GEOSS: Gene Expression Open Source System

[Nov 23, 2009 14:09] Web access to Microsoft Live@edu accounts now works.

GEOSS is an Open Source Gene Expression Database supporting storage and analysis of Affymetrix® GeneChip® technology, with planned support for two color spotted arrays and other gene expression measurement technologies. GEOSS is designed to serve as a secure repository and archive for many researchers' data. While we typically expect GEOSS to be installed as part of a microarray center, the software is compact enough to install for a single department or even a single user. GEOSS includes our innovative Analysis Tree package which includes an ever expanding set of analytical tools, and has plug-in architecture allowing easy expansion.

GEOSS Components and Description

GEOSS has a relational database, a web interface, and an analysis suite. There is also a repository for each user's files, and a comprehensive security system. The gene expression workflow begins with a researcher describing an experimental protocol in a minimal fashion. When samples are ready, the researcher creates an order for the microarray research center which at the University of Virginia (UVa) is the Biomedical Research Facility (BRF). The BRF hybridizes the samples as specified by the researcher, and imports the resulting data into GeneX. At this stage the data is available to the researcher. Data can be exported and/or analyzed. Our Analysis Tree package allows users to build a flow chart (yes, there is a graphical flow chart on the screen) of routines to analyze data and produce reports. Except for system administrators, all interaction with the system is via secure web pages. No special software is required for end users. All data is warehoused on powerful, secure servers.

Data is loaded by curators (personnel in the microarray center) via a web interface. Data is visible only to the researcher who owns it, unless the researcher explicitly allows other permissions. The security system allows all users to create groups, to control group membership, and to enable/disable group read and/or write permissions. Permissions apply separately to studies/experimental conditions, orders, data, and derived data (files).

Our Analysis Tree allows the user to build a graphical representation of the flow of data through various modules of an analysis. This system will be documented elsewhere.

The servers are running Fedora Linux. GEOSS is written in Perl. Most of the analysis routines are written in R, some in C, and some in Perl. We use PostgreSQL as our relational database. Its abilities to do transactions, and its high availability were critical to the project. Users must login to the system. All the web pages are accessed via SSL so that all data traveling between the user's web browser and the server is secure from eavesdropping.

Several Perl modules that aren't standard in Perl 5.8 are required. R is required, and of course Apache. All of the Perl used by GEOSS amounts to only 11,000 lines of code. GEOSS is compact. Installation has been vastly simplified and should only take a few hours.

While GEOSS is theoretically portable to Windows NT, 2000, or XP, there are a few aspects that would be nontrivial.

This information, the latest version of GEOSS, instructions, and more can be found on the GEOSS Home Page.

© 2009 by the Rector and Visitors of the University of Virginia.

The information contained on the University of Virginia’s Department of Information Technology and Communication (ITC) website is provided as a public service with the understanding that ITC makes no representations or warranties, either expressed or implied, concerning the accuracy, completeness, reliability or suitability of the information, including warrantees of title, non-infringement of copyright or patent rights of others. These pages are expected to represent the University of Virginia community and the State of Virginia in a professional manner in accordance with the University of Virginia’s Computing Policies.