medicine,Robert.Coleman2@einstein.yu.edu), Xavier Darzacq (Berkeley), Ibrahim Cisse (MIT), David Grunwald (UMASS Medical School), Brian English (HHMI, Janelia) | Bright organic fluorophores are required for single particle tracking (SPT) for an extended number of frames (millisecond or seconds time resolution), whereas fluorescent proteins can be used for imaging nascent transcription sites and single RNA molecules in live cells. Resolution estimates for live cell measurements can be complicated since labeled proteins are often freely diffusing or bound to genomic scaffolds that are also moving during time when individual frames are captured. In many cases throughput is also dictated by magnification, time resolution, time required to analyze movies. | ||||||||||||||||||
20 | SMLM/PALM/STORM | 50 20 | 30 10 | 30 6 | photoactivatable proteins organic dyes | none 100 | 80 35 | 60 25 | fluorescent proteins organic dyes | fixed cell | minutes | seconds | Time-lapse limited as most dyes are visible once | 1 | 2 | 2-3 | single fluorophores, limited by unspecific background | 10-60 min | minutes | seconds | Fast imaging only possible with organic dyes | Requires special ‘blinking’ fluorophores, either photoactivatable dyes or proteins or standard dyes in a special thiol-containing buffer. Techniques that provide z-resolution result in a loss in x,y resolution by a factor of ~1.5. Live-imaging severely limited by slow acquisition, phototoxicity (extremely high light levels) and loss in spatial resolution. | Review papers: http://www.annualreviews.org/doi/10.1146/annurev-biochem-060815-014801,http://science.sciencemag.org/content/361/6405/880 | Ries* (EMBL), Bewersdorf (Yale) | Resolution scales with 1/sqrt(number of detected photons). Thus, bright fluorophores (and low background) required. Live-cell SMLM very limited, reduced spatial resolution, high light-levels and photo toxicity | |||||||||||||||||
21 | interferometric SMLM | 40 20 | 30 10 | 20 4 | photoactivatable proteins organic dyes | 25 15 | 20 8 | 15 5 | fluorescent proteins organic dyes | fixed cell | minutes | seconds | 1 | 1 | 2 | single fluorophores, limited by unspecific background | 10-60 min | minutes | seconds | extremely complicated setup | http://linkinghub.elsevier.com/retrieve/pii/S0092867416307450, http://www.pnas.org/content/106/9/3125 | Bewersdorf (Yale), Ries* (EMBL) | ||||||||||||||||||||
22 | DNA-PAINT | 5 | 15 | fixed | unlimited (sequentially) | single fluorophores, but creates background | 1h - 16h per cell | hours | minutes | slow method, requires targets to be accesible easily by DNA-PAINT probes. Some unspecific labeling of cellular compartments. Resolution often limted by labeling (e.g. via antibodies) rather than by localization precision. | http://www.nature.com/doifinder/10.1038/nmeth.2835 | Ellenberg(EMBL) | Highest resolution of all methods due to the large brightness of the labels. | |||||||||||||||||||||||||||||
23 | Expansion microscopy | 100 | 50 | 30 | even better when combined with other superresolution methods | 250 | 100 | 60 | fixed only | 2 | 3 | unlimited | some dyes and labels do not survive expansion protocol | single fluorophores | fast | isotropic expansion on the nanometer scale still needs to be validated. | http://www.sciencemag.org/content/early/2015/01/14/science.1260088.full | Ries* (EMBL) | Swelling of fixed sample with subsequent imaging with diffraction limited or superresolution microscopy works surprisingly well on many samples. | |||||||||||||||||||||||
24 | MINFLUX | 10 | 5 | 1 | 750 | 30 | 3 | minutes | seconds | 0.1 sec | scales with size of FoV | 2 | 3 | 5 | single fluorophores | hours | minutes | seconds | best for small ROIs | requires photo-activatable probes. Very complex instrument, currently only working in the Hell lab. Commercial solution expected within 2 years | http://www.sciencemag.org/cgi/doi/10.1126/science.aak9913 | Ries* (EMBL) | Has the prospect of achieving nm 3D resolution in living cells, something not achievable by any other technique. Currently still in the proof-of-concept stage. | |||||||||||||||||||
25 | ||||||||||||||||||||||||||||||||||||||||||
26 | Electron microscopy | fixed | O'Shea / Ellisman | |||||||||||||||||||||||||||||||||||||||
27 | ||||||||||||||||||||||||||||||||||||||||||
28 | ||||||||||||||||||||||||||||||||||||||||||
29 | ||||||||||||||||||||||||||||||||||||||||||
30 | ||||||||||||||||||||||||||||||||||||||||||
31 | ||||||||||||||||||||||||||||||||||||||||||
32 | ||||||||||||||||||||||||||||||||||||||||||
33 | ||||||||||||||||||||||||||||||||||||||||||
34 | *Resolution: defined as FWHM | |||||||||||||||||||||||||||||||||||||||||
35 | ||||||||||||||||||||||||||||||||||||||||||
36 | ||||||||||||||||||||||||||||||||||||||||||
37 | ||||||||||||||||||||||||||||||||||||||||||
38 | ||||||||||||||||||||||||||||||||||||||||||
39 | ||||||||||||||||||||||||||||||||||||||||||
40 | ||||||||||||||||||||||||||||||||||||||||||
41 | ||||||||||||||||||||||||||||||||||||||||||
42 | ||||||||||||||||||||||||||||||||||||||||||
43 | ||||||||||||||||||||||||||||||||||||||||||
44 | ||||||||||||||||||||||||||||||||||||||||||
45 | ||||||||||||||||||||||||||||||||||||||||||
46 | ||||||||||||||||||||||||||||||||||||||||||
47 | ||||||||||||||||||||||||||||||||||||||||||
48 | ||||||||||||||||||||||||||||||||||||||||||
49 | ||||||||||||||||||||||||||||||||||||||||||
50 | ||||||||||||||||||||||||||||||||||||||||||
51 | ||||||||||||||||||||||||||||||||||||||||||
52 | ||||||||||||||||||||||||||||||||||||||||||
53 | ||||||||||||||||||||||||||||||||||||||||||
54 | ||||||||||||||||||||||||||||||||||||||||||
55 | ||||||||||||||||||||||||||||||||||||||||||
56 | ||||||||||||||||||||||||||||||||||||||||||
57 | ||||||||||||||||||||||||||||||||||||||||||
58 | ||||||||||||||||||||||||||||||||||||||||||
59 | ||||||||||||||||||||||||||||||||||||||||||
60 | ||||||||||||||||||||||||||||||||||||||||||
61 | ||||||||||||||||||||||||||||||||||||||||||
62 | ||||||||||||||||||||||||||||||||||||||||||
63 | ||||||||||||||||||||||||||||||||||||||||||
64 | ||||||||||||||||||||||||||||||||||||||||||
65 | ||||||||||||||||||||||||||||||||||||||||||
66 | ||||||||||||||||||||||||||||||||||||||||||
67 | ||||||||||||||||||||||||||||||||||||||||||
68 | ||||||||||||||||||||||||||||||||||||||||||
69 | ||||||||||||||||||||||||||||||||||||||||||
70 | ||||||||||||||||||||||||||||||||||||||||||
71 | ||||||||||||||||||||||||||||||||||||||||||
72 | ||||||||||||||||||||||||||||||||||||||||||
73 | ||||||||||||||||||||||||||||||||||||||||||
74 | ||||||||||||||||||||||||||||||||||||||||||
75 | ||||||||||||||||||||||||||||||||||||||||||
76 | ||||||||||||||||||||||||||||||||||||||||||
77 | ||||||||||||||||||||||||||||||||||||||||||
78 | ||||||||||||||||||||||||||||||||||||||||||
79 | ||||||||||||||||||||||||||||||||||||||||||
80 | ||||||||||||||||||||||||||||||||||||||||||
81 | ||||||||||||||||||||||||||||||||||||||||||
82 | ||||||||||||||||||||||||||||||||||||||||||
83 | ||||||||||||||||||||||||||||||||||||||||||
84 | ||||||||||||||||||||||||||||||||||||||||||
85 | ||||||||||||||||||||||||||||||||||||||||||
86 | ||||||||||||||||||||||||||||||||||||||||||
87 | ||||||||||||||||||||||||||||||||||||||||||
88 | ||||||||||||||||||||||||||||||||||||||||||
89 | ||||||||||||||||||||||||||||||||||||||||||
90 | ||||||||||||||||||||||||||||||||||||||||||
91 | ||||||||||||||||||||||||||||||||||||||||||
92 | ||||||||||||||||||||||||||||||||||||||||||
93 | ||||||||||||||||||||||||||||||||||||||||||
94 | ||||||||||||||||||||||||||||||||||||||||||
95 | ||||||||||||||||||||||||||||||||||||||||||
96 | ||||||||||||||||||||||||||||||||||||||||||
97 | ||||||||||||||||||||||||||||||||||||||||||
98 | ||||||||||||||||||||||||||||||||||||||||||
99 | ||||||||||||||||||||||||||||||||||||||||||
100 | ||||||||||||||||||||||||||||||||||||||||||