You’ve probably never heard of Synamatix, but that’s no surprise to the people who work for the small database company based in Kuala Lumpur, Malaysia. As Synamatix emerges from stealth mode, it seems to have two main messages: one, that the first product it’s officially launching this August will truly speed up whole-genome comparisons; and two — and perhaps more importantly — that it’s not like the other companies you associate with bioinformatics.
With a database that Synamatix refers to as “second generation,” the company itself seems to be looking for recognition as a sort of second generation to the predecessors that failed when the genomics bubble burst. Johan Poole-Johnson, Synamatix’s global marketing manager, says the company started up in 2001 with founder Robert Hercus and has since remained privately funded with some infusion of VC money. The focus is on scoring partnerships to gain credibility: “We are in dealings with some of the genome research institutes,” Poole-Johnson says, and the company has an alliance with Hewlett-Packard as well.
The main application, slated to launch this summer, is a structured hierarchical database designed to store many genomes in one repository — Poole-Johnson notes that the goal is to add non-sequence data as well, but that hasn’t been achieved just yet.
Zayed Albertyn, senior bioinformaticist at Synamatix, says the database is unique because it “stores information as patterns, based on things like Bayesian statistics.” The database is relational and every bit of information is linked, he adds, so that when a user adds new data, it “will change everything in the database.”
The rationale for storing patterns and predictions rather than fixed data, quite bluntly, is for a boost in speed. A project to align and compare sequences between mouse chromosome 11 and human chromosome 17 took “something like three minutes,” says Poole-Johnson, “and we were able to find … homology and even synteny.” Another test run took a comparison that had been run in five hours elsewhere and took just over 500 milliseconds on SynaBase, he adds.
Albertyn notes that the speed factor means this database will be targeted at organizations looking into personalized medicine with applications like genotyping whole genomes, such as medical schools or other “institutions that might want to offer high-performance computing solutions.”
That said, Poole-Johnson notes that Synamatix isn’t picky about customers. The company will be trying to get its database out to users in pharma and biotech as well.
— Meredith W. Salisbury