Correct particle intensities in Trackmate



Hello. We are using the Trackmate plugin to track single fluorescent protein molecules in living cells. Apart from the location of the spots, we are also interested in their intensity, because that reports on the proteins being monomeric, dimeric etc. What I noticed is that the intensity values that I can access, such as mean, median, max or total intensity, all represent pixel-values, and they all suffer from “pollution” by autofluorescence of the cells that decays (bleaches) gradually over time. Considering that the software fits a Gaussian funcion to the spot, I was wondering if it would be possible to log the height or the volume of the Gauss, because that would be background-corrected, I presume. Other suggestions are welcome.


Hi @yves_bollen

Actually TrackMate does not fit a gaussian on spots.

We detect spot by LoG (or DoG), with a sub-pixel localization scheme, and we measure the intensities within the radius specified by the user.


Thank you for clarifying, I was confused. But my problem remains, unfortunately. I now tried to export the spots in my tracks to the ROI, so I can use another plugin to perform a Gaussian fit on the intensity profile in the images. However, when I use the export-option (from the extras) ALL spots are exported. For my data it works better not to be too strict on the spot detection level, but more strict when filtering the tracks. As a results, many spots are not part of any track in the end. Is there a way to only export the spots that are in tracks to the ROI? That would be very useful!


What about taking another route and writing an action or a spot feature that would do the gaussian fit in TrackMate ?