Complete Genomics is the latest company to herald the arrival of cheap genome sequencing. Credit: Complete Genomics

The era of the $1,000 genome has arrived. That's the claim from Complete Genomics, a young company based in Mountain View, California, that revealed today that it plans to sequence 1,000 human genomes next year — and 20,000 in 2010.

But don't get your hopes up: Complete Genomics won't be selling to just anyone. It also hasn't released any data to back up its claims, although it has the backing of prominent scientists in the sequencing field, including Harvard University's George Church, who is sequencing the genomes of ten individuals in his Personal Genome Project and acts as an adviser to several sequencing companies including Complete Genomics.

"I think it's quite probable" that the company will achieve a $1,000 genome by the middle of next year, Church says, but "it's not going to be easy" to sequence that many genomes in that amount of time.

The company enters a market crowded with companies selling such 'next generation' sequencers, which use faster, cheaper technologies than those used in the Human Genome Project. But unlike the current crop of companies, Complete Genomics will not sell sequencing machines.

Instead, the firm is building a sequencing centre at its Mountain View base, where it plans to sequence human genomes for pharmaceutical companies, biotechnology firms, personal-genomics companies and government-funded sequencing centres. It will charge customers $5,000 to sequence a genome, says chief executive officer Cliff Reid, who estimates that the cost of materials used to generate each sequence will be around $1,000.

Exploding technology

DNA-sequencing technology has exploded since 2003, when the Human Genome Project produced a composite human genome sequence containing DNA from many people. In June 2007, James Watson became the first person to have his own complete genome sequenced at a cost of $1-1.5 million Since then a handful of other people have had their genomes sequenced, and costs have dropped to tens of thousands of dollars.

So far, cheaper genome scans have examined a subset of a person's DNA, looking at points of the genome called single nucleotide polymorphisms (SNPs) – the markers targeted by 'personal genomics companies'

But SNPs only mark areas of the genome that are associated with a disease; they do not reveal what precise genetic glitch causes the disease. Comparing whole genomes, however, might reveal the genes underlying an illness. Such studies will gain power as more genomes are sequenced.

Complete Genomics is proposing to sequence thousands of genomes at a speed and cost that no other company currently taking orders from customers is approaching. By keeping its machines in-house, it will not have to spend money making them 'idiot-proof' for outside customers. And the company is also using some cost-saving technological tricks.

Faster, cheaper

Its process breaks up genomic DNA into single-strand fragments which, along with segments of synthetic DNA called adapters, form small circles of DNA that contain about 80 bases. These adapters help to break the genomic DNA into smaller, easy-to-sequence chunks.

Hundreds of the DNA circles join together into a long chain that folds up into a 'nano-ball'. These nano-balls are then washed over a glass slide with activated sticky spots that bind the DNA. Once a nano-ball has stuck to a spot it repels other nano-balls, ensuring that the slide carries as many different sequences as possible. This high density means that a standard glass microscope slide holds about a billion spots. The miniaturization cuts down on the cost of sequencing chemicals, the firm says.

A ligase enzyme then helps to add short strands of complimentary DNA, which have been dyed with several fluorescent compounds, to the nano-balls. The fluorescence can be used to infer the sequence of the underlying DNA sample by monitoring the slide with cheap, off-the-shelf digital camera technology.

The whole system is very similar to methods developed by Church and used by the company Applied Biosystems in Foster City, California, in its SOLiD system. Church is using similar technology in his Polonator, a sequencing machine designed to use cheaper materials and open-source software.

In just seven days ...

So far, Complete Genomics has used its technology to sequence a human genome in seven days. By next year, it says that it will be producing finished genome sequences that include data from both the maternal and paternal contributions to an individual's DNA, with DNA from 99% of the protein-coding regions and 90% of the total human genome. But until the company releases more details, it is not yet clear how 'complete' its genomes are.

Elaine Mardis, co-director of the Genome Sequencing Center at Washington University, St Louis, is one of a handful of customers who have been briefed by the company. Mardis, who describes herself as "a sceptic by nature", says: "Before I would even know they were a 'fit' for us, I would need to see data produced by their method."

But Chad Nusbaum, co-director of the sequencing centre at the Broad Institute in Cambridge, Massachusetts, was more optimistic. "If it works, it's exciting, and I think there's a decent chance it will work," he says. If it does, "it puts us in a position to be able to sequence thousands of genomes".

And John Chant, a biotechnology entrepreneur based in San Francisco, is so bullish on the company that he says he convinced his former employer, San Francisco-based Genentech, to invest $1 million in it. "The selling point is, if you keep DNA from everyone in a clinical trial, when patients have an adverse reaction you can find out the gene that causes it," he explains. This would potentially allow companies to screen out the patients who have that gene from future trials. At $5,000 a genome, and perhaps 10,000 patients per clinical trial, the $50-million sequencing cost might be worth it to save a multi-billion-dollar drug, Chant says.

In the meantime, Reid says that the company has already signed up its first customer: Leroy Hood, president of the Institute for Systems Biology in Seattle, Washington. Hood, an adviser to Complete Genomics, has asked the company to sequence the genomes of more than 2,000 people as part of a project he is conducting with the government of Luxembourg.

Reid adds that the company will not compete for the X Prize in Genomics, which will award $10 million to the first team to sequence 100 human genomes in 10 days. He says that the project would consume too much capacity of the 96 machines Complete Genomics intends to build next year.