Opening 6D datasets in FIJI

Dear image.sc community,
I have a problem opening 6D datasets of a beating zebrafish heart from the Leica Digital Light Sheet microscope in FIJI.

With 6D I mean, XY, C, T, Z, L (loop). I can save as tiff file series, lif or lof files.

The desired outcome is to obtain small “.ome.tiff-files” of XYCT for each Z-plane and each Loop.

I created a macro-based workflow with a work around, using tiff files series. That works for basically all my datasets.

The problem is that it takes forever at the microscope to save big datasets with 750.000 individual tiff-files. It can take a whole weekend and more. To transfer these files afterwards to a server or harddrive takes another day. Similar problems occur when converting the files in the Leica Software from lif to tiff. I could solve the problem by reducing the hyperstack by one dimension directly at the microscope (by splitting Z-planes for example). Unfortunately the Leica-software does not allow this.

I need to save as .lif or .lof from the microscope and ideally load parts of the datasets in ImageJ. When I use Bioformats “Specify subset” the order of the imagestack is a mess. Of course there is no XYCTZL stack order. I spent a long time, trying to recognize a pattern that I can use macros to bring order into the chaos, but the channels, frames and Z-planes are jumping without an obvious sense.

I would be very happy for any solution, suggestion or idea.

Thanks a lot in advance.

Hi,
I have not worked with 6D datasets myself nor the lightsheet microscopy, but is it possible to reduce one dimension already at the microscope? I mean save generate multiple 5D datasets (x,y,c,z,t)?
What ‘loop’ means in your experiment?

Best regards,
Ilya

Thanks a lot for your answer. I had this possibility in mind, but it is not possible in the Leica Software.

The loop means, I record short movies (30 Frames in about 1.5 seconds) of every Z-plane every 10 min.

Hi Alexander,
thank you for your clarification, unfortunately I can’t suggest anything. May be someone from Bio-Formats can clarify the order of the planes for such datasets.
Ilya

Hi Alexander,

When you open one of these 6D datasets with the Bio-Formats importer is the 6th dimension (loop) showing up as part of the Time series? If you select ‘Display OME-XML’ during the import steps you will be able to see the DimensionOrder on the Pixels element. This will however only have 5 dimensions listed but will tell you the order of the frames.

When you were experimenting with specify subset, was the result dropping the correct dimensions or was it not behaving as expected?

Hey dgault,
Thanks a lot for your suggestions. Yes, the Loop is shown as part of the time series. Nevertheless, the specify subset was not working correctly. There was a major error in the stack order still. I now spent a while on trying different options.

I could solve the Problem now:
I loaded the .lif in the correct stack order, according to the .XML metadata, and before creating a hyperstack I split the stack by the number of loops. Afterwards, I was able to create the correct hyperstack.

Thanks a lot for your comments Ilya_Belevich and Dgault.

1 Like