Sequencing your genome just got cheaper

Jay Flatley, CEO of sequencing giant Illumina, announced at the Consumer Genetics Conference today that the company had reduced the price of its retail whole-genome sequencing service. At $19,500 this still isn't in the realm of an impulse buy for most of us, but it's a long way down from the $48,000 that Illumina offered at the launch of its service, and more than an order of magnitude below the $350,000 price paid for the first ever retail genome

Illumina's press release notes that bulk orders of 5 or more genomes drop the price to $14,500 per genome, and "[i]ndividuals with serious medical conditions for whom whole-genome sequencing could provide potential direct clinical value will be eligible for special pricing of $9,500 per genome". That latter price is very close to the reagent cost for a whole genome on Illumina's new HiSeq platform - so this is largely a market-creating rather than profit-making exercise.
Illumina's whole-genome sequencing service is not available directly to consumers, but only via a "physician-mediated process" that involves "pre-service consultation, consent, and a seven-day cooling off period, with final genome data returned to the physician". That's disappointing for those of us who feel we should be able to gain access to our own genetic information without a doctor's permission, but this approach - and the use of secondary services for all interpretation - will allow the company to dodge the world of pain currently being directed towards direct-to-consumer genetic testing companies
The most interesting aspect of this announcement is that Illumina's new price point puts it in direct commercial competition with sequencing service provider Complete Genomics, which has been offering whole genomes at around the $20,000 price point for some time. Complete currently appears to have substantially lower reagent costs per genome, so it seems likely to be able to retain the lead in the inevitable price war over the short-term - but Illumina's deep pockets and notoriously nimble technology development capabilities make it an intimidating competitor.
Add the plethora of emerging third-generation sequencing technologies to the mix, and the stage is well and truly set for a race to the bottom dollar. And regardless of who emerges victorious, those of us interested in affordable genome sequencing get to reap the rewards.
Update: Matthew Herper has a post on Illumina's announcement at Forbes.

More like this

Under HIPAA, patients have the right to a copy of their medical records, so people should still have access to their Illumina-sequenced genomes.
One thing that interests me about this method - it potentially escapes patent licenses on genetic tests. As long as Illumina just delivers raw data, they haven't broken any patents. At this point, it's just a database match against known disease associated alleles, and free software is available to the physican/patient for that. To make a formal diagnosis and proceed to treatment, they would probably still have to confirm by ordering the specific test from a licensed lab to keep the health and malpractice insurance companies happy though.

Too bad it only applies to human genomes. We're still having all sorts of problems sequencing some species (assembly just fails), much less a having cheap-ish sequencing for doing population genetics.


Too bad it only applies to human genomes. We're still having all sorts of problems sequencing some species (assembly just fails)...

If this is the case I'm not so sure Illumina is going to help you tremendously. Without a reliable reference, even the comparatively longer Illumina reads are still rather short and will have trouble mapping de novo.

I'm afraid you're going to have to go 454 or good old shotgun/BAC end-pair Sanger sequencing (or both) to get what you need.

By Jason Schultz (not verified) on 04 Jun 2010 #permalink

travc -- assembly "just fails" meaning what?

No contigs of any size?

Or contigs that don't overlap enough?

Is your species so far away on the tree of sequenced genomes that you don't have even a few trial reference genomes?