Hi Greg,

thank you for the explanation, could this also be included somewhere in the wiki? Although I do get far I don't get the tracks to display in my GBrowse. The link does redirect me to the appropriate gbrowse that I specified, the uploaded track is displayed as: "could not fetch data for http://galaxyserver/galaxy/root/display_as?id=4047&display_app=c_elegans..." with the label "galaxyserver/display_as". When I then follow that link I actually see the data as GFF, but it seems GBrowse can't retrieve it somehow.

This was tested under: Galaxy-central - 3528:48e83411aa91 and GBrowse 1.995

Two other things: - under python 2.6 the links appear, under 2.4 they don't. - The links are displayed as "display in wormbase" and the display_app="c_elegans". This seems to be harcoded at galaxy.datatypes.interval line 572, would be nice to have that dynamically generated from for example the ~/tool-data/shared/gbrowse/gbrowse_build_sites.txt

Cheers, Jelle

On Fri, Mar 12, 2010 at 3:25 PM, Greg Von Kuster greg@bx.psu.edu wrote:

This requires only configuration settings, not an actual tool like Epigraph. Displaying in GBrowse is done the same way as displaying in UCSC.

In you universe_wsgi.ini file, look for hte following setting, and set your GBrowse accordingly - you may only want 1 site instead of the 3 in the sample.

# Comma separated list of bx / UCSC / gbrowse / GeneTrack browsers to use for viewing bx_display_sites = main ucsc_display_sites = main,test,archaea,ucla gbrowse_display_sites = main,test,tair

Next, make the setting in ~/tool-data/shared/gbrowse/gbrowse_build_sites.txt - here are the sample settings ( note that each corresponds to what you configured in your universe_wsgi.ini file ). White spaces are tabs in this file, and the first value in each line ( e.g., main, test ) is the link label that points to the GBrowse site whose URL is in the 2nd column. The 3rd column of each line is the valid builds for which the link will be displayed ( see below ). Note that the sample builds are no longer valid, they should be value slike ce2, ce3, etc - see ~/tool-data/shared/ucsc/builds.txt for valid values ).

# wormbase sites / supported genomes main http://www.wormbase.org/db/seq/gbgff/c_elegans/ c_elegans,c_briggsae,c_remanei,c_brenneri,c_japonica,p_pristionchus,b_malayi test http://dev.wormbase.org/db/seq/gbrowse/c_elegans/ c_elegans,c_briggsae,c_remanei,c_brenneri,c_japonica,p_pristionchus,b_malayi tair http://arabidopsis.org/cgi-bin/gbrowse/ arabidopsis_tair8,arabidopsis

When you have the correct configuration settings, you'll see a link in history items ( here is an example - display in Wormbase main ) whose genome build matches one of the configured builds.

On Mar 12, 2010, at 3:27 AM, Jelle Scholtalbers wrote:

Hi,

I would like to be able to send data to my GBrowse2.0 instance. I have been trying some things based on the Epigraph tool but I can't seem to get it work. Below is a mail from Lincoln that was send to the GBrowse mailinglist about how to send data to GBrowse. If its possible, how should I configure the tool?

Cheers, Jelle

---

In version 2.0 it is relatively easy. POST to the GBrowse URL (including the datasource part of the URL) with the following parameters:

• action="upload_file"
• file=<a multipart/form-encoded file upload>
• upload_id=<any numeric id of your choosing>
• id=<the id of the user that you are uploading to>

Alternatively to providing the "file" argument you can provide a "data" argument with the uploaded data provided directly.

The "id" argument indicates the id of the user that you are uploading for. You can get this ID by bookmarking a page -- the "id" parameter records the info. Alternatively, if you don't provide an ID, GBrowse will generate it for you and return the ID in the response cookie, in an argument called "gbrowse_sess".

If you have administrator's mode enabled, then you can provide the ID of the administrator and the uploaded file will become part of the public set of tracks.

Lincoln _______________________________________________ galaxy-user mailing list galaxy-user@lists.bx.psu.edu http://lists.bx.psu.edu/listinfo/galaxy-user

Greg Von Kuster Galaxy Development Team greg@bx.psu.edu

What I now did to circumvent the issue with ldap, is to have this little piece in my httpd.conf: <Location /galaxy/root/display_as> SetEnv AUTHENTICATE_SAMACCOUNTNAME galaxy Satisfy Any </Location> RequestHeader set REMOTE_USER %{AUTHENTICATE_sAMAccountName}e

There is probably a better way, but this does the job for now.

Cheers, Jelle

On Mon, Mar 15, 2010 at 11:42 AM, Jelle Scholtalbers < j.scholtalbers@gmail.com> wrote:

Hi Greg,

I just figured it might have something to do with the proxy.. And indeed removing the proxy/adding exception did work (...), the track is displayed as expected. However this only works for the test case where Galaxy is not behind a LDAP auth. When it is behind LDAP, access is probably denied. I will see what I can do about this, but at least the basis is working :)

Cheers, Jelle

On Mon, Mar 15, 2010 at 10:46 AM, Jelle Scholtalbers < j.scholtalbers@gmail.com> wrote:

...

Hi Greg,

thank you for the explanation, could this also be included somewhere in the wiki? Although I do get far I don't get the tracks to display in my GBrowse. The link does redirect me to the appropriate gbrowse that I specified, the uploaded track is displayed as: "could not fetch data for http://galaxyserver/galaxy/root/display_as?id=4047&display_app=c_elegans..." with the label "galaxyserver/display_as". When I then follow that link I actually see the data as GFF, but it seems GBrowse can't retrieve it somehow.

This was tested under: Galaxy-central - 3528:48e83411aa91 and GBrowse 1.995

Two other things:

• under python 2.6 the links appear, under 2.4 they don't.
• The links are displayed as "display in wormbase" and the

display_app="c_elegans". This seems to be harcoded at galaxy.datatypes.interval line 572, would be nice to have that dynamically generated from for example the ~/tool-data/shared/gbrowse/gbrowse_build_sites.txt

Cheers, Jelle

On Fri, Mar 12, 2010 at 3:25 PM, Greg Von Kuster greg@bx.psu.edu wrote:

...

This requires only configuration settings, not an actual tool like Epigraph. Displaying in GBrowse is done the same way as displaying in UCSC.

In you universe_wsgi.ini file, look for hte following setting, and set your GBrowse accordingly - you may only want 1 site instead of the 3 in the sample.

# Comma separated list of bx / UCSC / gbrowse / GeneTrack browsers to use for viewing bx_display_sites = main ucsc_display_sites = main,test,archaea,ucla gbrowse_display_sites = main,test,tair

Next, make the setting in ~/tool-data/shared/gbrowse/gbrowse_build_sites.txt - here are the sample settings ( note that each corresponds to what you configured in your universe_wsgi.ini file ). White spaces are tabs in this file, and the first value in each line ( e.g., main, test ) is the link label that points to the GBrowse site whose URL is in the 2nd column. The 3rd column of each line is the valid builds for which the link will be displayed ( see below ). Note that the sample builds are no longer valid, they should be value slike ce2, ce3, etc - see ~/tool-data/shared/ucsc/builds.txt for valid values ).

# wormbase sites / supported genomes main http://www.wormbase.org/db/seq/gbgff/c_elegans/ c_elegans,c_briggsae,c_remanei,c_brenneri,c_japonica,p_pristionchus,b_malayi test http://dev.wormbase.org/db/seq/gbrowse/c_elegans/ c_elegans,c_briggsae,c_remanei,c_brenneri,c_japonica,p_pristionchus,b_malayi tair http://arabidopsis.org/cgi-bin/gbrowse/ arabidopsis_tair8,arabidopsis

When you have the correct configuration settings, you'll see a link in history items ( here is an example - display in Wormbase main ) whose genome build matches one of the configured builds.

On Mar 12, 2010, at 3:27 AM, Jelle Scholtalbers wrote:

Hi,

I would like to be able to send data to my GBrowse2.0 instance. I have been trying some things based on the Epigraph tool but I can't seem to get it work. Below is a mail from Lincoln that was send to the GBrowse mailinglist about how to send data to GBrowse. If its possible, how should I configure the tool?

Cheers, Jelle

---

In version 2.0 it is relatively easy. POST to the GBrowse URL (including the datasource part of the URL) with the following parameters:

• action="upload_file"
• file=<a multipart/form-encoded file upload>
• upload_id=<any numeric id of your choosing>
• id=<the id of the user that you are uploading to>

Alternatively to providing the "file" argument you can provide a "data" argument with the uploaded data provided directly.

The "id" argument indicates the id of the user that you are uploading for. You can get this ID by bookmarking a page -- the "id" parameter records the info. Alternatively, if you don't provide an ID, GBrowse will generate it for you and return the ID in the response cookie, in an argument called "gbrowse_sess".

If you have administrator's mode enabled, then you can provide the ID of the administrator and the uploaded file will become part of the public set of tracks.

Lincoln _______________________________________________ galaxy-user mailing list galaxy-user@lists.bx.psu.edu http://lists.bx.psu.edu/listinfo/galaxy-user

Greg Von Kuster Galaxy Development Team greg@bx.psu.edu