Circadian Gene Expression (CGE)

An ImageJ plugin for quantifying level of circadian gene expression by tracking individual cell

Written by Daniel Sage at the Biomedical Image Group, EPFL and Charna Dibner at the Circadian Endocrinology Group, HUG, Switzerland

splash image

Outline

This ImageJ plugin (CGE) is a semi-automatic tool to detect and track moving cell, and to measure the fluorescent protein expression level. CGE extracts the trajectory of the cells by tracking their displacements, makes the delineation of cell nucleus or whole cell, and finally yields measurements of various features, like reporter protein expression level, cell displacement.

Reference

[1] D. Sage, M. Unser, P. Salmon and C. Dibner, "A software solution for recording circadian oscillator features in time-lapse live cell microscopy", Cell Division 5:17, 2010.
[2] C. Dibner, D. Sage, M. Unser, C. Bauer, T. d'Eysmond, F. Naef, Ü. Schibler, "Circadian Gene Expression Is Resilient to Large Fluctuations in Overall Transcription Rates," The European Molecular Biology Organization Journal, vol. 28, no. 2, pp. 123-134, January 21, 2009.

Software

Specification

Circadian biologists are applying time-lapse microscopy using fluorescent reporters to study individual cell oscillation pattern. In an attempt to improve existing methodology of time lapse microscopy analysis for such experiments, we developed this ImageJ plugin, called CGE. It is a semi-automatic tool to detect and track moving cell, and to measure the fluorescent protein expression level.
CGE extracts the trajectory of the cells by tracking their displacements, makes the delineation of cell nucleus or whole cell, and finally yields measurements of various features, like reporter protein expression level, cell displacement, and granulometry index. In spite significant circadian variations in protein expression with extremely low expression level at valley phase, this software allows recording accurately and efficiently large number of cells parameters.

Pre-requirement

The software provided here is a plugin for ImageJ, a general purpose image-processing and image-analysis package. It runs also on the image-processing package Fiji. ImageJ has a public domain licence; it runs on several plateforms: Unix, Linux, Windows, and Mac OS X. It doesn't take more than a couple of minutes to install.

Download

Download CGE_.jar, the ImageJ's plugin. The plugin consists in one single JAR file; place it into the "plugins" folder of ImageJ. Do not unzip the JAR file.

Conditions of use

You are free to use this software for research purposes, but you should not redistribute it without our consent.
In addition, we expect you to include adequate citations and acknowledgments whenever you present or publish results that are based on it.

Instructions

The software package CGE consists in two ImageJ plugins. The preprocessing plugin CGE_Preprocessor is applied only one time to the image sequence. It precomputes all data required for the recording. It is highly computational and can required several minutes for a sequence of few hundreds images. The recording of the gene expression over time is performed by the second plugin CGE_Recorder. This plugin uses the data produced by the first plugin and interact with the user to quickly determine the trajectory and the outline of a nucleus.

Preprocess image sequence

Launch the plugin CGE_Preprocessor, menu Plugins » CircadianGeneExpression
Select a folder contains the sequence of images.
Give the time-lapse frame interval in minutes and the typical radius of a nucleus (object to track) in pixels.
Select the appropriate denoising operations. The choice of this operations and the parameters depends of the quality and of the level of noise of the input images.
Click on the "Run Preprocessing" button.
At the end, the preprocessing plugin generates a folder named with a "-cge" suffix. This folder contains the preprocessed sequence of images. The data of the folder will be used for the second plugin CGE_Recorder.

Record gene expression

Launch the plugin CGE_Recorder, menu Plugins » CircadianGeneExpression.
Select a folder named with a "-cge" suffix. Two stacks are laoded "rescale.tif" and "original.tif".
Create a new trajectory by clcik on the button "Create". Double-click on the name of the trajectory allow to rename the trajectory. At the beginning there is alreary one existing trajectory called "1".
Defintion of the trajectory
- Define the trajectory of a nucleus by clicking on its center and choosing "Add".
- Select at least two time position to create a trajectory.
- The trajectory is computed by a optimization procedure (dynamic programming) when the checkbox "Trace optimization" is checked or it is computed by spatio-temporal linear interpolation when the checkbox "Trace optimization" is unchecked.
- Better results is obtained by adding several positions, specially when the nucleus is closed to other ones or when the level of the nucleus is close to the background level.
- The optimization procedure for the tracking has a cost function expressed of a sum of 4 terms: the absolute and the variation of intensities and the absolute and the variation of the displacement.
- The optimization procedure has two constraints the maximum displacement from one the next frame (in pixels) and the confinement area (in pixels).
Outline of the nucleus
- Choose "Run" to start the shape outline detection.
- The detection of a nucleus is perfomed used the active rays technique, the number of rays (or vertices) is a parameters of the algorithm (see the "Settings" panel).
- Checking "Shape refinement" could give more accurate results but it highly computational.
- Checking "Postprocess (smooth)" applies a Gaussian smooth of the results, position and shape, the level of this regularization is set by three parameters in the panel "Settings". This prostprocess tends to reduce the number of false detection.
- The measurements are performed during the outline of the nucleus. Three zones of measurement are defined as percentage of the detected outine.
  - Zone 1: internal area of the nucleus, the percentage should be less than 100%.
  - Zone 2: periphery area of the nucleus, the percentage should be greater than 100%.
  - Zone 3: outer area of the nucleus, the percentage should be greater than 100%.
- The plugin has two mod to evaluate the background level 1) It could a uniform background for all the nucleus, this uniform background is estimated at the beginning of the plugin, but the background can also be entered manually (see the "Measure" panel). 2) The background can be computed individually for every nucleus has the mean intensity of the zone 3 (outer).

Exploitation of the results

In the plugin CGE_Recorder there are several options to visualize the results.
- Show as a kymograph.
- Show all results in table.
- Show some results in a graphical plot.
- Save all measurement in a CVS file. This file is stored in the folder "-cge".
After saving the results, take your favourite spreasheet to plot the results you want.

Preprocessing steps

Enlarge the screenshot

Recording trajectories

Enlarge the screenshot

Zone of measurements
zones

Features	Unit	Description
#	-	Frame number
Time	minutes	Time of the frame according the time-lapse frame interval parameters
T-test	-	Significance of the nucleus detection, based on a contrast measurement
Division	-	Estimation of the division of cell, 0 means no division, 1 means division in the corresponding frame
Displ	pixel	Displacement of the nucleus beetwen the current frame and the previous frame
Express	level of intensity	Relative gene expression. It is the average of intensity inside the nucleus minus the background level
Var. Dir.	-	Local variation of directions
Area1	pixel	Size of the zone 1 (nucleus)
Area2	pixel	Size of the zone 2 (periphery), 0 is the background is uniform
Area3	pixel	Size of the zone 3 (outside), 0 is the background is uniform
XC	pixel	X position of the gravity center
YC	pixel	Y position of the gravity center
Mean1	level of intensity	Average of intensity inside the zone 1 (nucleus)
Mean2	level of intensity	Average of intensity inside the zone 2 (periphery), 0 is the background is uniform
Mean3	level of intensity	Average of intensity inside the zone 3 (outside) or constant value is the background is uniform
Sum1	level of intensity	Sum of intensity inside the zone 1 (nucleus)
Sum2	level of intensity	Sum of intensity inside the zone 2 (periphery), 0 is the background is uniform
Sum3	level of intensity	Sum of intensity inside the zone 3 (outside), 0 is the background is uniform
Min1	level of intensity	Minimal value of intensity inside the zone 1 (nucleus)
Max1	level of intensity	Maximal value of intensity inside the zone 1 (nucleus)
StdDev1	level of intensity	Standard deviation of the intensity inside the zone 1 (nucleus)
Granules1	energy	Estimation of the level of granulometry inside the zone 1 (nucleus)
Corr Min	level of intensity	Correction: minimal value, use only for internal purpose
Corr Min	level of intensity	Correction: maximal value, use only for internal purpose
Direction	degrees	Estimation of the direction of displacement [-180°..180°]

Results

Experiment 1

Download

Original sequence of image: RFP8.zip
Stack with the 3 good marked cells on original background: RFP8-7-goodcells-original.zip
Stack with the 3 good marked cells on rescaled background: RFP8-3-goodcells-rescale.zip
Measure - 7 good marked cells: RFP8-3-goodcells-measure.xls Excel file

Experiment 2

Download

Original sequence of image: pos10s.zip
Stack with the 10 marked cells on original background: pos10s-10-cells-original.zip
Stack with the 10 marked cells on rescaled background: pos10s-10-cells-rescale.zip
Measure - 10 marked cells: pos10s-10-cells-measure.xls Excel file

daniel.sage@epfl.ch • 30.10.2020