The battle for sequencing supremacy

http://www.nature.com/nbt/journal/v30/n11/full/nbt.2412.html

Today, the HiSeq 2000 instrument can generate up to 600 Gb of sequence data in about 10 days, sufficient to sequence four entire human genomes with each base sequenced an average of 30 times (30× depth of coverage). This throughput and quality also come with a high price tag (~$700,000 for the instrument and $6,000–8,000 per genome), but with over 1,000 instruments placed worldwide, the HiSeq 2000 is the backbone of most cutting-edge sequencing centers. 
 
It also faces stiff competition from Illumina, whose benchtop MiSeq is more expensive upfront and slower, but also generates more data—more accurately, according to some studies. Both Ion Torrent and 454 sequencing tend to encounter problems with homopolymers, misreading stretches of sequence where the same nucleotide is repeated many times and generating inappropriate insertions or deletions (indels). “Based on our diligence, homopolymers continue to be an issue for Life Tech, and I think the MiSeq has a considerable lead in data accuracy,” says William Quirk, managing director and senior research analyst at PiperJaffray. 
 
At present, though, most clinically minded companies see the greatest opportunity in developing assays for targeted resequencing of individual loci associated with disease—applications where small, high-speed benchtop instruments are ideal. “Life Technology recently announced that more than 50% of their PGM placements now are going into the clinic,” says Quirk. Some centers, like the Ontario Institute for Cancer Research, are already using tumor sequencing to guide patient treatment, and McPherson reports that instruments like the MiSeq have proven quite effective in enabling his team to act quickly and confidently. “We're trying to return results—from the time the patient is consented to the time the clinician gets a report—in three weeks,” he says. “That requires that the sequencing doesn't take more than a day or two.”
 
Illumina and Life Technologies have been quick to recognize this opportunity, and both MiSeq and PGM are undergoing US Food and Drug Administration (FDA) evaluation for general use as clinical instruments.
 
Targeted clinical diagnostics is also the primary focus of Cambridge, Massachusetts–based GnuBio, a startup that has largely stayed below the radar, but already has a working instrument installed at Canada's Montreal Heart Institute. In their system, a microfluidic workflow combines PCR amplicons from a genomic sample with nanodroplets containing one of ~5,000 different hexameric sequencing primers, each associated with a specific dye barcode.