segmentation method for OGB signal?

[Marta Gajowa]

Hi all,

I am new with SIMA and I just started using it. Do anyone of you could recommend me a method for segmentation but for OGB, not for GcamP. I tried normalized cuts method and PlaneCA1PC till know.  In first case the ROI boundaries were in the middle of my cell bodies (I do not have ROI Buddy ( it's not working on the newest mac system with the PyQt4 library) so I only can look for the extract summary file in pdf). This I was trying on my own data, so I tried on the workflow data form you the PlaneCa1PC method. At first when I was just trying the tutorial code delivered by you - it was fine and ROI was really nice, but after failing on my own data I decided to go back to your data. The only thing I did is to change channel names in a way that the segmentation was done on tdtomato channel (which fluorescent picture is much more similar to the OGB pictures). In this case the ROI's were completely not at the cells. I suppose I should play more with parameters, which I will do, but in the meantime If anyone has a suggestion for the segmentation method for OGB I will be grateful.

I attach to ROI maps - the nice one and the second made by CA1PC method, Unfortunately I lost the one from normalised cuts method.

best,

Marta Gajowa

[Patrick Kaifosh]

Hi Marta,
Automatic segmentation is still a rather tricky subject, with some methods working well on some datasets, but no methods really working amazingly on all types of data, so I can't guarantee that any of the included methods will work satisfactorily for you. In general, if your cells are densely packed and have dark cytoplasmic regions, the CA1PC method may work, but otherwise you probably want to use the stICA approach. It's important to indicate which channel has the dynamic signal (e.g. GCaMP or OGB), since these methods use the dynamic nature of the signals to extract the ROIs.
Cheers,
Patrick

--
You received this message because you are subscribed to the Google Groups "sima-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sima-users+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sima-users/c5fae80f-5508-4917-9d66-b6a3568470eb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.