IBM Credits Life Sciences for Growth in Database Market Share
A recent study by market research firm Gartner Dataquest reported that IBM nudged Oracle out of the top spot in the worldwide market for database management systems in 2001.
While the overall DBMS industry experienced flat growth over the course of the year, IBM’s share grew by 4.3 percent, while Oracle witnessed a drop of 4.9 percent.
The report points out that IBM’s acquisition of data management firm Informix in 2001 gave it the edge over Oracle, but IBM also credited adoption of its DB2 system in the life sciences market as a contributing factor.
The Gartner report did not break down the database industry by market segment, but an IBM spokeswoman said the company is witnessing “triple-digit growth” for DB2 in the life sciences. Current customers for the product include Aventis, Eisai Pharmaceutical, Mayo Clinic, MDS Proteomics, NuTec Sciences, Structural Bioinformatics, University of California, San Diego, University of Pennsylvania, and Vitea AB.
While IBM overtook Oracle in the overall database market, Oracle maintained its lead in the relational database segment, which makes up 80 percent of the industry, according to the report.
TimeLogic’s DeCypher Exhibits 82-Fold Speedup over Linux Cluster
In a recent benchmark with the San Diego Supercomputer Center and Sun Microsystems, Crystal Bay, Nev.-based TimeLogic proved that its DeCypher biocomputing accelerator offered an 82-fold speedup against the SDSC’s Linux cluster.
Adam Godzik, a researcher at SDSC in the Joint Center for Structural Genomics, designed the test, which compared a cluster of 32 x 1-GHz Pentium III Linux PCs running HMMER to an 8-CPU Sun Fire 6800 server containing the DeCypher XD-4G FPGA reconfigurable computing processor array.
Using NCBI’s NR75 non-redundant protein database of 372,119 sequences as the query set and SDSC’s HMM database of 19,192 models as the target of the search, the Linux cluster required 144 days while the DeCypher system completed the task in 41 hours, 46 minutes.
According to TimeLogic founder and CTO Jim Lindelien, the DeCypher system works out to the equivalent throughput of 2,624 CPUs for the application.
Sia Zadeh, group manager for life sciences at Sun, said that while clusters offer many benefits for lower-cost, high-performance computing, the test demonstrated some of the limitations of the approach. Sun markets its Grid Engine server farm technology to the life sciences market, he said, but the approach is not always the most cost-effective choice. “Server farm operators face daunting administrative workload, scaling and load balancing challenges, and per-CPU licensing costs as farms scale to huge sizes,” he said.
Spotfire Reports Record-Breaking First Quarter
Spotfire of Somerville, Mass., said its revenue for the first quarter of 2002 grew 107 percent over the year-ago period, and credited the growth to 23 new customers for its DecisionSite product and orders from 73 existing customers. The news provided one of the few bright spots in an otherwise dismal earnings season for the bioinformatics sector.
“At a time when most software companies are struggling to stay alive through recession aftershocks, Spotfire has managed to grow in nearly every capacity,” said Spotfire board member Philippe Chambon, general partner of The Sprout Group.
The company also expanded its staff during the quarter, adding new employees in sales, consulting, customer service, and support in the US and in Europe.
3rd Mill Evaluates Integration Tools
A recent white paper from bioinformatics consulting firm 3rd Millennium confirms the ugly truth about data integration in the life sciences: there are so many options available that companies often make the wrong choice and end up with an ineffective solution.
“Companies are often undermined by how much more difficult it is to match their objectives with the right data integration solution than for other software investments. And if a company selects incorrectly, it might end up with a solution that only partially addresses its requirements,” said Roland Carel, senior systems architect for the Cambridge, Mass.-based company.
The paper, available at the company’s website (www.3rdmill.com) assesses the pros and cons of the four primary approaches: database federation, data warehouses, specialized databases, and point solutions. The paper also reviews integration products and technologies such as SRS, DiscoveryLink, the K1 integration middleware, GenoMax, and XML.
AnVil Wins Two Software Development Grants
AnVil of Burlington, Mass., was awarded two Small Business Innovation Research grants from the National Institutes of Health last week to develop drug discovery software tools for cancer research.
The first grant, “Cluster Comparison Methods and the NCI Expression Dataset,” provides $98,438 to cover development of new techniques for integrating quantitative cluster comparison metrics with the company’s high-dimensional visualization technology. AnVil will apply these new tools to the analysis of the National Cancer Institute Drug Information System (DIS) compound data set, which comprises data on chemical compounds applied to 60 cancer cell lines, combined with an external source of gene expression data for the same 60 cell lines.
The second grant, “Very High Dimensional Visual Data Mining of the NCI Dataset,” awards $99,225 to support development of new high-dimensional visualization technology to analyze and mine cancer data, which the company also plans to apply to the NCI DIS compound dataset.
The awards, which encourage commercial development of new technologies, will help expand the company’s offering and “provide analytic evaluation of the results of different clustering methods applied to the same data set,” said John Hotchkiss, CTO at AnVil.
Genomatix Software Contributes to US Bioterror Research
Genomatix, a bioinformatics company based in Munich, Germany, said last week that its software is being used in an undisclosed “leading United States national research laboratory” as part of its research into biological warfare agents and microbial pathogens.
The company’s software, which is able to detect specific DNA signatures to uniquely identify specific microbial pathogens, recently identified the DNA signature of the foot and mouth disease virus, according to Genomatix. This signature detects all seven serotypes of the virus, the company said, which is an improvement over previous tests that could only detect six types.
The US researchers are also using the software to study signatures for Variola (smallpox), Vibrio cholerae (cholera), and West Nile Virus.