Whose genome is it anyway? Big Data and your DNA

DNA is effectively data. If we are concerned about the data stored on our phones and computers, on social media sites and government servers, should we extend our concern to our DNA?

Over the past five years the 100,000 Genomes Project has carried out whole genome sequencing of consenting NHS patients with cancer or rare diseases, and their relatives. The amount of genomic data on file is set to increase massively, as in the next five years the UK government has declared that one million genomes – minimum, with an ambition of five million – will be sequenced (see BioNews 970).

Genomic medicine holds great potential for improving understanding, diagnosis, and treatment of disease, but every person's DNA is their own unique data. At the Progress Educational Trust event 'Whose Genome Is It Anyway? Big Data and Your DNA', held at the Royal Society of Edinburgh on 10 October 2018, an audience heard a panel of experts discuss the various issues which accompany this burgeoning field.

First to speak was Zosia Miedzybrodzka, professor of medical genetics at the University of Aberdeen, and clinical lead for the 100,000 Genomes Project at NHS Grampian. She reflected that as a clinician, she encounters plenty of educated patients who nonetheless do not know what DNA is, or what it does. After all, the professor challenged the audience, could you define what a genome is?

While genetic testing in the NHS has usually focused on coding DNA and a few particular known genes, the Scottish Genomes Project – like the 100,000 Genomes Project– will carry out whole genome sequencing. This will record the data of both coding and noncoding DNA, looking for variants of the 10,000 known genes implicated in rare diseases, and more. As research will be ongoing, the project will have long-term, follow-through electronic records.

Addressing discussion about consent and children's genetic data, Professor Miedzybrodzka added that in accordance with NHS guidelines, consent can only be given when a patient has capacity to give it, and upon reaching maturity every individual must consent for their data to remain in or by default be taken out. In further discussion of how information about genetic variants could be divulged to patients by doctors, she urged caution until the clinical significance of variants is clearly known.

But to find results, genetic data may not be enough. Dr Austin Diamond, a quality manager at NHS Lothian's Molecular Genetics Service, gave insight into how gene variants can be linked to genetic disorders. An estimated one in 25 children are affected by a genetic disorder, of which some may be apparent at birth, while others develop at later stages of life. The likelihood that a variant is pathogenic (and therefore may contribute to a genetic disease), or alternatively is benign, is calculated using certain criteria. In particular, how common the variant is compared to the incidence of disease in a population.

Consequently, it is important to have access to different genetic datasets and population data, in order to improve the accuracy of research. As a case study, Dr Diamond explained how the distribution of genetic variants can differ geographically. The disease hypertrophic cardiomyopathy (HCM) is linked to six common genes and more broadly to 20 additional genes. Certain variants in these genes are rare in European populations, but are surprisingly common in East Asian populations. Nonetheless, a low incidence of disease in an East Asian population indicates that these variants are not pathogenic.

'This is why I'm really excited about the prospect of these very large datasets like 100,000 Genomes becoming available to us,' Dr Diamond said.

Professor John Mattick, chief executive of Genomics England and chair of the event, added that research will also look for beneficial variants, such as those associated with longevity and good health in old age. In further questions, the panel discussed whether private companies should contribute datasets, and Professor Mattick commented that the NHS can currently only take information from accredited laboratories. As for aggregating genetic data internationally, he stated that individual rights and privacy according to jurisdiction should be maintained.

What does the public think of genomic research? Natalie Frankish, policy and engagement manager at Genetic Alliance UK, gave insight into patients' perspectives. Based on two surveys carried out in 2015 and 2016, it seems patients want as much information as possible from their genomes. A high number are willing to allow their genomes to be used in further research, provided they know which organisations will be using the data, and what they will be looking for. Patients show a keen understanding that for research into rare genetic conditions, every bit of data counts. But they also express a desire for dynamic consent, with opportunities to change preferences or opt out as research progresses.

Genetic Alliance UK also found a scale of trust when it came to organisations that might carry out this research. Around 80 percent would trust the NHS, 77 percent would trust universities, 63 percent charities, around 40 percent would trust private companies, and fewer would trust the government.

Frankish emphasised that trust is key to enable the relationship between the public and researchers to flourish. The public wants its consent to be sought, and data to be stored securely. After all, only three out of 63 participants would be comfortable for private companies to use genetic data without further detail on how it would be used.

Phil Booth, coordinator of medConfidential – an organisation campaigning for citizens' control of their personal data – spoke with caution about data security and the potential for misuse. The stakes are high when it comes to trust, he warned, referencing the case of the UK government's care.data programme which was launched in 2013. The project, which would have seen the creation of a national database of medical records, had to close in July 2016 as public mistrust led to nearly two million patients opting out. 'So we have to get this right,' he said.

DNA is unique information, especially if sequence data is linked to records of dated, medical events. With increasing and varied genomic research on the horizon, the value of the information in a stored genome would not decrease over time, either. He advised that strong, statutory protections needed to be put in place early to protect access to data, and cited warning examples such as the use of DNA data for identification and reidentification, as has been carried out by police. He also mentioned the fear that any dataset may be 'fair game' to use in strict government policies.

Last to speak was Dr Sarah Chan of the University of Edinburgh, and member of Genomics England's Ethics Advisory Committee. She turned the focus onto the question of 'whose genome it is', reimagining it as 'to whom does it belong'. There are many rights this question could encompass. Who has the right to profit from a genome, or to access it or manipulate it?

Only a tiny amount – around 0.1 percent – of the human genome is unique to each person. But when compared with other people in a population, the variance reveals information about them. If this information could improve medical diagnosis and treatment, then bioethicists may argue there is a moral obligation to participate in genomic research. It is possible that the real problem is how to protect people from misuse of their data, as no data system is ever going to be 100 percent secure.

But if people just do not participate in genomic research, then research will be done on small or irrelevant populations that will produce little benefit. Dr Chan gave the example of clinical trials where lack of diversity in participants meant that adverse effects to drugs were not detected.

Fielding questions from the audience, she emphasised that the public does not need merely to understand the usefulness of its data, but also to participate and be engaged in decisions over the use of this data. While currently it is most urgent and cost-effective to offer sequencing to patients who may have genetic diseases, we may be on a journey to a place where genomic research is offered more commonly to the public.

The public needs to be involved in what is done with its genomes, Dr Chan advised. Even though genomic research is commonly associated with personalised, individual treatment, 'we are all in this together.'

The Progress Educational Trust is grateful to Genomics England for collaborating on this event.