Stitching an image file with multiple 'fields' (made with Andor iQ)


I have a single tiff file containing multiple time points, ‘fields’, montage tiles, z-slices and channels. It was made by Andor iQ software.

By ‘fields’ I mean multiple positions on the microscope stage (e.g. representing multiple samples) and a montage array of the same shape has been imaged at each of those fields.

I’ve been trying to simply stitch each field separately using a combination of the Bio-Formats importer and Grid/Collection stitching plugins but haven’t found a solution. The best I have managed so far is when in the stitching plugin I choose:
Type=Positions from file
Order=Defined by image metadata
Compute overlap=unchecked
In this case the multiple fields are all stitched correctly into the same image but with huge black spaces between them (I guess this is representing the real distance between samples). What I would want as output is separate images for each stitched field.

One problem is that with Bio-Formats all the tiles and fields are read as separate ‘series’… I can’t find a way to separate fields in Bio-Formats like you can with channels/focal planes/timepoints.

One solution would of course be to make a macro which separates out the fields before applying the stitching but there are some complications with this too like losing the metadata and (?) needing to save the individual field files before stitching.

So I’m mainly wondering if I’m missing something, like a way with Bio-Formats to automatically separate the ‘tiles’ from ‘fields’?

Many thanks in advance for the help,

Hi Tom, would you be able to post the OME-XML output here to see what the metadata for the tiles and fields looks like?

Hi David, thanks for the reply!

I’ve attached a screenshot of the top of the ome-xml, does that help? (I couldn’t think of a way to export ome-xml metadata?)

The 18 ‘planes’ in the ‘image’ are 9 z-slices with 2 channels. In the whole file there are 80 ‘images’ representing the 8*5 montages tiles for 2 fields.

Thanks again for the help,

I’m assuming all of the 80 images all have the same dimensions? In the XML is there anything distinguishing the 2 fields from one another?

Yes each of the 80 images is the same in all dimensions.

Hmm no I can’t see anything that distinguishes the fields… the only thing is it could be inferred from X and Y positions but it’s not super easy since there could be overlap.

Andor IQ saves a .txt file along with the tif files that details the number/order of fields and montage tiles but it seems like the XML doesn’t know anything about it.

I guess this means I should just write a macro that separates them out first? That’s not a problem, I only was assuming I was missing something and Bio-Formats would do it.

Many thanks for your help again!

Yeah, unfortunately it sounds like Bio-Formats wont be able to automatically separate them and you will have to split them out yourself first. Its a bit of extra hassle but hopefully it can be worked around.

Ok thanks a lot for your help!