|
Caltech Update - Diane Trout
Accomplishments:
-Unhacked some workstations :(
-Discovered Poseidon 1.5.1 is no longer using the novosoft UML reader
instead they're using one from the netbeans project. (it supports a
newer version of the XMI/UML file format).
-Fixed bug in pymerase that prevented the bool type from being saved
to the database correctly
Next Week:
-Fix several bugs that brandon found.
-Choose and start implementing a solution to UML parsing.
Problems:
-Dealing with the switch of UML parsers
--Ended up being distracted by trying to understand netbeans.
--I'm getting tempted to extend pymerase to support generating C++
classes. Using the current UML reader I could hopefully generate C++
classes representing the UML specification.
Add a little XML parsing and some swig wrapping and I'm hopefully
free of this annoying java dependency with my own scripting language
friendly UML parser.
UVa Update - Jodi Kanter
Accomplishments this week:
-spent a lot of time on schema design for two other projects.
- updating Va-GeneX schema drawings for web site.
- ongoing Va-GeneX user assistance with training, accounts, data load, etc..
- first historical data load to Va-GeneX complete.
For Next Week:
- clean up schema SQL script to reflect tables, indexes, etc.. actually in use.
- focus on Postgres and query performance
- work on adding foreign key constraints, necessary indexes, etc. to GeneX schema
UCI Update - Harry Mangalam
Updates:
Over Holidays did not do much of anything :) & :(
Recently, used indexing, and experimented with float -> int conversion to represent data in a custom table to increase speed of query > 1000X on GMU server on some queries. Also learned many unpleasant but useful things about GeneSpring, DB drivers, PostgreSQL, Perl, & Python.
GeneSpring connectivity is back (SiG changed the way it distributes it over the holidays) and once Jennifer tells me more about their Genome format, we can start putting together a complete connection package (finish HOWTO, provide example views and tables to query with it, etc).
Working on:
- Learning about Python, Qt, & SWIG (thanks, Diane), so as to be able to design independent apps to connect the DB to other apps.
- continuing to optimize the DB so as to make it responsive enough for real work (in realish time).
- writing some PG routines to do the table conversion in-place rather than exporting tables, converting with Pl/Py, then importing them (latter is site dependent; former will be site-independent).
--
Note: Michael Pear, an old-time GeneX contributor who went to work for ISIS's GeneTrove div contacted me to tell me that they have decided not to release the SW as a separate product, but will instead be Open-Sourcing it (the value to them is the content, not the SW). It's a relatively generic gene expression DB browser and query composer, written in Java, for MySQL, but relatively agnostic. I'll be meeting with them at the O'Reilly BioCon in San Diego Feb 3-6 to get the details, but it sounds like it might be worth looking into as another front-end for GeneX.
I'll post a follow-up abstract from their group in another note.
Caltech Update - Brandon King
Accomplishments:
-Unhacked two computers
-Discoverd the cause thanks to Diane and a few others (Affymetrix software installs MSDE with a huge security hole!)
-Documented how to secure a computer with Affymetrix software installed.
Next Week:
-Back to auto-generating python/tk gui library using pymerase
Problems:
-Windows and Security, or lack there of!
|
