Data from: Towards drift-free high-throughput nanoscopy through adaptive intersection maximization
Data files
Apr 25, 2024 version files 4.56 GB
Abstract
Single-molecule localization microscopy (SMLM) often suffers from suboptimal resolution due to imperfect drift correction. Existing marker-free drift-correction algorithms often struggle to reliably track high-frequency drift and lack the computational efficiency to manage large, high-throughput localization datasets. We present an adaptive intersection maximization-based method (AIM) that leverages the entire dataset's information content to minimize drift correction errors, particularly addressing high-frequency drift, thereby enhancing the resolution of existing SMLM systems. We demonstrate that AIM can robustly and efficiently achieve an angstrom-level tracking precision for high-throughput SMLM datasets under various imaging conditions, resulting in an optimal resolution in simulated and biological experimental datasets. We offer AIM as simple and model-free software for instant resolution enhancement with standard CPU devices.
README: Towards drift-free high-throughput nanoscopy through adaptive intersection maximization
AIM
Adaptive Intersection Maximization (AIM) is a high-speed drift correction algorithm for single molecule localization microscopy.
The details are presented in our paper entitled "Towards drift-free high-throughput nanoscopy through adaptive intersection maximization".
All the codes under \DME_RCC are from https://github.com/qnano/drift-estimation published in Jelmer Cnossen, Tao Ju Cui, Chirlmin Joo, and Carlas Smith, "Drift correction in localization microscopy using entropy minimization," Opt. Express 29, 27961-27974 (2021).
Hardware requirement:
AIM requires only a standard computer with a minimum of 16GB of RAM.
RCC and DME require a minimum of 32GB of RAM to handle the large datasets generated from systems with a large field of view (e.g., 2048 x 2048 pixels).
Software requirement:
The provided codes have been tested on MATLAB version 2020b to 2023a on Windows 10 Operating System.
Installation:
Users can direacly download the codes and run the demo code on MATLAB.
Users need to replace the file name when processing users' own datasets.
Demo datasets:
We provided four experimental datasets (Origami_PAINT, Microtublue_3d, Tissue_colon and CTCF_MCF10A_DRB_6h) along with one simulated dataset (simulationSMLM) in MATLAB '.mat' format available at Dryad. Please download these datasets and put them in the Data folder.
The 'simulationSMLM.mat' (Figure 2) and 'Microtubule_3D.mat' (Figure 6) datasets contain 3D localization lists, consisting of four variables: 'F', 'X', 'Y', and 'Z'.
'F': Frame index of the localization (unit: frame). Note that the time duration for each frame is 10 ms for both 'simulationSMLM.mat' and 'Microtubule_3D.mat'
'X': X position of the localization (unit: pixels)
'Y': Y position of the localization (unit: pixels)
'Z': Z position of the localization (unit: pixels)
The remaining datasets, 'Origami_PAINT.mat' (Figure 3), 'CTCF_MCF10A_DRB_6h.mat' (Figure 4), and 'Tissue_colon.mat' (Figure 5), contain 2D localization lists with three variables: 'F', 'X', and 'Y'.
'F': Frame index of the localization (unit: frame). Note that the time duration for each frame is 50 ms for 'Origami_PAINT.mat', 20 ms for 'CTCF_MCF10A_DRB_6h.mat', and 10 ms for 'Tissue_colon.mat'
'X': X position of the localization (unit: pixels)
'Y': Y position of the localization (unit: pixels)
Example files:
We provide four MATLAB codes as examples to demonstrate how to use AIM.
example_ExperimentalData.m: This code performs drift correction with AIM on 2D or 3D localization coordinates of experimental data. Sample experimental data are available at Dryad
example_code_2D.m: This code compares the performance of drift correction for AIM, RCC and DME using 2D localization coordinates for the experimental dataset of DNA origami (Origami_PAINT.mat) or simulated data (simulationSMLM.mat) available at Dryad
example_code_3D.m: This code compares the performance of drift correction for AIM, RCC, and DME using 3D localization coordinates of experimental data of simulated data or experimental data of microtubules (Microtublue_3d.mat) available at Dryad
example_code_FigureS1.m: This code is used to reproduce Supplementary Figure S1, which shows drift tracking precision under a wide range of image sizes from 128×128 pixels to 2048×2048 pixels.
Other files:
simulationSMLM.m: This code is used to generate the simulated SMLM dataset from DNA origami structures used in Figure 2 in the main text.
save_imSR.m: This MATLAB function is used to save the SMLM dataset into a tif image.
Load_ThunderSTORM.m: This code is used to provide a MATLAB function to read the localization dataset (csv files) from the commonly used ThunderSTORM software.
Methods
We provided the following dataset.
- Two-dimensional single-molecule localization point list from an DNA origami structure (Origami_PAINT.mat). This data is used to produce Fig. 3 in the main text.
- Two-dimensional single-molecule localization point list from CTCF of a cell line MCF10A treated with DRB (CTCF_MCF10A_DRB_6h.mat). This image size is 2048 x 2048 pixels with a pixel size of 100 nm. This data is used to produce Fig. 4 in the main text.
- Two-dimensional single-molecule localization point list from a colon tissue (Tissue_colon.mat). This image size is 2048 x 2048 pixels with a pixel size of 100 nm. This data is used to produce Fig. 5 in the main text.
- Three-dimensional single-molecule localization point list for microtubules from COS-7 (microtubule Microtublue_3d.mat). This image size is 2048 x 2048 pixels with a pixel size of 100 nm. This data is used to produce Fig. 6 in the main text.
- Simulated three-dimensional single molecule localization point list for DNA origami structure (simulationSMLM.mat). This dataset is used to produce Fig. 2 in the main text. It can also be generated using the provided code "simulationSMLM.m"