John Timmer gives a great summary of the new paper in Science covering the Complete Genomics sequencing method.
A surprising (to me, at least) fact:
But, because each of these nanoballs was so small, it's possible to do massive amounts of sequencing in parallel, simply by washing different solutions across the surface they were stuck to. The authors were able to get anywhere from 45- to 87-fold coverage of the genomemeaning that, on average, each base in the genome had been sequenced 45 or 87 times, respectively. That's just an average, of course, as chance will ensure that some sequences are under- or overrepresented. Still, given an algorithm designed to work with these specific results and the reference human genome sequence, Complete Genomics was able to get excellent coverage of the genome.
Oh, yeah – the lede from the paper:
Validation of one genome data set demonstrates a sequence accuracy of about 1 false variant per 100 kilobases. The high-accuracy, affordable cost of $4,400 for sequencing consumables and scalability of this platform enable complete human genome sequencing for the detection of rare variants in large-scale genetic studies.