[ome-users] Issue opening Leica GSD lif files in bioformats.

[Melissa Linkert]

Hi Ian,

> > According to the Leica LAS software the lif contains 10 entries, whereas
> > bioformats thinks there are 12 entries. Bioformats has repeated entries
> > for the extremely large image sets. Accessing the image sets produces the
> > following symptoms....
> 
> Thank you for the bug report.  I don't believe we have any data from the
> GSD systems, so if you could upload one of the problematic files that
> would be much appreciated.  I will be sending FTP instructions
> privately, as it sounds like those files will be too large for our QA
> system.

Thank you for uploading a file privately, and my apologies for the delay.

I can confirm this problem, and there is now a ticket for it on our
issue tracking system:

https://trac.openmicroscopy.org.uk/ome/ticket/12585

You have been CC'd on the ticket, and dso will be automatically notified 
of any changes.  If you prefer not to receive notifications, please let us
know

Regards,
-Melissa

On Tue, Sep 23, 2014 at 02:26:02PM -0500, Melissa Linkert wrote:
> Hi Ian,
> 
> > I have found a rather annoying bug when opening .lif files from the
> > relatively new GSD system. We are trying to do two colour STORM and have
> > lif files with two extremely large time series (10,000 - 20,000 frames
> > each). Fiji copes fine with the first set of images but doesn't get the
> > second right. Additionally we seem to get extraneous data sets added to
> > the lif in the bioformats series options dialogue. 
> > 
> > Unfortunately this is on data from a demo session with the GSD so all we
> > have are the extremely large test data sets, 2GB +, although I'm pretty
> > sure size is not the issue. I can easily provide the data but I can
> > provide a smaller example, I'm sorry. 
> > 
> > According to the Leica LAS software the lif contains 10 entries, whereas
> > bioformats thinks there are 12 entries. Bioformats has repeated entries
> > for the extremely large image sets. Accessing the image sets produces the
> > following symptoms....
> 
> Thank you for the bug report.  I don't believe we have any data from the
> GSD systems, so if you could upload one of the problematic files that
> would be much appreciated.  I will be sending FTP instructions
> privately, as it sounds like those files will be too large for our QA
> system.
> 
> Before uploading, though, could you please confirm that your Fiji is up
> to date ("Help > About Plugins > Bio-Formats Plugins" should show
> version 5.0.4)?
> 
> Regards,
> -Melissa
> 
> On Tue, Sep 23, 2014 at 07:43:05PM +0100, Ian Dobbie wrote:
> > Hi Everyone, 
> > 
> > I have found a rather annoying bug when opening .lif files from the
> > relatively new GSD system. We are trying to do two colour STORM and have
> > lif files with two extremely large time series (10,000 - 20,000 frames
> > each). Fiji copes fine with the first set of images but doesn't get the
> > second right. Additionally we seem to get extraneous data sets added to
> > the lif in the bioformats series options dialogue. 
> > 
> > Unfortunately this is on data from a demo session with the GSD so all we
> > have are the extremely large test data sets, 2GB +, although I'm pretty
> > sure size is not the issue. I can easily provide the data but I can
> > provide a smaller example, I'm sorry. 
> > 
> > According to the Leica LAS software the lif contains 10 entries, whereas
> > bioformats thinks there are 12 entries. Bioformats has repeated entries
> > for the extremely large image sets. Accessing the image sets produces the
> > following symptoms....
> > 
> > 
> > 1) Selecting the first repeat of the first large image works fine.
> > 
> > 2) The second repeat of the first large image set opens junk data. Some
> > regions appear to have diagonal stripes like it is trying to read data
> > with the wrong xy size, and probably starting at the wrong byte boundary.  
> > 
> > 3) The first repeat of the second large image set, incorrectly has the
> > number of frames from the first large image set. If this is larger than
> > the real number of frames then you can open it, get reasonable data
> > until the end of the expected data and then junk. If the first image set
> > is smaller than the second then you truncate the image set when you open
> > it.
> > 
> > 4) opening later image sets which are labelled with the correct number of
> > frames fails, but I think that bioformats is trying to read beyond the end
> > of the lif file. 
> > 
> > Thanks, 
> > 
> > Ian
> > 
> > -- 
> > Ian Dobbie
> > Micron Bioimaging Unit Manager,
> > Biochemistry,
> > University of Oxford,
> > South Parks Road,
> > Oxford
> > OX1 3QU
> > Tel: 01865 613323
> > Email: ian.dobbie at bioch.ox.ac.uk
> > _______________________________________________
> > ome-users mailing list
> > ome-users at lists.openmicroscopy.org.uk
> > http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users