Supporting time for inCarta data

[tischi]

Also, regarding multidimensionality. Is time lapse handled somehow? I was just looking at one big INCarta folder, with 4D data, >90000 files, and all time-lapse steps are in the same folder (t1... to t48...)?
Is it possible to select one one timepoint. Or does it assume a whole folder is just one timepoint?

Hi @ThomasMicroscopy, I think I should be able to integrate time-lapse support for the InCarta data, such that you can open all time points at once and then there would be a time slider at the bottom of the BigDataViewer window.

For testing, could you please share a minimal dataset with me, e.g. containing two timepoints?

[ThomasMicroscopy]

Hello again. Thank you for giving this a look.
I created a bunch of minimal recordings with MetaXpress. This is the software that Molecular Devices uses to record from their high throughput microscopes like ImageXpress Confocal. Since the HTD, XDCE, and JDCE files are rather similar at the .tif level, I sent now small datasets for all them.

When recording into the SQL Databank of MetaXpress, this cannot handle directly 4D formats. Either only Z-stacks, or Time-lapse. If you do z and time-lapse at the same time, it stores only the Z-projection for the the z-planes.
From this database one can either export as .hTD, or as .xdce (in Carta)
Once does have a choice to bypass the SQL database (at the expense of not being able to review that data - hence Mobie), and then the tif files are directly stored as .xdce.
Opening the .xdce files in In Carta, converts them the .jdce format, whereby time-lapse is split into folders, as I noticed.

So in the dataset you have:
HTDMetaTime: A timelapse (no Z) exported from MetaXpress in .htd
HTDMetaZ: A Z-stack series (no timelapse) exported from MetaXpress in .htd
HTDMetaZPlusProjection: A Z-stack series (no timelapse) exported from MetaXpress in .htd (adds a Step O folder)
MetaInCartaTime: A timelapse (no Z) exported from MetaXpress in .xdce for Incarta (actually same data as HTDmetatime)
MetaInCartaZ: A Z-stack series (no timelapse) exported from MetaXpress in .xdce for Incarta (same data as HTDMetaZ)

Now these files are directly recorded in INCartaformat (with MetaXpress, but bypassing the SQL storage of MetaX)
InCartadirect: 4d dataset (Z and Time)
InCartadirectZfolder: 4d dataset (Z and Time)
[in the MetaXpress software - in INCarataMode - there is an option "Z-subfolder" ]
InCartadirectTimefolder: 4d dataset (Z and Time) [using the option "time-lapse" subfolder in MetaX - INcartamode ]

As far as I can see, there is no difference in the file arrangements, not subfolders in these three datasets.

I then opened all three datasets in the IN Carta Software, which does the conversion to jdce:
InCartaOpened_JDCE folder this contains the three InCarta datasets above, all opened with IN Carta.

So, what it looks like is that during recording, either in MetaXpress mode with exporting to In Carta, or
with direct recording InCarta mode,
the files are all exported into a single folder (tx_Wellname_sx_wx_zx.tif).
When then opening into IN Carta (no matter what options for subfolders were selected), it seems
all timelapse (t1, t2 etc.) are reorganised into Folders with Timepoint0 etc.) - similar to the .htd format, which makes timepoint0 folders.

So, it perhaps looks more complicated than it is, I just wanted to make sure that I don't miss some odd
data rearrangement into folders that would cause issues for a 4D viewer that needs to find the z and timelapse files
in the right places.
So it seems primarily: Either all files in one folder (in carta), or files split into timepoint folders (in carta), or
files split further (MetaXpress) into time folders and Z-folders.

[ThomasMicroscopy]

Just a short comment. I can load 2d Projections saved in xdce in Mobie, and look at them. That's great, because recording directly in InCarta format gives no overview, so MOBIE to the rescue. after Incarta converts it to jdce though, it gives crashes, which I don't quite understand, since the tifs look superficially the same.

[ThomasMicroscopy]

If it's easier to implement, just being able to select one timepoint at start to open in MoBIe would already be useful.

[tischi]

Hi, thanks!

Sorry, I am travelling right now, might take a bit until I can do this....

[ThomasMicroscopy]

Hi, thanks for the update. Don't worry.
I used a utility to compare an INCarta xx.tif file (in an .xdce folder) with the same INCarta xx.tif file after conversion (by opening the data with IN Carta) to .jdce . The tif files are identically with this test, so really it's only the metadata (.xdce -> .jdce and other files) which changes in this process. So maybe something with the extra files in .jdce which causes a crash.