Click-ExM data process and example

Matlab code and example files for manuscript Click-ExM enables expansion microscopy for all biomolecules, by De-en Sun, et al.

Uploaded and last edited by Yujie Shi (yujieshi@scripps.edu) Oct-06-2020.

Introduction

The script Process_click_EXM.m in this repo is created for expansion microscopy data processing. This includes rigid registration, calculation of expansion factor, B-spline non-rigid registration, calculation of distortion field and RMS error. For detailed explanation of the code, see the annotations inside the script. All dependencies (registerImages.m, nr_registerImages.m, RMSE_cal.m, and RMSE_MS.m) are included in this repo. The code was written and used under Matlab R2018a on Windows 10 system.

Description of the method

Rigid registration and expansion factor calculation

The pre-expansion images were first stretched by a scaling factor F_esti of 3.8~4.8 with bicubic interpolation to approximately match the scale of the post-expansion images. This estimated expansion factor was decided based on the macroscopic measurement of gel size. The stretched pre-expansion image was used as the ‘fixed image’ for rigid registration, and the post-expansion images as the ‘moving image’. Rigid registration was performed using the monomodal intensity-based registration with imregtform function. Affine transformation was used in the function to correct the shear deformation of the gel. The X- and Y- scaling elements (x_corr and y_corr) in the transformation matrix calculated by the imregtform function were used to correct the rough scaling factor (F_esti). The final corrected expansion factor (F_corr) was therefore calculated as F_corr = F_esti×√(x_corr×y_corr)

Non-rigid B-spline registration and distortion analysis

The registered pre- and post-expansion images were then subjected to B-spline non-rigid registration. To exclude regions with no features, masks of the rigid registered pre- and post-expansion images were generated by Gaussian blur to suppress the background. The displacement field and B-spline registered images were then acquired with the imregdemons function, and the distortion vector field was visualized with the quiver function. RMSE was quantified by calculating the difference of distance between each pair of matching features before and after B-spline registration, and plotted as a function of the distance between the matching features. The calculated RMSE-to-measurement length curve was smoothed for better visualization.

Instruction for use

1. Download all files into a same folder.
2. Run scirpt Process_click_ExM.m in Matlab. This will process the included example files pre_Click_ExM_tubulin.tif and post_Click_ExM_tubulin.tif included in this folder.
3. The rigid registration and non-rigid registration will finish automatically.
4. Select a small region to calculate RMS error when prompted (Use your mouse to draw a square on the figure, double click the selected region when finished).
5. Wait for RMSE calculation (may take a long time if a large region is selected). A "Process finished" message box will show up when all calculations are finished.
6. All results will be written into a newly created folder called processed_data

Explanation of expected output

There will be five .svg files and four .xlsx files in the newly created processed_data folder:

The .svg files shows registration results:

• (i) compare pre and post (registered).svg shows the rigid-registered post-ExM image and the original pre-ExM image;
• (ii) rigid registration and exp fac.svg merges the rigid-registered post-ExM image with the pre-ExM image; the estimated and corrected expansion factors are shown in the title;
• (iii) non-rigid registration and distortion field.svg shows the post-ExM image before and after non-rigid registration and the distortion field;
• (iv) Cropped area for RMSE analysis (Distortion field).svg shows the cropped area for RMSE analysis selected manually by the user;
• (v) RMSE plot.svg shows the RMSE plot, with mean value ± standard deviation.

The .xlsx files saves RMSE related values, which can be used to make RMSE plot using other softwares

• RMSE mean.xlsx is the calculated mean RMSE value
• RMSE mean+std.xlsx is the calculated mean RMSE value + standard deviation
• RMSE mean-std.xlsx is the calculated mean RMSE value - standard deviation
• RMSE measurement_length.xlsx is the measurement lengths correlated to the mean RMSE values