Microarray users eager to integrate their lab’s data storage and analysis have a new platform available: The BioArray Software Environment (BASE), developed by Carsten Peterson and his colleagues at Lund University in Sweden. Its main attraction: It’s free. “If you have a small lab, you don’t have to buy anything but a naked computer,” said Peterson, a professor of theoretical physics who also develops algorithms for microarray analysis.
BASE consists of a MySQL relational database, an application layer, and an Apache web server. Its structure is described in some detail in a software report published in Genome Biology last week; the system was developed on Linux and provides a plug-in architecture for data transformation, data viewing, and analysis modules. In addition, it has array-production LIMS features that can be integrated with the data analysis. BASE supports the Microarray Gene Expression Database group’s MIAME (minimum information about a microarray experiment) standard.
Peterson, who has been working closely on microarray projects with Lund University’s department of oncology for several years, said he first sensed a need for such a comprehensive system about a year and a half ago. Unable to find a suitable platform among either commercial or freely available software, he decided to write his own. Most of the work was done by three graduate students who started the project last summer. “Initially it was only intended for internal use,” Peterson recalled, but late last year they decided to make it available to others under the GNU general public license. Eight academic beta testers in several countries signed on in January and have since made “significant contributions” to the software, Peterson said.
Since BASE first became generally available about two months ago, it has already attracted a fair amount of interest, indicating the Swedish researchers have hit a nerve: So far it has been downloaded at about 200 locations, most of them in the US and at academic institutions.
Peterson acknowledged that wariness of the GNU GPL may hold some commercial users back from adopting the platform. “It could be that the inclusions they are making [to the software] signals what they are up to in terms of proprietary business,” said Peterson, who nevertheless hopes that some companies might contribute in the future.
And BASE is not the only open source gene expression database and analysis system. GeneX, available under the LGPL, was originally developed by researchers at the National Center for Genome Resources in Santa Fe. GeneX also uses a relational database and an Apache web server as well as a number of analytical routines and visualization tools.
In the commercial sector, BASE might compete with products like PerkinElmer Life Sciences’ ArrayInformatics, which is priced at $20,000 for a “personal edition.” “[BASE] certainly has some potential of doing the kinds of functions, providing the tools that microarray researchers are looking for,” said Dile Holton, product manager for ArrayInformatics. “Its capabilities and how they have drawn the schema and the things they want to track; absolutely, we are doing the same thing,” he said. However, he cautioned that BASE has yet to prove its compatibility beyond Linux-based systems.
A new version of ArrayInformatics, coming out in early September, will allow data to stream directly from spotters and scanners into the database and will include some other new features like principal component analysis, Holton said.
But BASE is constantly evolving as well: Users have already contributed clustering tools, for example, and within a few weeks, BASE will support the MAGE-ML standard. Also, although the system can be used for all types of microarrays already, future alterations will “make it a little easier for Affy users” and support non-gene chips like tissue arrays better, Peterson said.
BASE 1.0.3 is available for download at base.thep.lu.se.