Illumina, the largest maker of DNA sequencers, is launching a new DNA sequencer with new architecture it says could push the cost of decoding a human genome from $1,000 to $100–although that decrease will not come for years.

“This is good news, affirming that the field is still so healthy that price-plummeting is still considered good for business,” says George Church, the Robert Winthrop Professor of Genetics at Harvard Medical School.

The new DNA sequencers, called the NovaSeq 5000 and the NovaSeq 6000, could be about 70% faster than existing machines, by Church’s math–not a big increase in this field. But Illumina promises rapid increases as new parts and software upgrade the machines, meaning that by the end of the year the machines will be six times faster than its predecessors. Illumina has annual sales of $2.4 billion.

In 2006, Illumina’s first DNA sequencer could sequence a human genome at a cost of $300,000. For Illumina’s highest-end products, the cost is currently $1,000 per genome, including reagents and the amortization of its machines. The machines themselves are still costly: The NovaSeq 5000 and 6000 cost $850,000 and $985,000, respectively.

But those drops in cost have generated a huge amount of medical knowledge. Five years, ago, only a small number of human genomes had been sequenced. Two years ago, when Illumina made its last big product announcement, 65,000 human genomes had been sequenced. Now more than 500,000 have been sequenced, Illumina says.

“We feel like in the high end of the market this continues to put distance between us and any possible player in the market on any dimension: Quality, throughput or cost per genome,” says Francis deSouza, Illumina’s chief executive. “This is a phenomenal machine.”

Recommended by Forbes

Illumina CEO Francis deSouza (Image courtesy Illumina)

However, deSouza said the $100 number was more a roadmap–something that would probably happen in more than three years and fewer than ten.

Prominent genomics researchers were happier about more prosaic features. First, new machines are individual machines–Illumina’s current top-line products, the X5 and X10, are sold in bundles and can’t be bought one at a time. Second, they don’t come with restrictions on what kind of research they can be used for. Illumina does not allow the X5 and X10 to be used for looking at what are called exomes: looking at only the parts of the genome that contain known genes. Those restrictions will still exist for the older machines, Illumina says.

Continued from page 2

“It’s good they are moving away from the model of requiring users to buy a certain number of instruments as a bundle, and not allowing you to do some types of assays on an instrument you bought and paid for (dearly), which was just silly,” wrote Elaine Mardis, co-director of the Institute for Genomic Medicine at Nationwide Children’s Hospital in Ohio, in an email.

DeSouza says that the new machines are also easier to use, with the number of steps in its workflow cut from 38 to 8. The machine has been made more idiot-proof, with RFID chips checking to make sure components are loaded properly. Creating the new architecture, he says, required 17 major innovations across every part of the sequencer, although it uses the same basic chemistry as previous models.

Another question is read length. DNA sequencers actually only read small bits of DNA, which are assembled like a puzzle. The Illumina machine is still stuck at 150 kilobases (thousands of DNA letters). But some new technologies are allowing much longer readlengths–at a much higher cost per DNA base pair.

One is Oxford Nanpore, which already sells a portable DNA sequencer and is expected to launch one for laboratories this year. The question is whether Oxford can even make a dent in the Illumina marketing machine.

“I think there’s going to be blood in the water at some point in 2017 or 2018,” says Ewan Birney, joint director of the European Bioinformatics Institute. He notes that Nanopore, which he consults for, has some advantage. An experiment on its machine can be run in a matter of minutes or hours (one run on an the NovaSeq takes 40 hours) and that the Nanopore machine will only cost $75,000, which could help labs start to try it out.

Many other DNA sequencing upstarts–Ion Torrent, Pacific BioSciences, Complete Genomics–have tried and failed. None has even made a dent. At present, the company holds the vast majority of the market for DNA sequencing worldwide. But in order to make its investors happy, it’s also going to have to make that market grow beyond research. That’s as big a challenge as any competitor.

Page 2 / 2

Comment on this story

This article was sourced from