JMP, a business unit of statistical software provider SAS, has released a trio of bioinformatics tools that promise to bring both statistical rigor and user-friendly graphics to the biologist's desktop.
This week, JMP released JMP Genetics, JMP Microarray, and JMP Proteomics — desktop applications for the statistical analysis of DNA, RNA, and protein data, respectively.
The products are essentially a reworking of the SAS Scientific Discovery Solutions suite — an enterprise-scale system that included a centralized data repository called SAS Research Data Management, along with SAS Microarray, SAS Genetic Marker, and SAS Proteomics.
"This was a server-based solution, it had a Java front end, and then ultimately in order to view your results, you actually launched JMP," said Charles Ellis, global manager for genomics sales at JMP. However, he noted, SAS Scientific Discovery Solutions "carried a large price tag and basically wasn't accessible for a biologist doing microarray analysis or proteomics analysis."
In response, SAS and JMP decided to co-develop a new line of products — the first such collaborative effort for the business units — in order to make the statistical capabilities of the SAS platform more accessible to bench biologists.
SAS Scientific Discovery Solutions "carried a large price tag and basically wasn't accessible for a biologist doing microarray analysis or proteomics analysis."
The three JMP bioinformatics tools use the same stored processes that underlie the SAS suite of genomics software, but run them in a lower-cost and user-friendly desktop environment. Pricing ranges from $1,000 to $10,000 per seat, Ellis said, depending on the number of licenses. The JMP software can also be installed on a server, as well, he said.
SAS Scientific Discovery Solutions, meanwhile, has been rolled into a broader offering for pharmaceutical companies called SAS Drug Development.
The decision to reconfigure the bioinformatics lineup came from the realization that there is still a large communication gap between statisticians and biologists when it comes to data analysis, Ellis said.
"A lot of biologists get their microarray results and then pass them over the wall to the statistician, which creates a lot of work for these people," he said.
JMP addresses this issue with a feature called the "journal," which enables statisticians to build analytical workflows that they can share with their colleagues running the experiments. From a statistician's standpoint, "it's a very easy way to standardize data analysis," Ellis said. As for the biologists, he said, the software is a welcome option for "people who don't want to know what's behind the statistical analysis."
The software suite also includes JMP's Design of Experiment tools for creating "unconfounded" experiments. In addition, the software is customizable for users familiar with either SAS or JMP. A SAS user comfortable with SAS statements can add his or her own algorithms to any of the software tools, while those comfortable with the JMP Scripting Language can customize the front end to include links to internal or external databases or other features.
JMP's claim that it can offer both high-end statistics and user-friendly graphics in a single desktop system places it in direct competition with vendors from both directions, however. On the statistical side, the company is going head-to-head with Partek, while customers looking for ease of use are more likely to turn to a familiar package like Agilent's GeneSpring.
Ellis acknowledged that the microarray analysis market is crowded, but said that it's not quite saturated yet. He estimated that JMP's direct competitors in the microarray analysis market currently generate around $30 million a year in revenues. "Is the market going to grow from there? I don't know. But what I can tell you is that there still is a desire for stronger statistical tools in this area," he said.
In addition, he said, new higher-density microarrays coming online in new application areas should help grow the user base for software that can handle large data sets without compromising statistical analysis.
Around 15 customers have been beta-testing JMP's bioinformatics tools since October, and Ellis said the initial response as been "very positive." While the company is initially targeting current customers for SAS and JMP, Ellis said that it looks like one of the first sales will come from an undisclosed customer who never used either type of software before.
— Bernadette Toner ([email protected])