ImageJ scripts and R code

Overview

A collection of R scripts I have put together over the course of my PhD research project to process and plot data generated from image analysis performed in ImageJ. For each project, example data and outputs are provided so that the user can run the R scripts and produce the plots. I have tried to generalise the scripts in the hope they could be useful for other projects.

Analyse Particles | 3D object counter | Microtubule plus-end tracking

Data organisation

• A similar directory structure is used for all projects. The parent directory is named according to the project (i.e. Analyse particles) and contains the subdirectories; Data, where the output from ImageJ, the lookup.csv and log.txt are located; Output, where the dataframes and plots are saved to the directories Dataframe and Plots, respectively; and Scripts, where the ImageJ and R code are located.
• Within Data there are subdirectories containing the ImageJ output from separate experiements, each labelled with their unique experiment number (e.g. JS103).
• Where log.txt is used, analysis was performed blind to the conditions of the experiment and a lookup.csv is loaded in R to add the original labels. The lookup table is also used to set the levels of the independent variable, enabling the user to change the plotting order by editing the csv file.

R code

• For each project there is a Project_name_process.R script that is used to process the files in the Data directory experiment-by-experiment to generate dataframes that are saved to Output/Dataframe. The working directory should be set to the parent directory of the project. At the start of the script you will be prompted to select the subdirectory containing the files to be processed, e.g. Analyse_particles/Data/JS103. Re-run the script until all experiments in the project have been processed.
• A Project_name_combined.R script is used to combine the dataframes from individual experiments, calculate statistics and generate plots that are saved to Output/Plots. The working directory should be set to the parent directory.

Analyse particles

This was initially written to quantify transferrin uptake in fixed cells using images acquired by light microscopy.

Typical image used to quantify tranferrin uptake (purple spots)

Running the code

• Analyse_particles.ijm will threshold the transferrin channel to isolate vesciular structures and use analyse particles to create a mask of the particles based on parameters set by the user. The output is a csv file containing the results from analyse particles and a txt file containing the values of the threshold limits used for the analysis. Run the script on a directory containing all of the tiff images to be analysed and select a separate directory to output the data.
• Analyse_particles.R will process the Data (output from ImageJ) to generate a dataframe, plots and calculate statistics. The dataframe and plots are saved to Output.

Notes

• There is only a single experiment in this project so a Analyse_particles_combined.R script is not required.

3D object counter

This was written to detect and measure objects in z-stack images acquired by confocal microscopy. The example provided here detects and measures pericentrin (centrosome marker) in images of HeLa cells fixed in metaphase.

Example micrographs of HeLa cells fixed in metaphase to visualise pericentrin (green) and tubulin (red)

Running the code

• 3D_object_count.ijm will require the user to draw around the outline of the cell to be analysed and use 3D Objects Counter to measure the objects in the image. The output is a csv file containing the measurements and a txt file containing the size filter and threshold limits used in 3D Objects Counter as set by the user. Run the script on a directory containing all of the tiff images to be analysed and select a separate directory to output the data to.
• 3D_Object_Count.R will process the Data (output from ImageJ) for each experiment and generate a dataframe that is saved to Output/Dataframe.
• 3D_object_count_combined.R will combine the dataframes generated in the previous script and produce plots that are saved to Output/Plots.

Microtubule plus-end tracking

I wrote an R script to process and plot output from u-track, a multiple-particle tracking MATLAB software developed by the Danuser Lab. I used this software to detect and track fluorescently-labelled EB3 (a microtubule plus-end tracking protein, (+TIP)) in movies taken from live-cell imaging experiments, to measure microtubule growth parameters under different conditions.

Still image of growing microtubles in a cell expressing fluorescently-labelled EB3

Running the code

• Microtubule_tracking.R will process the Data (output from u-track) for each experiment and generate a dataframe that is saved to Output/Dataframe.
• Microtubule_tracking_combined.R will combine the dataframes, calculate statistics and generate plots. Plots are saved to Output/Plots.

Notes

• The organisation of Data is slightly different to the other projects in that there are multiple subdirectories and the analysis was not performed 'blind'. This is because the analysis was performed using the MATLAB based software u-track, which is an automated particle tracking software. The output from u-track includes mat files in addition to the txt file used here for processing. The mat files were removed from the Data directory here for simplicity.