Now we can cheaply sequence DNA, how do we store all that data?

If we are to continue learning about the human genome, making huge amounts of data scalable and available to physicians and patients will be key

In 2003, the international consortium of scientists working on the Human Genome Project completed the final first draft for the human genome - a DNA blueprint for human life. This monumental achievement involved thousands of dedicated people, took more than a decade, and cost over $2.5 billion (£1.95bn). The public availability of a completed human DNA map ushered in the genomics era, giving rise to personalised, or precision, medicine.

Now, 15 years later, major advances in DNA- sequencing technology and its commercial development have driven down the time and cost of sequencing a human genome. In 2018, we will at last start to understand the commercial, clinical, regulatory, ethical and legal issues unlocked by the Human Genome Project.

Today, a fully analysed whole-genome-sequencing test costs about $600 and takes just a few weeks to complete. This next-generation technology is changing the way doctors manage patients with rare inherited diseases and cancer. In 2018, DNA sequencing will continue to improve, with new applications such as screening for disease in newborns.

But, as is common with other disruptive technologies, our advances in genomics have outpaced the ethical and legal frameworks we need to responsibly and sustainably deploy them. In 2018 we will need to find ways to provide better assurances of patient safety and practical protections against losses of privacy. Patients and physicians alike must also be presented with a realistic approach to expectations, so that public trust in this transformative technology grows. It will require co-operation on a scale that does not yet exist.

Dame Sally Davies, the chief medical officer of England, has recently called for making genomic sequencing as common as other routine blood tests in patients with cancer and rare inherited diseases. This may seem quite reasonable, and is now technically plausible. But, aside from the cost of professional analysis and interpretation, there's the issue of storing all this information.

Current low-cost cloud-based storage is available for £0.004 per gigabyte per month. Sounds cheap, right? But sequencing a single tumour creates about two terabytes of data, so a lab doing just 1,000 cases a year would have to spend around £96,000 annually. Further, readily accessible secure cloud storage that meets regulatory requirements costs much more, and these data storage costs are cumulative - thus, our hypothetical sequencing lab might be spending more than £1 million annually after five years.

Healthcare workers will need this information widely available if precision medicine is to work. The data must be scalable, harmonised, and above all available not only to physicians and patients, but to scientists and clinical investigators if we are to continue to learn about the human genome. In 2018, we will work out how we can handle it, store it, share it and make the best use of it. By cracking that, we will open the way for a raft of other big-data applications that will make the world a better place.

This article was originally published by WIRED UK