[DAS2] Notes from the weekly DAS/2 teleconference, 20 Nov 2006

[Steve Chervitz]

[Note: No DAS teleconference on 27 Nov. Next one is on 4 Dec]

Notes from the weekly DAS/2 teleconference, 20 Nov 2006

$Id: das2-teleconf-2006-11-20.txt,v 1.1 2006/11/27 04:31:31 sac Exp $

Note taker: Steve Chervitz

Attendees: 
  Affy: Steve Chervitz, Gregg Helt, Ed Erwin
  Dalke Scientific: Andrew Dalke
  UCLA: Brian O'connor

Action items are flagged with '[A]'.

These notes are checked into the biodas.org CVS repository at
das/das2/notes/2006. Instructions on how to access this
repository are at http://biodas.org

DISCLAIMER: 
The note taker aims for completeness and accuracy, but these goals are
not always achievable, given the desire to get the notes out with a
rapid turnaround. So don't consider these notes as complete minutes
from the meeting, but rather abbreviated, summarized versions of what
was discussed. There may be errors of commission and omission.
Participants are welcome to post comments and/or corrections to these
as they see fit. 


Agenda
-------
 * Spec discussion
 * Status reports


Topic: Spec Discussion
-----------------------
gh: has everyone reviewed steve's re-orged html doc of the retrieval
spec? 

[To summarize the re-org: everything has been organized into three
main sections: general, overview, and detailed. There's a table of
contents and all sections and subsections are numbered a la W3C specs.
The summary table at the top has been simplified, and in-page
navigation has been fixed and improved.]

consensus: no, haven't looked at in detail yet.

[A] all give the new html retrieval spec doc a read through again.

gh: need to add examples of alignments.
I plan to announce in the next two weeks that it's ready.
When reviewing, pay special attention to comments marked 'XXX'
sc: esp in third para from the top, ambiguousness.

ee: need commit privileges
sc: sent email to support at open-bio.org. They are quick.

Topic: Status reports
---------------------

gh: First half of last week, getting affy das2 server up to snuff with
spec changes and compliance with spec, had spotty
compliance. correctly handle errors. email message about test server,
different url than the public server.

[A] gregg/steve to update public affy das2 server with new IGB release

gh: lots of testing in last few days, ready to replace the existing server.
ee: some zooming issues when switching servers.

gh: worked on fixing bugs in das/2 client, problems w/ biopackages
server due to problems in IGB, also w/r/t spec changes.

also working on using das/2 in another context in igb, retrieving
genomic locations. Expression console, recommended for processing affy
chips (all expr), generates chp files. big request esp for whole exon
folks, need igb to load these, but the chp files have no genomic
location info, had to pre-load these in past. Now, when you load a
chip file, igb automatically goes out via das2 and retrieves
it. hardwired is what server to go to, based on genome + types from
server, figures out what das request to make to load it. per-sequence
basis, and lazy, doesn't load all locations for whole genome. big
files: 1M probe sets, + 4 probes w diff locs. Also, igb gets it back
in compact binary format (using alt format)  -- new use of das in
igb. not committed yet, but will be cool.

ee: data?
gh: yes. we need more bp2 files there. Will try and have igb prompt
user for file if it can't look it up automatically.

[A] gregg send ed a write up so he can get it in the release notes.

gh: bottom line: efficient way to look at expr data in igb

gh: Third thing: prepping for this release of igb. targetting
wed. server change over tues, igb on wed. may cause a day of hassles.

ee: I'll be working in Paris on Wed, will be Tues day for US..
gh: so I'll be done by noon wed or earlier.

gh: happy with progress on affy client server now.

ad: cleaning up code I did on validator while in EBI. will check into
dasypus CVS on sf. 

ee: on vacation, but working now.

sc: worked on affy das2 server set up for testing
( http://netaffxdas.affymetrix.com/das2/test ) and worked on the igb
keystore update (digital signature for affy jars). Ran into issue with
the error codes sent by the affy das2 server getting altered by apache
into another error (502, I believe). Need to figure out how to get
apache to not alter these error responses.

[A] steve figure out how to prevent apache from changing das/2 error
responses

gh: would assume not doing redirection through apache. but plan b is
to not use apache. [affy das2 server uses jetty servlet engine, and
apache forwards request to it via a rewrite rule.]

sc: quick load access now requires apache.
gh: should be able to load and serve these through jetty running on
port 80. need to get apache to stop mucking with the http headers.

bo: no new progress, but will look into filtering issue this
week. 'so:' stuff. either allen of I will look into it this week.
gh: i get right feat response for some but not all request. wonder if
the 'so:' is involved. that's the only remaining issue I know of.

both servers are passing validation. this was a high priority. for
brian gilman and lincoln to make use of the das/2 spec.

ad: any more feedback from bg?
gh: no. he was going to start working on a server as well.

[A] gregg contact brian gilman to see how things are going.

ee: good news, bug I reported about zooming out was not a bug, but
cause by me pressing the wrong button. related to changing genome
version.

[A] steve set up jar signing cert today so Ed can use tomorrow.

Wrap up:
--------
[A] review and modify das/2 html retrieval docs over the next few days.

[A] Next meeting in two weeks (4 Dec 06)