4D scatter plot

Hi Hannah,
I used the 4D XYZColor coded scatter plot for my biofilms…
I could get the images, however the scatter plot are not as dense as those shown in your tutorials.
why is that??
attached are my imagesSend.tif (111.3 KB)

Hi Srinivas,

the density of the points in the image depends on the density of cubes in your data. You can increase it by lowering the cube side length in the objects declumping option of the segmentation (see screenshot below). The cube side length determines the resolution with which you measure properties in your biofilm, so it should always be smaller than the size of the pattern that you want to observe (e.g. fluorescence pattern, density differences…). At the same time, decreasing the cube size will slow down the property measurements, so it is best to find a balance.

Let me know if you encounter any more questions.

Best,

Hannah

Hi Hannah,
these are the new images wherein in use the cube side length as 10…are these images good??
would you suggest me to reduce the cube side length further to 5??
if we are using this software, do we cite the biorixv paper or the nature microbiology paper??
HA Discs_biovolume_10 cube size.tif (137.8 KB) New_10cube size.tif (136.9 KB)
regards

srinivas

Hi Srinivas,

without knowing what the original images look like and which patterns can be observed in your communities, I cannot tell whether the plot you sent captures the essence of your data. I suggest that you take a few example images for which you know or can tell from the images how the properties you are measuring should behave. Then calculate them and look at the results carefully to see if they make sense. Once you have established a workflow that works well with your data, you can run it on the rest of your imaging data.

A very useful method to view measured properties in BiofilmQ is the segmentation preview. To open it, click the “Load Segmentation results” button in the Segmentation preview area below the image preview in the center of the GUI (see below). Be sure to select an image before clicking it.

image

When you do so, a list pops open, where all locally measured properties are shown. Select one, then press “Ortho view” to get to the view shown in the screenshot below on the right: Here, your data is shown, with the colors corresponding to the value of the selected property for each cube. I usually find this view very helpful to determine if the results make sense. You can place it side by side with the image overlay (“Overlay” button next to the image preview) to get a feeling for how the raw data is transferred into the calculated properties.

I hope this description was helpful to you. For citing this software, please refer to the biorxiv paper for now, thank you :grinning:

Best,

Hannah