[ome-users] BioFormats Bug report: MetaMoprh .nd files parsing incorrectly

[Melissa Linkert]

Hi Wendy,

> I just found a bug related component parsing when importing files from a MetaMorph .nd file. When we try to open one of our files by loading the .nd file into Fiji, it shows the right parameters (number of channels, number of time points and positions) and shows a thumbnail image for each position, but the images that appear are not parsed into the correct parameters. In this case, a time series with 4 channels, multiple positions and no z series opens as a single stack with only three of the channels interlaced--the forth channel is missing. 
> 
> 
> I eventually figured out that the individual channel Z stack designations is the root cause. The .nd file for this dataset shows "WaveDoZ1" "WaveDoZ2" and "WaveDoZ3" set to TRUE, WaveDoZ4 set to FALSE and "DoZSeries" set to FALSE. The images that open correspond to the wavelengths that have a TRUE value, indicating that the "WaveDoZ#" value is over-ruling the "DoZSeries" value. And indeed, c hanging the "WaveDoZ#" values all to FALSE restores proper image parsing. 
> 
> 
> 
> I uploaded the original .nd file (B+T.nd), the 4 images from the first position and time point (B+T_w1...tif, etc), and a version of the .nd file that has the "TRUE" values reset to "FALSE" (B+T_adjusted). 

Thank you very much for the bug report.  Unless I misunderstand the problem, this
dataset seems to open correctly using Bio-Formats 4.4.9 and the original
.nd file.  Could you please confirm that Fiji is up to date, and that the
version of Bio-Formats being used is 4.4.9?  "Help > About Plugins >
LOCI Plugins" in Fiji should show the version number.

Regards,
-Melissa

On Mon, Oct 28, 2013 at 05:44:06PM -0400, Wendy Salmon wrote:
> Hi, 
> I just found a bug related component parsing when importing files from a MetaMorph .nd file. When we try to open one of our files by loading the .nd file into Fiji, it shows the right parameters (number of channels, number of time points and positions) and shows a thumbnail image for each position, but the images that appear are not parsed into the correct parameters. In this case, a time series with 4 channels, multiple positions and no z series opens as a single stack with only three of the channels interlaced--the forth channel is missing. 
> 
> 
> I eventually figured out that the individual channel Z stack designations is the root cause. The .nd file for this dataset shows "WaveDoZ1" "WaveDoZ2" and "WaveDoZ3" set to TRUE, WaveDoZ4 set to FALSE and "DoZSeries" set to FALSE. The images that open correspond to the wavelengths that have a TRUE value, indicating that the "WaveDoZ#" value is over-ruling the "DoZSeries" value. And indeed, c hanging the "WaveDoZ#" values all to FALSE restores proper image parsing. 
> 
> 
> 
> I uploaded the original .nd file (B+T.nd), the 4 images from the first position and time point (B+T_w1...tif, etc), and a version of the .nd file that has the "TRUE" values reset to "FALSE" (B+T_adjusted). 
> 
> 
> Please let me know if you need any more information! 
> 
> 
> 
> Thank you, 
> Wendy 
> 
> ~~~~~~~~~~~~~~~~~~~~~~~ 
> Wendy Salmon 
> Light Microscopy Specialist 
> Whitehead Institute for Biomedical Research 
> W.M. Keck Imaging Facility 
> 9 Cambridge Center, Rm 447 
> Cambridge, MA 02142 
> c: 617-429-0158 
> e: wsalmon at wi.mit.edu 
> w: http://staffa.wi.mit.edu/microscopy/ 
> 

> _______________________________________________
> ome-users mailing list
> ome-users at lists.openmicroscopy.org.uk
> http://lists.openmicroscopy.org.uk/mailman/listinfo/ome-users