Clone wiki

NanoJ-SRRF / Home

Super-Resolution Radial Fluctuations (SRRF) - ImageJ Plugin

News: we've migrated the source code repository to github, you can find the latest source here.

Check out the SRRF paper in Nature Communications or our SMLMS2016 short talk about NanoJ and SRRF.

We have now developed SRRF-Stream in partnership with Andor Technology. SRRF-Stream allows real-time analysis of data being acquired in Andor cameras. Check out our webinar on NanoJ-SRRF and SRRF-Stream.

SRRF (pronounced as surf) is as a novel open-source and high-performance analytical approach for Live-cell Super-Resolution Microscopy, provided as a fast GPU-enabled ImageJ plugin. SRRF is capable of extracting high-fidelity super-resolution information in modern microscopes (TIRF, widefield and confocals) using conventional fluorophores such as GFP. Compared to other methods, SRRF is capable of live-cell imaging over timescales ranging from minutes to hours, using sample illumination orders of magnitude lower than methods such as PALM, STORM or STED.

Actin labelled with LifeAct-GFP and imaging performed on a standard TIRF microscope. Each SRRF frame was produced by running SRRF analysis on 100 frames of the raw TIRF acquisition; for comparison the average of these 100 TIRF frames is displayed on the left.

Tutorials on how to get started

To get set up with SRRF please click here to find tutorials on how to install, run and compile NanoJ-SRRF.

Hall of Fame

Here is a list of researchers that have either posted data analysed with SRRF or given us feedback that helped improve the algorithm.

About SRRF

SRRF is part of the NanoJ project - a collection of analytical methods dedicated to super-resolution and advanced imaging compatible with ImageJ. Both SRRF and NanoJ are developed by the Henriques laboratory in the MRC Laboratory for Molecular Cell Biology at University College London.

Microtubules labelled with tubulin-GFP and imaging performed on a spinning disk confocal microscope. Z sections were taken at 300nm intervals, with each SRRF image produced from SRRF analysis of 100 raw frames acquired at each interval; for comparison the average of these 100 spinning disk frames is displayed on the left. The right hand panel shows the cumulative projections of the SRRF and spinning disk images colour-coded by z position.


  • Super-resolution with standard microscopesSRRF is capable of super-resolving cellular structures imaged with widefield, TIRF or confocal modern microscopes without the need for specialized optics. Additionally, SRRF has sample illumination intensity requirements orders of magnitude lower than other super-resolution methods such as PALM, STORM or STED.
  • Super-resolution with conventional fluorophores such as GFP: we have shown that SRRF is able to produce super-resolution images from samples labelled with a wide range of conventional fluorophores, such as GFP.
  • Live-cell super-resolution with minimal phototoxicity: as SRRF is able to extract high-fidelity super-resolution information from low signal-to-noise ratio samples, it requires lower sample illumination than most other super-resolution methods. For this reason, SRRF enables live-cell imaging over timescales ranging from minutes to hours. Imaged cells generally remain capable of undergoing mitosis, mitochondrial motility and cytoskeletal reorganisation as expected in normal healthy conditions.
  • Speed: SRRF is very fast!! It has been fully optimized to take advantage of GPU high-performance computing in modern graphics cards. However, its analytical framework has been developed to work in almost any computer, independently of its architecture. SRRF, generally will process images and generate super-resolution data in real-time.
  • Drift correction: drift is a major challenge in super-resolution microscopy and in most cases the limiting factor for resolution. Since the acquisition can take from several minutes to hours, drifts of even a few tens of nanometers can drastically deteriorate resolution, or worse, create anomalies in reconstructed images (e.g. artifactual doubling of filamentary structures). SRRF provides an easy drift correction method based on dynamics sample tracking (cross-correlation).
  • Availability, ease of use and open-source: both the software and its source code are freely available as a Fiji or ImageJ plugin. This allows maximal dissemination of the software in the biological research community, optimal usability, and will offer users the ability to modify and improve the software at will.

SRRF and QuickPALM

SRRF follows the ideology of our previous algorithm QuickPALM published in Nature Methods, QuickPALM remains one of the most cited analytical packages for PALM and STORM super-resolution since 2010, provided freely to the community. However SRRF goes far beyond the first generation of analytical approaches, of which QuickPALM is part of, allowing for Super-Resolution information to be extracted from almost any modern microscope, including those non-specialised for super-resolution.

Microtubules labelled with GFP and imaging is performed on a standard TIRF microscope. Each SRRF frame was produced by running SRRF analysis on 100 frames of the raw TIRF acquisition; for comparison the average of these 100 TIRF frames is displayed above.