gmane.science.biology.gmod.apollo

New WebApollo release (WebApollo-2013-05-16)

Ed Lee — 2013-05-17T18:06:42

Hello everyone,

We'd like to announce a new WebApollo release.  You can get the newest
release at:

http://genomearchitect.org/webapollo/releases

A public demo with the newest release can be accessed at:

http://genomearchitect.org/WebApolloDemo

The updated server setup instructions can be accessed at:

http://www.gmod.org/wiki/WebApollo_Installation

The updated user guide can be accessed at:

http://genomearchitect.org/webapollo/docs/webapollo_user_guide.pdf

Here's a summary of changes:

WebApollo 2013-05-17 release:
+ Fully revamped genomic sequence selection screen
        + sorting by name and length
                + uses customizable JavaScript function
        + filtering of genomic sequence names
+ Annotation info editor
        + allows editing of symbols
        + editing of comments
        + editing of dbxrefs
+ Script for bulk loading gene/transcript/exons to annotation track
+ Improved login system
        + allows logging in from either genomic region selection screen or
editor
+ Configur

RE: WebApollo and Postgres 9.2

Shane Brubaker — 2013-05-15T21:43:09

Thanks much!


Sincerely,

Shane Brubaker
Director of BioInformatics
Solazyme, Inc.
225 Gateway Blvd.
S. San Francisco, CA 94080

From: elee-/3juihCSby0< at >public.gmane.org [mailto:elee-/3juihCSby0< at >public.gmane.org] On Behalf Of Ed Lee
Sent: Wednesday, May 15, 2013 2:39 PM
To: apollo-M45m+dTNDDscvh0c176lvQ< at >public.gmane.org
Subject: Re: [apollo] WebApollo and Postgres 9.2

Hi Shane,

We haven't tested with 9.2 (works fine up to 9.1) but it sounds like adding the schema to the search path in your PostgreSQL configuration might resolve this issue.

Look at the documentation for more information:

http://www.postgresql.org/docs/9.2/static/runtime-config-client.html

Cheers,
Ed

On Tue, May 14, 2013 at 3:12 PM, Shane Brubaker <sbrubaker-SB+zGAu0R7RWk0Htik3J/w< at >public.gmane.org<mailto:sbrubaker-SB+zGAu0R7RWk0Htik3J/w< at >public.gmane.org>> wrote:
Hi, I was just wondering if WebApollo works with Postgres 9.2.  I've been using some of the setup scripts, and it seems that 9.2 requires that everything be prefaced by a schema

Re: WebApollo and Postgres 9.2

Ed Lee — 2013-05-15T21:39:07

Hi Shane,

We haven't tested with 9.2 (works fine up to 9.1) but it sounds like adding
the schema to the search path in your PostgreSQL configuration might
resolve this issue.

Look at the documentation for more information:

http://www.postgresql.org/docs/9.2/static/runtime-config-client.html

Cheers,
Ed


On Tue, May 14, 2013 at 3:12 PM, Shane Brubaker <sbrubaker-SB+zGAu0R7RWk0Htik3J/w< at >public.gmane.org>wrote:

WebApollo and Postgres 9.2

Shane Brubaker — 2013-05-14T22:12:20

Hi, I was just wondering if WebApollo works with Postgres 9.2.  I've been using some of the setup scripts, and it seems that 9.2 requires that everything be prefaced by a schema name and so that is causing errors.

Re: next release of Web-Apollo

Ed Lee — 2013-05-10T19:38:20

Hi Jeff,

Is there a hard date for the next release of Web-Apollo, and a list of

We're shooting for a release in the very near future (in the next few days)


That will be included in this release.


Any GFF3 loaded to the user annotation track will preserve non-GFF3 tags
from column 9 (although you currently won't be able to view them until
exporting the data).


Won't make it this release, we're hoping to have that for the next release
(which we're shooting for early August).

Cheers,
Ed

next release of Web-Apollo

Hoover, Jeffrey — 2013-05-10T19:10:44

Is there a hard date for the next release of Web-Apollo, and a list of features to be included?

We are particularly interested in three things:

1.       scripts to pre-load the annotation tier with our existing gene models.

2.       being able to  load and preserve meta-data with the gene models, such as our local identifier and a comment

3.       configurable codon tables

What's your best estimate for the next release date and expected features?

Jeff

Re: Multiple organisms?

Ed Lee — 2013-05-10T17:43:32

Hi Jim,

The latter would work as well and it might be feasible for smaller genomes
with a small number of genomic regions (a best scenario being one genomic
region per organism e.g., chromosome).  I'd imagine that if you're dealing
with large genomes with 1000s of genomic regions (e.g., scaffolds), using
that approach could become somewhat cumbersome.  If you go with that
approach, there's an undocumented way to configure your instance so each
"chromosome" can have its own organism (which is used in the genomic region
selection screen before launching the editor).

One caveat with the the latter approach is that you'll want to make sure
that any user uploaded data is preprocessed as well.

Cheers,
Ed


On Fri, May 10, 2013 at 10:20 AM, Jim Hu <jim.hu.biobio-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

Re: Multiple organisms?

Jim Hu — 2013-05-10T17:20:32

I was thinking either that or ... because they are little phage genomes, I'm wondering if I could catenate in some way to treat them like separate chromosomes and then decatenate on export.

Jim

On May 10, 2013, at 12:17 PM, Ed Lee wrote:

Re: Multiple organisms?

Ed Lee — 2013-05-10T17:17:13

Hi Jim,

For now I'd suggest having a separate instance per organism and then
perhaps setting up a proxy if you wanted to access all organisms from a
single entry point.

Cheers,
Ed


On Fri, May 10, 2013 at 10:10 AM, Jim Hu <jim.hu.biobio-Re5JQEeQqe8AvxtiuMwx3w< at >public.gmane.org> wrote:

Re: Multiple organisms?

Jim Hu — 2013-05-10T17:10:57

Is there a suggested workaround? This will make it useless to us for our phage annotation plans where we want to do dozens to hundreds in parallel as community annotation projects.

Jim

On May 10, 2013, at 11:59 AM, Ed Lee wrote:

Re: Multiple organisms?

Ed Lee — 2013-05-10T16:59:00

Hi Scott,

We currently don't support multiple organisms in a single instance.
 However, that's something that is on our plate for sometime in the future.

Cheers,
Ed


On Thu, May 9, 2013 at 1:20 PM, Scott Cain <scott-XkniDE3L5z0cWVvVuXF20w< at >public.gmane.org> wrote:

Multiple organisms?

Scott Cain — 2013-05-09T20:20:51

Hi All,

Can WebApollo be configured to work with multiple organisms?  I was just
chatting with someone who wants to use GMOD in the Cloud with WebApollo and
put 5-10 newly sequenced genomes on it and set his experts loose on it.
I'm guessing you could just make multiple WebApollo directories under
tomcat, but that is somewhat unattractive (not to mention difficult to do
in the context of GMOD in the Cloud but that's my problem).

Thanks,
Scott

Re: Adding Scaffolds does not update selectTrack.jsp

Hans Vasquez-Gross — 2013-05-09T19:47:03

Hi Ed,

    I had a PermGenSpace error in the logs.  What is the MaxPermSize you
are using for tomcat?  Currently, our tomcat7 installation is configured
with -XX:MaxPermSize=512M, but I'm debating bumping this up to 1G.   In the
mean time, restarting the tomcat7 service on the server fixed the load
time.  It's back down to ~2-4 seconds.  Thanks for the help.

Cheers,
-Hans

Re: Adding Scaffolds does not update selectTrack.jsp

Ed Lee — 2013-05-09T19:42:29

Hi Hans,

It shouldn't take very long.  Do you see any error messages in the Tomcat
logs?

Cheers,
Ed


On Thu, May 9, 2013 at 12:28 PM, Hans Vasquez-Gross <
havasquezgross-ZnEz5tD0I2KVc3sceRu5cw< at >public.gmane.org> wrote:

Re: Adding Scaffolds does not update selectTrack.jsp

Hans Vasquez-Gross — 2013-05-09T19:28:04

Thanks Ed.   After running add_tracks and set_track_permissions, the second
track is showing up in the selectTrack.jsp page.   However, now after
clicking edit, it takes orders of magnitudes longer to load.  Previously,
the page loaded in ~1second.  Now, it takes about 80-100 seconds.  Is this
because of authentication?  Below you can see my Web Console output for the
Jbrowse window that opens after clicking 'Edit' for a scaffold.

Thanks,
-Hans

[12:21:15.579] Use of getAttributeNode() is deprecated. Use getAttribute()
instead. < at >
http://sswap.dendrome.ucdavis.edu/WebApollo/jbrowse/src/dojo/dom-attr.js:163
[12:21:15.579] Use of attributes' specified attribute is deprecated. It
always returns true. < at >
http://sswap.dendrome.ucdavis.edu/WebApollo/jbrowse/src/dojo/dom-attr.js:164
[12:21:15.968] GET
http://jbrowse.org/analytics/clientReport?ver=webapollo_1.0&refSeqs-count=2&refSeqs-avgLen=8211128&tracks-count=4&plugins=WebApollo&scn-h=1050&scn-w=1680&win-h=604&win-w=1580&el-h=604&el-w=1580&t=1368127275.652&tzoffse

Re: WebApollo writing to Chado

Scott Cain — 2013-05-09T19:05:34

Hi Ed,

Yep, it looks like that fixed it.  Now I want to figure out a nice way to
highlight user edited features in GBrowse and JBrowse.  In GBrowse, I can
check for the featureprop that WebApollo sets via a perl callback, but I'm
not sure what I'd do in JBrowse.  If WebApollo set the features source (ie,
what you get from GFF column 2), I'd be able to create a track that had
just WebApollo created features.  The GFF source is stored as a dbxref with
db.name='GFF_source' and is linked to the feature via feature_dbxref.  It's
not a high priority, but it would be nice.

Scott



On Thu, May 9, 2013 at 2:50 PM, Ed Lee <elee-TVLZxgkOlNW371VqleWjrQ< at >public.gmane.org> wrote:

Re: Adding Scaffolds does not update selectTrack.jsp

Ed Lee — 2013-05-09T18:59:23

Hi Hans,

Sounds like it might be a permission issue.  You'll want to either run
add_tracks.pl and set_track_permissions.pl for the newly added scaffold or
use the web interface for user management (with an user with the user_admin
role) then unchecking the read/write permissions for the users you want to
change, clicking "Update", then rechecking the read/write permissions for
those users and then click "Update" again.

If you choose to do the script route, check out the user database section
in the server documentation:

http://www.gmod.org/wiki/WebApollo_Installation#User_database

Cheers,
Ed


On Thu, May 9, 2013 at 11:45 AM, Hans Vasquez-Gross <
havasquezgross-ZnEz5tD0I2KVc3sceRu5cw< at >public.gmane.org> wrote:

Re: WebApollo writing to Chado

Ed Lee — 2013-05-09T18:50:37

Hi Scott,

Yes, I think that should resolve this issue (will probably consider doing
this in the future as well to allow users to configure their own schema if
necessary).

Cheers,
Ed


On Thu, May 9, 2013 at 11:34 AM, Scott Cain <scott-XkniDE3L5z0cWVvVuXF20w< at >public.gmane.org> wrote:

Adding Scaffolds does not update selectTrack.jsp

Hans Vasquez-Gross — 2013-05-09T18:45:07

Hi Ed,

    Last week I loaded up WebApollo and originally loaded a fasta file with
one scaffold.  Today, I just added a second scaffold with GFF3 information.
 All scripts completed successfully (bin/prepare-refseqs.pl, tools/data/
split_gff_by_source.pl, bin/flatfile-to-json.pl, and bin/generate-names.pl)
which then I reloaded the Webapp through the tomcat manager.

   After logging into WebApollo and clicking 'Edit annotations', I am only
provided the option to edit the first scaffold I loaded last week.
 However, once I click edit, I am brought to the JBrowse interface where
both of the scaffolds are available within the JBrowse interface dropdown
menu.  If I select the newly loaded scaffold, all configured gff3 features
are properly being displayed.

Any suggestions on how to debug this would be great.  I took a look at web
console after clicking 'Edit annotations' with no errors being reported
when arriving at the selectTrack.jsp

Thanks,
-Hans

Re: WebApollo writing to Chado

Scott Cain — 2013-05-09T18:34:30

Just to make sure I'm doing the right thing: there are over 200 of the
*.hbm.xml files, and near the top of each is a line like this:

    <class
name="org.gmod.gbol.simpleObject.generated.AbstractPhenotypeDescription"
table="phendesc" schema="public">

and you want me to remove the schema="public" part, right?

Thanks,
Scott



On Thu, May 9, 2013 at 2:27 PM, Scott Cain <scott-XkniDE3L5z0cWVvVuXF20w< at >public.gmane.org> wrote:

Re: WebApollo writing to Chado

Scott Cain — 2013-05-09T18:27:44

Sounds fun!  I'll let you know how it goes :-)



On Thu, May 9, 2013 at 2:20 PM, Ed Lee <elee-TVLZxgkOlNW371VqleWjrQ< at >public.gmane.org> wrote: