[ome-users] [Software Feedback] MATLAB bfopen discards series names

[Curtis Rueden]

Hi Allon,

> The code you sent works with a minor modification: it is necessary to
> convert the name into a string using char().

Great, thanks for the information. Perhaps Sébastien (the primary bfopen
maintainer) will integrate your changes.

One word of caution: it occurred to me after sending that code snippet that
there may be a discrepancy between the Bio-Formats/Java 0-indexing and
MATLAB's 1-based indices. Off the top of my head, I am not sure whether
"getImageName(i)" should actually be "getImageName(i-1)" but would be worth
double checking, just to be sure.

Regards,
Curtis


On Thu, Jun 13, 2013 at 1:42 PM, Allon Klein <Allon_Klein at hms.harvard.edu>wrote:

> Hi Curtis,
>
> Many thanks for the quick response. The code you sent works with a minor
> modification: it is necessary to convert the name into a string using
> char().
>
> The code below adds the sequence name to the label without deleting any of
> the other information that bfopen helpfully includes. I added it to bfopen
> right after:
>         % build an informative title for our figure
>         label = id;
>
> % NEW CODE
>  % add series name, if available
>  sequenceName = char(r.getMetadataStore().getImageName(i));
> if ~isempty(sequenceName)
> label = [label '; ' sequenceName];
>  end
>
> Thanks again,
> Allon
>
>
> On 13 Jun 2013, at 13:24, Curtis Rueden wrote:
>
> Hi Allon,
>
> I am CCing the ome-users list, which is the best place to ask Bio-Formats
> questions.
>
> > I just started using your matlab functions with loci-tools.jar. Thanks
> > for making a wonderful tool!
>
> Glad to hear it is working for you.
>
> > I noticed a problem, which I suspect has a quick fix. When loading
> > multiple series from a .lif file with bfopen, the image "labels" do
> > not include the original series title. Instead, the bfopen script
> > creates a "meaningful" label that includes information such as "series
> > 1/23", but not the actual series label.
>
> Indeed, it happens here:
>
>
> https://github.com/openmicroscopy/bioformats/blob/v4.4.8/components/bio-formats/matlab/bfopen.m#L200
>
> > This is clearly a matlab-implementation specific issue, because the
> > series names are correctly recognized in the imageJ importer.
>
> Right.
>
> > I wonder whether you can suggest a quick hack to add this information?
> > I tried looking up a user guide for loci-tools.jar, to see what
> > command might retrieve the series names. But I couldn't find any
> > information.
>
> Sure, the code would be something like:
>
> # override label with series name, if available
> imageName = r.getMetadataStore().getImageName(i);
> if imageName ~= ""
>   label = imageName;
> end
>
> But I have not tested this, so it may need minor tweaking.
>
> Regards,
> Curtis
>
>
> On Thu, Jun 13, 2013 at 9:13 AM, <allon_klein at hms.harvard.edu> wrote:
>
>> Allon Klein sent a message using the contact form at
>> http://loci.wisc.edu/contact.
>>
>> Hi,
>>
>> I just started using your matlab functions with loci-tools.jar. Thanks
>> for making a wonderful tool!
>>
>> I noticed a problem, which I suspect has a quick fix. When loading
>> multiple series from a .lif file with bfopen, the image "labels" do not
>> include the original series title. Instead, the bfopen script creates a
>> "meaningful" label that includes information such as "series 1/23", but not
>> the actual series label.
>>
>> This is clearly a matlab-implementation specific issue, because the
>> series names are correctly recognized in the imageJ importer.
>>
>> For me having the series names is critical, since each series refers to a
>> different condition and the details are stored in the names.
>>
>> I wonder whether you can suggest a quick hack to add this information? I
>> tried looking up a user guide for loci-tools.jar, to see what command might
>> retrieve the series names. But I couldn't find any information.
>>
>> Thanks very much, in advance, for your help!
>>
>> Best wishes,
>> Allon Klein
>> Systems Biology
>> Harvard Medical School
>>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openmicroscopy.org.uk/pipermail/ome-users/attachments/20130613/af69704d/attachment.html>