Adaptive image reconstruction for high-fidelity, fast and easy-to-use 3D live-cell super-resolution microscopy
eSRRF (enhanced Super-Resolution Radial Fluctuations) is an extension of the SRRF method developed by the Henriques lab, described in Gustafsson et al. (2016). For more details check out our preprint on bioRXiv.
eSRRF aims at improving the fidelity of SRRF images with respect to the underlying true structure. Below is shown a representative dataset obtained from high-density emitters for which the underlying structure was obtained via DNA-PAINT (SMLM).
The (e)SRRF approach is based on
- A spatial analysis of the high-density emitter data using Radial symmetry transform;
- A temporal analysis of the obtained temporal stack using similar appraoch to SOFI.
Some of the new features available in eSRRF include:
- Improved fidelity of reconstructions;
- Adaptive reconstruction schemes allowing to explore the compromise between fidelity and resolution. This is enabled by an integration of our SQUIRREL approach, described in Culley et al. (2018);
- Estimation of the maximum number of frames to use for eSRRF analysis from a dataset, based on SSIM (or optical flow magnitude) calculation;
- Rolling analysis;
- Integrated drift correction;
- Better memory management;
- Full OpenCL integration, enabling GPU acceleration;
- Direct saving to disk for large dataset analysis;
- eSRRF 3D reconstruction!
The latest stable version of eSRRF can be directly obtained from our Fiji update site: https://sites.imagej.net/NanoJ-LiveSRRF/
Information about update sites can be found here.
Video guide: Installation
There have been some issues reported with OpenCl and running NanoJ-Squirrel and NanoJ-eSRRF on Windows10/11. You can find an instruction with the temporal fix in the wiki.
✨ ✨ Update November 2023 ✨ ✨:
eSRRF is now also available in Python 🐍! You'll find the code and notebooks here: 👉 https://github.com/HenriquesLab/NanoPyx
✨ ✨ ✨ ✨ ✨ ✨ ✨ ✨ ✨ ✨ ✨ ✨
We have published test datasets including eSRRF parameter suggestions on Zenodo. Download and get started right away!
Video guide: Getting started
)
eSRRF comes packed with useful Tools plugins to perform a range of things, such as (but not limited to):
- Fluorescence fluctuation simulator;
- Rescale individual slices within a stack and convert it to RGB (useful to visualise the parameter sweep output);
- Save all current open images as Tiff files;
- Perform linear rescaling on stack;
Many people are involved in developing and testing this method, here are some of the key players:
- Romain F. Laine (@LaineBioImaging)
- Ricardo Henriques (@HenriquesLab)
- Guillaume Jacquemet (@guijacquemet)
- Christophe Leterrier (@christlet)
- Siân Culley (@SuperResoluSian)
- Bassam Hajj (@Bassam_A_HAJJ)
- Hannah S. Heil (@Hannah_SuperRes)
- Simao Coelho (@simaopc)
- Jonathon Nixon-Abell (@AbellJonny)
- Angélique Jimenez
- Tommaso Galgani
- Aki Stubb (@akistub)
- Gautier Follain (@Follain_Ga)
- Samantha Webster
- Jesse Goyette
Exciting news! enhanced Super-Resolution Radial Fluctuations (eSRRF) is now accessible in Python through the NanoPyx package. This integration brings the power and versatility of eSRRF to Python users, opening up new possibilities for analysis and integration within Python-based workflows.
NanoPyx seamlessly integrates eSRRF capabilities into Python environments. With NanoPyx, users can now leverage eSRRF's high-performance analytical approach within their Python scripts, pipelines, and interactive sessions. Through NanoPyx, eSRRF is also available as "codeless" Jupyter Notebooks and a napari plugin.