It cost $2.7 billion to sequence the first human genome, which was completed in 2003. But even then scientists talked about it someday being possible to do it for $1,000, since that could make it practical to read out all your DNA for some routine medical care, the way MRI or CT scans are used now.
In January 2014, Illumina said it had essentially hit that milestone with its HiSeq Ten sequencing system, which it said could spell out a genome for about $1,000. But that number is a bit misleading; $1,000 might be the rough average cost of reading out a single genome in a facility that can achieve the economies of scale that come from constantly running 10 sequencing machines in parallel. Each machine costs $1 million, so not many sequencing centers can afford 10 of them.
The Broad Institute of Harvard and MIT was one of the first buyers of the 10 machines that make up the Illumina HiSeq Ten system; the Broad now has 16 and churns out almost 2,000 human genomes a month for researchers. Even at that scale, the cost of sequencing is “not quite $1,000 yet,” says Stacey Gabriel, senior director of the genomics platform at the Broad, but “we’re close.”
Keep in mind that the cost to a sequencing facility is generally lower than the price it charges its customers. The Broad charges researchers $1,800 to sequence a human genome. Other big centers have similar rates. For example, the private company Macrogen, headquartered in South Korea, charges $1,600. And the one-off rate that these centers (or Illumina itself) would charge you or me for a genome sequence is higher still.
Gabriel says the cost of sequencing includes the chemicals and other materials needed to read out DNA, labor, basic data analysis (stringing the whole sequence together from end to end and putting it in a readable format), and some slice of administration and overhead (building electricity, for example). It does not include the cost of data storage.
Even with those expenses, continued refinements to the technology should bring the cost of sequencing to $1,000 “within less than a year,” says Jeffery Schloss, director of Genome Sciences at the National Human Genome Research Institute. For example, faster imaging systems and improvements to the sequencing chemistry and the thumb-drive-sized chips where the sequencing occurs have sharply reduced the time to process each genome, says Joel Fellis, associate director of product marketing at Illumina.
Driving costs down further
For clients who pay $1,600 to $1,800 for sequencing, each letter of DNA is read an average of 30 times. The more a DNA sequence is read, the greater the certainty that the sequencing machine read the sequence accurately.
Sequencing a genome 30 times over is enough for some medical purposes, like understanding disorders with a small number of genes involved. But for research into genetically complex disorders, you’d need to sequence thousands of individuals, which means $1,000 per genome would still be prohibitively expensive. For research into certain cancers, which can have different mutations in different individual cells, researchers recommend 60 or 80 times. That obviously means higher prices, and as a consequence, some applications still need sequencing to become about 10 times cheaper: more like $100 or $200.
Illumina’s Fellis won’t predict when that could happen, but he says the cost reductions should be doable with continued refinements to existing sequencing systems. “I don’t think a fundamentally new technology is needed,” he says.
It costs large facilities not much more than $1,000 to sequence a genome if they are doing a big batch of them. Before long the obstacle to widespread genome sequencing will no longer be cost, but proving that it’s medically useful. For many medical applications, sequencing the exome, a key subset of the genome, is sufficient, and already costs less than $1,000.
Thanks to Henrik Grabner for this question. If you have one, send it to email@example.com.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
Driving companywide efficiencies with AI
Advanced AI and ML capabilities revolutionize how administrative and operations tasks are done.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.