[opdev] documentation

2 views
Skip to first unread message

Pierre Lindenbaum

unread,
Dec 12, 2008, 5:31:01 AM12/12/08
to foglio, Simon Heath, opero...@googlegroups.com
Hi All,
I've started to write the documentation about the Operon API (at
least, its architecture)

http://code.google.com/p/polymorphism/wiki/Architecture


Simon, I also forgot to put you in recipients of the mail I sent
yesterday: Basically, Florence asked for a graphical tool that would
help her team to manage some big tables without any command line: A
first result is available here:

http://code.google.com/p/cephlib/wiki/Workbench


Pierre

Simon Heath

unread,
Dec 12, 2008, 5:38:30 AM12/12/08
to Pierre Lindenbaum, opero...@googlegroups.com
Pierre,

For Workbench - the table being worked on is stored in a temporary BerkeleyDB structure?  So in theory the maximum size is limited by the available storage yes?  Could this be used as the core of an analysis system (i.e., identify sample identifiers and phenotype columns, and then allow a set of simple analyses to be run).  If getting the SNP info from UCSC takes time, why not get this information from a local Operon installation?

Simon

Simon Heath

unread,
Dec 12, 2008, 6:03:42 AM12/12/08
to Pierre Lindenbaum, opero...@googlegroups.com
On Fri, Dec 12, 2008 at 11:58 AM, Pierre Lindenbaum <plindenb...@googlemail.com> wrote:
> For Workbench - the table being worked on is stored in a temporary
> BerkeleyDB structure?  So in theory the maximum size is limited by the
> available storage yes?


yes.
And, to be honest, by the way JAVA implemented its widgets: The number
of rows in a table is said to be an integer: so you cannot get a
scrollbar with a maximum-value larger than INT_MAX= 2E31-1

I don't think this should be too much of a limitation...
 




>  Could this be used as the core of an analysis system
> (i.e., identify sample identifiers and phenotype columns, and then allow a
> set of simple analyses to be run).

Yes, no problem at first glance.
But I'd rather see this kind of application as a command line tool...

Yes, but others might prefer a visual tool.  
 


> If getting the SNP info from UCSC takes
> time, why not get this information from a local Operon installation?

Because, anybody in the world can already use this application without operon.

Yes true.  It could always be an option to connect to a local server if available.
 

You can already run this application with java webstart
http://anybody.cephb.fr/perso/lindenb/tinytools/workbench.php (Java
1.6 required)
FYI: Java WebStart is a technology where your java program is
downloaded from a server each time you need it: as a consequence your
application is always up-to-date.

The aim of this tool was to help Florence's team in managing her data
while waiting for a stable version of operon.

Sure
 


Of course I can later add Operon to this tool but I'd rather use the
dedicated graphical tool I've already created (see
http://code.google.com/p/polymorphism/wiki/Architecture  )

OK
 

It would takes time only if you query a about a huge number of snp.


Pierre

Simon
Reply all
Reply to author
Forward
0 new messages