[ome-users] OMERO

[Andrew Patterson]

Hello Kees,

I am copying this discussion onto the OME-USERS mailing list.

I think some of your questions are general enough to be of interest to  
others and we may get some other advice from our users.

On 24 Mar 2009, at 17:58, Straatman, Dr K.R. wrote:
>
> I run the advanced light microscopy facilities in the University of  
> Leicester, UK. We have microscopes in different buildings and will  
> expand further during this year. Therefore, I look for a way to  
> store data and have easy access to the data at multiple locations.  
> We have some funding for a web server and I have been testing OME  
> and OMERO in collaboration with the computer centre. At the moment  
> we have around 55 active users.

> From what I have seen OMERO seems to be easier to use. However, I  
> have some questions/problems and I am not sure where to ask them as  
> most of the OME website seems to focus on OME and not OMERO.

The OMERO section for the site is located at:
http://www.openmicroscopy.org/site/documents/data-management/omero
We are preparing an update to this section at the moment to go along  
with the release of OMERO4.

A lot of the data links here go to our Trac systems. This is updated  
by the developers and has general documentation as well as detailed  
work tickets and version information.
>
> May be you can direct me in the right direction or you might be able  
> to answer some of the questions
> -          I was able to upload images to the database, but I seem  
> not to be able to delete images. Is that correct?

Deletion is present in OMERO 4 which is being packaged for released as  
we speak.

> -          I have the same problem with Tags. I have created some  
> tags and they are partly nested where I  wanted single layers (don’t  
> know how this happened yet). However, I don’t seem to be able to  
> undo this.

We have tagsets in OMERO.insight. We don't have a flattened view of  
them, they remain in their tagset in the tag tab of the datamanager.  
When adding a new tag to an object it shows all tags flattened.  
Searching will search on a tagset or tag regardless.
Tags don't have to belong to a tagset just as datasets don't have to  
belong to projects anymore. Tagsets are just a tag of a tag.

> -          How can I do a search for images with the same Tag?

There is a search option at the top of the window in the new  
OMERO.insight client. This will search within your current datasets  
for tags, ratings, comments, attachments, etc.

You can also change from a top level project->dataset view to a top  
level Tag view. This allows you to look at all the images you tagged  
with a keyword like "metaphase" for example.

> -          Is it possible to move data within the database. For  
> example, a PI has over the years several students working on the  
> same project, each student with his/her own database under the PIs  
> account and wants to put all the data in one collection and not  
> dependent on the students anymore.

Each user has there own account. Several accounts can belong to the  
same group so they can see each other's data. Using the current  
clients you cannot move the data from belonging to one user to  
another. It would be possible to do this in the server using custom  
code against the API or a script running on the DB directly. Using the  
API would be the safer option.

> Or does this has to be solved by choosing the right tags?

Not really, as I think you are only viewing the Tags for one user  
account at a time.

> -          In our situation we would image on a microscope, add the  
> images to the database and move to a different location to analyse  
> the images. This could be ImageJ, Volocity, Imaris, Huygens  
> deconvolution, or any other software package needed. As far as I can  
> see there seem not to be any option  to do this. I tried the ImageJ  
> OME download option but it failed. Would this be possible and how  
> would this work?

The ImageJ OMERO plug-in is being re-written to work with the new  
server at the moment.

> -          Would it be possible to write images directly from the  
> microscope into the database? This would save a lot of time.

We have a product called OMERO.fs that will watch directories on a  
file-system and import the files automatically to the OMERO.server.  
The resulting import is a metadata only import with the pixel data  
remaining in the original file on the file-system. Thus, the original  
file will be reread by OMERO.fs if pixel data for the image is  
requested. The OMERO.fs process can either be on a different machine  
to the OMERO.server process or the same machine.

> -          What if the analysis software does not read OME-tiff  
> files? For example, if we want to analyse data in the future with  
> not yet developed software?

An OME-TIFF file conforms to the TIFF standard so can be read by any  
application that can read a TIFF. If the application does not  
understand OME-TIFF it will not see the metadata in the file and may  
or may not preserve it.

If you are developing you own software you can use our Bio-Formats  
Java library to read/write OME-TIFF.

> Are there somewhere options to upload the data in its original  
> format? Or will other software just read the tiff information?

You can at import time archive the original file on the server. The  
OMERO.insight client will then let you retrieve this original format  
file at a future date. This means the OMERO.server can also function  
as your main image archive. This does use additional disk space so may  
or may not make sense depending on you IT setup.

> -          What is the support in the community for the OMERO format?
> 6-7 years ago we changed images to the ICS/IDS format,
> but never got funding for a server system and now there seems no  
> development on the SIDB anymore. As we have many users I need a  
> system that is as much future proof as possible.

OMERO can read ICS/IDS files.

This is a hard question to answer but I think we are doing well. For  
example, just looking at some stats I see we have over 500 unique  
server instances that have been started up in the last year.

We have good funding in place for the next few years and have no  
intention to stop development any time soon. We have had several  
successful collaborations with academia and also with companies and  
publishers through our commercial partner Glencoe Software. See:
http://glencoesoftware.com/
for details about Glencoe, and:
http://jcb-dataviewer.rupress.org/
for a good example of something which utilises parts of OMERO.server /  
OMERO.web in a commercial application. Some of our other commercial  
partners include PerkinElmer and Deltavision.

Hope this helps,

Andrew

Andrew Patterson
ajpatterson at lifesci.dundee.ac.uk
Software Developer, Open Microscopy Environment
Wellcome Trust Centre for Gene Regulation and Expression, University  
of Dundee