[ome-users] PerkinElmer Opera flex files import

[Melissa Linkert]

Hi Manuel,

> > I've got some problems importing images acquired with an Opera system from PerkinElmer into Fiji via Bioformats:
*snip*
> If you see the same problem with 4.4.0, then it would be very helpful if
> you could send at least one dataset that exhibits this problem.  If you
> need somewhere to put files, please let me know and I will send
> instructions for connecting to our SFTP server in a private email.

Thank you for providing a dataset privately, and for clarifying that
this problem does occur with the 4.4.0 version.

I have filed a ticket for this on our issue tracking system:

https://trac.openmicroscopy.org.uk/ome/ticket/9375

You have been CC'd, and so will receive an automated email each time we
update the status of the ticket.  If you prefer not to receive these
emails, please let me know.

Regards,
-Melissa

On Wed, Jul 18, 2012 at 09:47:40AM -0400, Melissa Linkert wrote:
> Hi Manuel,
> 
> > I've got some problems importing images acquired with an Opera system from PerkinElmer into Fiji via Bioformats:
> > Images were acquired in 32 Wells of one 64-wellplate,  9 sub-positions in each well, 288 sub-positions in total. For each sub-position one flex file is stored. Each flex file consists of 10 images: 5 z-layers and two color channels. If I open the res file which should contain all the info about the whole image plate (wells, sub-positions acquired) and which is stored together with the flex files, I can specify which series (sub-position) I want to import. If I specify series 1, the first image from the first flex file is imported. Specifying for instance series 1-10 imports the first 9 (not all 10) images of the first sub-position, the 10th image is the first image of the next well. So each time for each sub-position as many images can be imported as sub-positions within the well exist. In my case, 9 sub-positions per well, but 10 images within each flex file, or in other words: from the 90 images acquired in well 1 only the first 9 from the first sub-position can be imported
> >  .
> > If one flex-file is stored separately, without the res file being in the same folder and I import it in Fiji, I can select which series I want to import (up to 9 = 9 sub-pos.). Series 1 is the first z-layer in the first color channel at the sub-pos., series 2 the first z-layer in the second color channel, series 3 the second z-layer in the first color channel and so on. Layer 10 is not selectable, because there are only 9 sub-positions in the well.
> > The only workaround I see for this problem would be to acquire as many sub-positions in one well as there are image planes (z-pos, channels, ...), but this is not always desired. is there a fix? To me it looks like there is some problem with multilayer flex files. With proprietary PerkinElmer software, all image layers are visible (and can be exported to multilayer tiffs which seem to be readable nicely).
> > 
> > I hope you understand what my problem is, I can provide example files if desired!
> 
> It's not clear to me which version of Bio-Formats you are using, so
> could you please try using the latest stable version (4.4.0, from
> http://loci.wisc.edu/bio-formats/downloads) and see if that works?  Note
> that this will be a much newer version than what you see in the Fiji
> updater.  You can also double-check which version you have installed
> using "Help > About Plugins > LOCI Plugins".
> 
> If you see the same problem with 4.4.0, then it would be very helpful if
> you could send at least one dataset that exhibits this problem.  If you
> need somewhere to put files, please let me know and I will send
> instructions for connecting to our SFTP server in a private email.
> 
> Regards,
> -Melissa
> 
> On Wed, Jul 18, 2012 at 01:31:51PM +0200, Manuel Gunkel wrote:
> > Hi,
> > 
> > I've got some problems importing images acquired with an Opera system from PerkinElmer into Fiji via Bioformats:
> > Images were acquired in 32 Wells of one 64-wellplate,  9 sub-positions in each well, 288 sub-positions in total. For each sub-position one flex file is stored. Each flex file consists of 10 images: 5 z-layers and two color channels. If I open the res file which should contain all the info about the whole image plate (wells, sub-positions acquired) and which is stored together with the flex files, I can specify which series (sub-position) I want to import. If I specify series 1, the first image from the first flex file is imported. Specifying for instance series 1-10 imports the first 9 (not all 10) images of the first sub-position, the 10th image is the first image of the next well. So each time for each sub-position as many images can be imported as sub-positions within the well exist. In my case, 9 sub-positions per well, but 10 images within each flex file, or in other words: from the 90 images acquired in well 1 only the first 9 from the first sub-position can be imported
> >  .
> > If one flex-file is stored separately, without the res file being in the same folder and I import it in Fiji, I can select which series I want to import (up to 9 = 9 sub-pos.). Series 1 is the first z-layer in the first color channel at the sub-pos., series 2 the first z-layer in the second color channel, series 3 the second z-layer in the first color channel and so on. Layer 10 is not selectable, because there are only 9 sub-positions in the well.
> > The only workaround I see for this problem would be to acquire as many sub-positions in one well as there are image planes (z-pos, channels, ...), but this is not always desired. is there a fix? To me it looks like there is some problem with multilayer flex files. With proprietary PerkinElmer software, all image layers are visible (and can be exported to multilayer tiffs which seem to be readable nicely).
> > 
> > I hope you understand what my problem is, I can provide example files if desired!
> > 
> > Best regards, I'm thankful for any help
> > 
> > Manuel
> > _______________________________________________
> > ome-users mailing list
> > ome-users at lists.openmicroscopy.org.uk
> > http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users