NEW YORK, Nov 14 - Several months after announcing the establishment of a $100 million war chest to fight for a foothold in the genomics sector, IBM has embarked on its latest effort to win new business.
Big Blue is now conducting extensive market surveys and customer focus groups to try to figure out how to muscle their way into the already crowded bioinformatics and genomics market.
“We’re doing a lot of them,” said Anne-Marie Derouault, director of business development for IBM life sciences. “Both focus groups which are qualitative and also quantitative surveys, like doing interviews over the phone.” The company has also hired outside consultants to help.
But with Sun Microsystems’ hardware and Oracle’s data management software quickly becoming the standard for the broader market and Compaq’s Alpha servers already crunching data for some of the biggest high-performance computing consumers in genomics, IBM has its work cut out for itself.
“They were asking what it takes to move people off their current platforms and off their current software apps,” said TimeLogic president Jim Lindelien, one of several genomics executives Big Blue invited to a focus group conducted by Smith Research at Scientific American ’s recent BioSilico conference in New York. “And everybody said it takes a hell of a lot.”
While IBM has had little success so far in winning over competitors’ customers, the company has had moderate success in winning contracts with newly formed companies, such as First Genetic Trust.
IBM has also begun to challenge Oracle with it own database management solution called DiscoveryLink, which creates what IBM calls a “virtual database,” allowing for a single complex query to search various databases regardless of format.
In September Incyte agreed to embed DiscoveryLink into its new Genomics Knowledge Platform, which is designed to allow researchers to integrate and analyze a variety of data types. And it’s on the data management front that IBM may have the best shot.
“There was almost universal agreement that dealing with Oracle is a necessary evil rather than a pleasure,” said Lindelien. “Everybody is concerned about the scalability of their solution for life science.”
While IBM might be late to the genomics game, the company is hoping that it is well-positioned to gain a foothold in the proteomics sector.
“Going from the genomic stage into the proteomics stage, the complexity both in terms of computing performance and size of database and data management is so mind boggling that people are a little concerned about whatever system they have today and how they are going to scale it up to the level it has to be,” Derouault said.
The company plans to take on Compaq with Blue Gene, a petaflop-scale computer 1,000 times more powerful than the Deep Blue machine that beat world chess champion Gary Kasparov in 1997. Blue Gene is expected to begin modeling human protein folding by 2005.
On the hardware front, IBM is faced with the challenge of finding a way to eclipse Sun’s grip on the market. Sun gets potential customers used to its system by providing steep discounts to universities.
“They get people hooked on their machines when they are in grad school or even in undergrad, so that when people get out and they are in companies they want to continue on Suns because that’s what they are used to,” said Incogen president and CEO Maciek Sasinowski, another focus group participant. “So, when I was at that point when I was actually the decision maker, I got a Sun.”
Even though many industry insiders said an increase in computer companies catering to the genomics sector would result in greater innovation and cheaper products, some industry players said they were not prepared to give up their current systems. Even if a computer maker gives away the technology, switching platforms often requires too much effort and money for the company to benefit much.
For companies to overhaul their current IT platforms and give IBM a try “there has to be a big enough carrot dangled in front of the decision makers,” Lindelien said. “That carrot is what I think IBM is trying to define."