By Kevin Davies
February 7, 2011 | MARCO ISLAND, FL – It is a testament to the remarkable progress in next-generation sequencing and analysis that when neurobiologist Tim Yu described the complete sequencing of 40 human genomes in a successful search for gene mutations that cause autism, it barely registered a ripple from the large audience.
That presentation was one of many standout presentations at this year’s Advances in Genome Biology and Technology* conference, which concluded over the weekend. While the conference lacked some of the technology razzamatazz of recent years, there were clear signs of the early maturation of newer technologies from companies such as Pacific Biosciences, Ion Torrent, and Complete Genomics (CGI), which supplied Yu with those 40 genome sequences.
Complete Genomics CEO, president and chairman Clifford Reid spoke exclusively to Bio-IT World about CGI’s recent accomplishments and near-term plans. Having taken the company public late last year, Reid was celebrating the company’s landmark of 1,000 sequenced human genomes and public release of 60 completed genomes that have been well studied, including a 17-member CEPH family. (The data are available from CGI’s website and mirrored on the Bionimbus Cloud.) One repeat customer, the Institute of Systems Biology in Seattle, recently placed an order for 615 genomes, following the receipt of 104 genomes last year.
CGI is now sequencing up to 400 genomes per month, all to at least 40X coverage, and Reid expects to be “in the vicinity of 1,000 genomes per month” by the end of the year. CGI’s customer base remains firmly academic, including engagements with a handful of physician scientists. As for big pharma, Reid says interest is starting to grow. “Pharma tends to be more of a follower, with a few notable exceptions. We’re working with the exceptions,” he said. Those include deals with “three flagship pharma accounts” -- Genentech, Pfizer and Lilly – and he is expecting more.
Cost and Quality
“We’re still the only company that’s published a 10-5 error-rate [human] genome,” Reid says (average 1 error/100,000 bases). He asserts that Illumina’s current system consumes $5,000 in reagents, and that cost swells to $20-25,000 when the full cost of informatics and labor is included.
After claiming last year that CGI had cracked the $1,000 genome threshold for reagent costs, Reid now says that CGI’s all-in cost for a complete human genome is under $10,000. “With all of it added in, we’re below $10,000 now. We’ve got a 2-3X cost advantage [over Illumina], and a 10X quality advantage.”
CGI currently charges $9,500 per genome for a minimum order of eight genomes. “You can’t pay $20,000 [per genome] any more, even if you try. We just send the money back!”
While more and more service providers are cropping up, the biggest competition to CGI is BGI – the former Beijing Genomics Institute. Or so it seems. “We don’t entirely know yet,” says Reid. “They have a big facility, but they’ve also said they’re all things to all people. In the complete human genome space, they’re using Illumina technology, and we have distinct advantages for complete human genomes over that technology. So it’s more likely people would use BGI for one set of projects, and CGI for others.”
One of CGI’s original goals was to build a network of genome centers around the world, in collaboration with various foreign governments or organizations. Reid signaled that CGI might take some steps in that direction. “From Mountain View, we can service North America and Western Europe quite effectively, though we might open a center in Europe to get closer to customers. But I think opening a center in Asia will get us access to customers that we really don’t have access to.”
Noting that CGI has just one 1 sales representative in Asia (Japan), Reid said: “We haven’t pulled the trigger yet [on expansion], but we certainly expect to use that model to expand into Asia.”
Reid also indicated that CGI would consider expanding its suite of services, which until now has focused exclusively on complete human genomes. “We’re not going to do exomes,” he said. “We think that’s an interim product.” But while there is no stated plan when it comes to transcriptomes, Reid said, “that’s squarely within our strategic footprint. We can and should be doing transcriptomes.”
The CGI team continues to drive advances in its proprietary DNA nanoball technology, increasing the density of the DNA on the slides. “Right now, the commercial systems are running slides with 3 billion spots at 700 nm center-to-center separation,” said Reid. “So the spots are separated by the wavelength of red light. We’re on our way to the wavelength of blue light!”
Reid also said that plans are advanced to commercialize CGI’s long-fragment read (LFR) technology, for which a patent has been issued. The LFR technology will be “an overlay on our current sequencing technology,” said Reid, with a DNA engineering front-end and an informatics back-end. A paper describing the approach and applying it to a genome sample is about to be submitted.
Among the advantages to the LFR approach are chromosome separation, which phases out the two parental sets of chromosomes. “It also has a wonderful property of error detection and correction,” says Reid. “There’s a whole category of errors it can detect – the number of errors it is catching is very high. Right now our genomes are 10-5. These LFR genomes will be 10-7 -- another factor of 100.
“From a clinical perspective, a 10-5 genome is 30,000 errors in the genome, which gives you pause. But a 10-7 genome is 300 randomly distributed errors in the genome. That’s virtually perfect! We think we’ll be doing clinical grade genomes by the end of the year.” CGI is currently in the early stages of preparing for CLIA certification.
If Reid’s claims are verified, then CGI’s ability to sequence genomes to high accuracy and phasing and at less than $10,000 per sample would appear to put it in line for a shot at the Archon Genomics X PRIZE. But CGI’s Achilles’ heel is turnaround time. The X PRIZE calls for a total turnaround time of ten days. “Our turnaround times are longer than that,” says Reid. “But in terms of price, phasing, quality, we’re well within the X Prize spec.”
At AGBT, Tim Yu provided further validation of CGI’s genome sequencing service and the ability to ferret out single-gene mutations that cause or predispose to disease from the sea of benign or irrelevant base changes.
Yu and colleagues have focused on families from the Middle East, where a high degree of cousin-cousin marriages (consanguinity) contributes to genetic diseases. Yu’s team collected some 200 families with cases of autism, analyzing 40 sequenced genomes from CGI. On average, an individual genome contained some 10,000 potentially pathogenic variants when cross-referenced to the databases. Those are then filtered further to focus on novel variants that follow a consistent inheritance pattern in the affected families.
For example, in one Pakistani family with three affected children, the likely candidate gene on chromosome 6 proved to be PEX7, a known metabolic disease gene. Yu suspects that a mild mutation in this gene is responsible for the autism-like symptoms. In a Saudi family, Yu suspects a mild mutation in another metabolic disease gene called AMT. Other gene candidates play roles in histone modification and axon guidance.
“There will be no ‘autism gene’ but many autism genes,” said Yu.
*Advances in Genome Biology and Technology 2011, Marco Island, FL; Feb. 3-6, 2011.