SODA STORM 2D results interpretation

[ezra bruggeman]

I would like to use the SODA STORM 2D protocol to analyse some 2D dSTORM data. I can load my localization files and run the protocol, but I'm not sure how to interpret all the values in SODAResults.xls output file. As far as I know, there is no documentation available?

[Stephane]

Hi,

Did you had a look on the plugin documentation here :

http://icy.bioimageanalysis.org/plugin/SODA_suite

I think the video explains what you obtain from the excel file.

Best,

- Stephane

[ezra bruggeman]

Hi,

Yes it does! In the paper (https://www.nature.com/articles/s41467-018-03053-x) they retrieve the mean coupling distance, but in the SODAResults.xls output file, it gives me a different mean coupling distance for every radius (which makes sense). After some radius the values in the mean coupling distance column plateaus. Do I take this plateau as the mean coupling distance? Or do I need to calculate some weighted average?

Or, do I calculate the mean of the distances column in the other output file ProbAndDist.xls, and have the mean be the mean coupling distance?

Best,

Ezra

Op maandag 23 juli 2018 10:27:17 UTC+1 schreef Stephane:

[Thibault Lagache]

Le mercredi 25 juillet 2018 05:43:58 UTC-4, ezra bruggeman a écrit :

Hi,

Yes it does! In the paper (https://www.nature.com/articles/s41467-018-03053-x) they retrieve the mean coupling distance, but in the SODAResults.xls output file, it gives me a different mean coupling distance for every radius (which makes sense). After some radius the values in the mean coupling distance column plateaus. Do I take this plateau as the mean coupling distance? Or do I need to calculate some weighted average?

Dear Ezra,

Actually, results are given for every intermediate radius. The fact that it plateaus at some point indicates that increasing search radius would be useless as SODA "captured" all the coupling events (i.e. there is no more coupling at further distance).

You should thus consider plateau's values (as done in the paper).

Cheers

Thibault Lagache 

[Lockett, Stephen (NIH/NCI) [C]]

Hi Thibault,

 

I am wanting to use SODA to measure the spatial proximity of different cell types in tissue.  The cells are immunofluorescence labeled in two different colors.  Each positively stained cell is detected in each color and represented by a ROI.  First, I want to eliminate those cells that are positive for both markers (i.e. their ROIs overlap).  My intention for finding these overlaps was to use SODA with a short maximum radius.  However, I get different numbers of colocalized detections when I reverse the inputs.  This might be understandable if a threshold is applied to the likelihood that a pair of cells overlap.  However, I get different results again if I change the input sequence images in various ways and keep the list of detections the same.  This I can’t understand at all, since SODA is clearly based solely on the detected spots (ROIs) and has nothing to do with the underlying images that the spots were previous detected from.  Thank you for your help.

 

Sincerely,

 

Stephen Lockett

--
You received this message because you are subscribed to the Google Groups "Icy imaging" group.
To unsubscribe from this group and stop receiving emails from it, send an email to icy-software...@googlegroups.com.
To post to this group, send email to icy-so...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/9f0b6f90-355d-413e-8e59-245a381e7b53%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Lagache Thibault]

Hi Stephen,

Actually, changing the input sequence images can change the output of SODA because:

- It can change the size of the field of view and thus the statistics for 2 objects to be considered coupled

- the « position » of an object is computed as the center of intensity of its ROI (thus changing the sequence and the intensity map changes the computed positions of objects)

Cheers

Thibault

You received this message because you are subscribed to a topic in the Google Groups "Icy imaging" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/icy-software/mW2nnuu9K5g/unsubscribe.
To unsubscribe from this group and all its topics, send an email to icy-software...@googlegroups.com.

To post to this group, send email to icy-so...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/SN6PR0901MB23364563AC0A2A9F8DA9EC93FFF90%40SN6PR0901MB2336.namprd09.prod.outlook.com.

[Lockett, Stephen (NIH/NCI) [C]]

Hi Thibault,

 

Thanks.  Your reply partially answers my questions, but there are still problems.  In the attachments I show that I get different results when I connect images into “input sequence 1” and “input sequence 2” of SODA, or not make these connections and instead select the same sequences a in the fields to the right of “input sequence 1” and “input sequence 2”.  I have tried this for images from my studies (SODA_issues_1 and SODA_issues_2) and when I use images generated by the colocalization simulator using the default setting (SODA_issues_3 and SODA_issues4).

 

Furthermore, I get different results when I change the step. I get what appears to be correct results when the step size is similar to the object size, but the cells I am analyzing have a range of sizes.  So I have no idea how to set the step in this case.

 

Sincerely,

 

Stephen Lockett

To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/7F187A75-6823-4E59-9762-E58A7463DD8B%40yahoo.fr.

[Lagache Thibault]

Hi Stephen,

Thanks for all your detailed reports! Unfortunately I cannot reproduce your findings: I used the coloc simulator with default parameters and I found the same results with or without connecting sequence 1 & 2  (see attached)…

Thus, I have no idea of what’s happening.

Sorry about that

Thibault

To view this discussion on the web visit https://groups.google.com/d/msgid/icy-software/SN6PR0901MB2336E1599369F85520CAC67BFFF60%40SN6PR0901MB2336.namprd09.prod.outlook.com.

For more options, visit https://groups.google.com/d/optout.

<SODA_issues_1.PNG><SODA_issues_2.PNG><SODA_issues_3.PNG><SODA_issues_4.PNG>