Skip to main content
Premium Trial:

Request an Annual Quote

For 454 Technology, Roche Bets on Specific Apps, Taking Share from Sanger


By Julia Karow

This story was originally published on August 5.

Roche plans to take "significant share" from the Sanger sequencing market as well as focus on specific applications, including medical ones, for its 454 technology, according to 454 Life Sciences' CEO.

In addition, with its partner IBM, the company is developing a successor technology that it estimates will take five to seven years to bring to market.

Sequencing was a billion-dollar market last year, 454 Life Sciences President and CEO Chris McLeod told investors during a Roche event at the American Association for Clinical Chemistry annual meeting last month, and is poised to grow to $1.6 billion over the next five years.

The growth will be driven by next-generation sequencing, driven by an increase in large-scale resequencing of human genomes and because it will start taking market share from Sanger sequencing, McLeod said.

The next-gen sequencing market is split into different applications, he said, and while Roche expects human genome resequencing to be the largest one, it believes the 454 technology — due to its long reads and high accuracy — best addresses de novo genome sequencing, metagenomics, and targeted sequencing.

But the company also aims to take a slice from the existing Sanger market, both by extending its read length to Sanger-like 800 bases and by selling its lower-cost, lower-throughput instrument — the GS Junior — which "puts next-generation sequencing on the benchtop of the individual researcher."

In addition, the company will be "trying to leverage the strength of Roche Diagnostics" to apply its technology in the diagnostic field.

McLeod said that early-access customers have already been testing a long-read package for the GS FLX that takes them "into the neighborhood" of 600 bases per read. In a second testing phase, they will reach an average of 800 bases per read, and up to 1,000. Roche plans to launch the long-read package globally in early 2011, he said.

So far, 454 has seen "good acceptance" for the GS Junior, a desktop sequencer it launched in May (IS 12/1/2009). "The initial response is above our forecast, and we are working hard to make sure that we meet demand and that our customers have a great experience," McLeod said. The instrument costs about a fifth of the GS FLX — or on the order of $100,000 — and produces about 35 megabases of data with 400-base pair reads in a 10-hour run.

In order to facilitate certain targeted sequencing applications, 454 is also developing a number of primer sets — the first one for high-resolution HLA genotyping, which is currently mostly performed by Sanger sequencing. While that method is "very expensive and very slow," because it requires multiple steps to achieve unambiguous allele calls, "454 sequencing, in a single run of the Junior, will be able to deliver tens of different high-resolution HLA genotypes, with very unambiguous allele calls," he said.

Roche sees two main areas where next-gen sequencing could have diagnostic applications, according to McLeod: One is in virology, for example to detect drug-resistant strains of HIV present at low levels, or to identify emerging novel pathogens. The other is in immunology, where sequencing can be used to profile immune receptors, for example to characterize lymphomas, as well as for HLA genotyping.

Overall, McLeod said, the 454 technology is currently "mid-way up its S-curve" of growth, implying that it will soon plateau, like the Sanger technology. That is why Roche, with its recent partnership with IBM, is already looking for a real-time single-molecule sequencing technology to succeed it (IS 7/6/2010).

The nanopore platform, which will use IBM's DNA transistor technology to control the translocation speed of the DNA, promises very long read length, fast runs, and electrical detection, "which would mean that the instrument itself would not be that expensive," he said. It will also require no amplification, have an easy sample prep, and inexpensive reagents and disposables.

The technical development of the platform will take approximately three years, Roche estimates, followed by another two years or more to turn it into a commercial product, so "we are looking at a five- to seven-year window before we will be able to bring this technology to market," McLeod said.

The nanopore technology, he said, has the potential to sequence a human genome for less than $1,000, "and potentially on the order of $100 per genome."