How to quantify bioluminescence in roots?

Hi,

I’m new to image J and i’m trying to figure out how to quantify bioluminescence in root images (at tissue level). I have a set of 733 images showing DR5:LUClFERASE (bioluminescent) signal over 24 hours. The signal oscillate with a periodic timing of 4-6 hours. I would like to quantify the signal over time in order to know the periodicity of these oscillations in the roots of three plants. I don’t know what kind of macro i should use ?

This is the graph, i want to reproduce: the unit are ADU = analog digital unit (y) and time (x).

Without an example image I would select (possibly with threshold or other segmentation methods) the (probably) static region with a ROI and add it to the ROI Manager.

I then would import all your images as an image Sequence which results in a stack, see:

https://imagej.nih.gov/ij/docs/guide/146-26.html

You can now apply a measurement on all slices of the stack with the ROI Manager Menu: More->Multi Measure
to measure the region in all slices (your time interval images), see:

https://imagej.nih.gov/ij/docs/guide/146-30.html#fig:The-ROI-Manager

In the set measurements dialog you can determine which measurements you would like to apply (e.g., mean grey value or integrated density):

https://imagej.nih.gov/ij/docs/guide/146-30.html#sub:Set-Measurements

There is also another approach if you need to preprocess the images individually (split color channels, etc.).

Then you could also write a small macro (use the macro recorder) and then apply this macro on your images, see:

First of all, thank you for you answer. It confirmed some of the step i wanted to follow to quantifiy my signal. I need to precise that my signal is not static, it’s oscillating so i changing over time, more like a wave than a static point

I still have a few questions:

  • when importing my images, do i need to convert them to RGB or 8-bit-grayscale?

  • which zone should i measure with integrated density, would it be better to measure the whole part of my root with the signal or measure several spots where i see the oscillation (i have three spots on each root over time)

  • what can i do with the data i got with the multi-measure tool: i don’t know if i must use the values named “Intden” or “RawIntDensity”.

  • do i need to normalise my signal? how? i thought about quantifying with integrated density the black background and then substract this values (Intden) to my signal values.

when importing my images, do i need to convert them to RGB or 8-bit-grayscale?

Which types have your source images (devices, output)? Without any information (image calibration) or example images it is hard to say.

Also a look at the calibration methods to possibly calibrate the luminiscense (from known standard signals):

https://imagej.nih.gov/ij/docs/guide/146-30.html#sub:Calibrate

http://imagejdocu.tudor.lu/doku.php?id=gui:analyze:calibrate

which zone should i measure with integrated density, would it be better to measure the whole part of my root with the signal or measure several spots where i see the oscillation (i have three spots on each root over time)

That depends on your question. If you located your spots (segmented them) you can do that with several ROI’s and measure them over the time.

what can i do with the data i got with the multi-measure tool: i don’t know if i must use the values named “Intden” or “RawIntDensity”.

See here for a similar answer and please consult the ImageJ documentation or Wiki:

http://imagejdocu.tudor.lu/doku.php?id=gui:analyze:set_measurements

do i need to normalise my signal? how? i thought about quantifying with integrated density the black background and then substract this values (Intden) to my signal values.

What does the literature about this (common practice)?. Do you have any references or technical details?