Health & Medicine

Does a Chinese company's combination of genomics, big data and AI offer hope for a healthier future?

WangJianGrandChallenges.jpg

Wang Jian, in athletic jacket at Gates Foundation's Grand Challenges meeting in Beijing, has built BGI (formerly Beijing Genomics Institute) into the world's leading sequencer of genomes, and a serious contender on genomic research.

Credit:

Mary Kay Magistad

You’ve got your Fitbits, and Apple Watches and assorted gadgets that keep track of how and what you’re doing — how far you’re walking, or running, and how fast, at what heartrate, how well you sleep, how much you fidget, whether you slouch.

Player utilities

Listen to the Story.

And you can take it to the next level. Companies like 23&Me will sequence your genome. You can see where your ancestors came from, how much Neanderthal DNA you have compared to the general population, and what risk factors you have for disease.

Perhaps you don’t want to know. But perhaps, if you do have elevated risk factors — you can take steps to reduce the risk. After all, genes only go so far in determining what happens to you. Your lifestyle choices — how much you sleep and exercise, what you eat — can have profound effects on how some genes are expressed.  And now, research is also focused on the role your gut bacteria plays.

All this is fascinating to scientists, and promising for a global population that’s living longer. After all, it benefits everyone if those who live longer are healthier, longer.

In southern China, in the city of Shenzhen — right next to Hong Kong — is a company called BGI, originally Beijing Genomics Institute, that is the world’s leading sequencer of genomes — human, animal, plant. It has big ambitions to be a global leader in life-sciences research. It’s already working with top research institutes around the world, and has published in world-class journals. BGI’s work hasn’t been without controversy — one of its many focuses has been on finding the genetic basis for genius, raising concerns about a future of designer babies. But much of BGI's other research has gained respect among global peers.  

BGI sequencing achievements
Credit:

Mary Kay Magistad

In October, BGI hosted a conference that drew hundreds of geneticists and other researchers from leading institutions around the globe. One focus was the promise of precision medicine, based on collecting lots of data. This area so fascinates BGI’s former chief executive, Wang Jun, that he stepped down from BGI this past summer to start a new venture of his own, iCarbonX. I is for internet. X is for what is still unknown in how different biological systems work together and influence each other. And carbon?

“We are carbon-based,” Wang Jun says. “We are basically a program, right? And DNA is just a form of programs. Why is that? We all know every individual is running a life program, if you think of the genome.”

The journal Nature has called Wang Jun visionary, and revered by those he, along with founder Wang Jian, has helped lead at BGI. He sees great promise in synthesizing Big Data, sophisticated technologies and ever smarter and faster computers, to better understand life.

President Barack Obama appears to have been on the same track when earlier this year he announced a national project to sequence a million genomes, to move away from “one size fits all” medicine, and toward precision medicine — medicine targeted for your specific genotype and phenotype.  Wang Jun says — a million genomes, great.  Why stop there? 

“I don’t think that’s the right goal,” he says. "We have to collect one million omics data for the same individual."

Omics is a layer of information about life, such as genomics.  Wang Jun envisions sequencing not just genomes, but behavior data, data from biosensors, imaging data, environmental data, all for the same individual, over time. Now that, he says, should yield important insights about what interactions matter in causing genes to be expressed in a certain way, or not at all.

Wang Jun, former BGI Chief Executive, with Bill Gates, in photo displayed at BGI
Credit:

Mary Kay Magistad

But how to get so much data from so many people? Wang Jun says his company aims to start by offering to sequence people’s genomes for free.

“The reason is simple,” he says. “The value of the genome right now, to a person, to an ordinary person like you, is not worth $100 now. Well, maybe you don’t really agree with me. But how many people want to pay $100 to get their genome sequenced, if you’re not a researcher like me? But I strongly believe, after accumulating a million genomes there, the value of that is more than $10,000, a million dollars, to you.”

But what about you — the patient — sitting there in the doctor’s office, wanting to know what your genome tells you about what to do about your health, in your life, right now? Well, that’s still a work in progress.

“When you talk about precision medicine from a genome sequencing perspective, you start with your traditional sequencers, one to four days,” says Ketan Paranjape, Intel’s general manager of Life Sciences and Analytics, speaking at the BGI conference. “The analysis takes a few weeks, depending on what tools you use, what technologies you use. And finally, you get into the mode of actually finding clinically actionable components, that you can make a treatment decision on. That is basically the key problem. This takes forever.”

What if all that sequencing and processing time could be cut down to one day, Paranjape asks. What if that could happen by 2020? That could certainly change the way medicine is practiced. Patients could be diagnosed and prescribed treatments that have a better chance of working with their own specific physiological make-up.

But with the speed of sequencing, and processing, and analyzing, the range of new techniques in how to use genomes, comes a host of ethical questions.  

A group of international scientists met in Washington in early December, and called for a moratorium on editing the human genome, with a technique, just three years old, called CRISPR.  This technique allows scientists to, fairly easily and quickly, delete, replace or change DNA. These changes can be passed on to the next generation. That’s desirable when trying to breed a malaria-free mosquito; more complicated when editing the human genome.

It’s tempting to fantasize about a world in which genetic predispositions to cancer, Alzheimer’s or multiple sclerosis, can simply be snipped out of an individual’s genome, sparing that person and all his or her descendents that misfortune. But what might the unintended consequences be, even if excising of the targeted gene goes smoothly — itself, not a given?

Related: 4 things you wanted to know about gene editing

The one experiment done on humans, on nonviable human embryos at Sun Yat Sen University in southern China, was a cautionary tale. The researchers used CRISPR on 85 different embryos, trying to remove just one gene without affecting others — and they failed in each case

Wired magazine put the quandary succinctly in a tagline in a recent article it did on CRISPR: “We now have the power to quickly and easily alter DNA. It could eliminate disease. It could solve world hunger. It could provide unlimited clean energy. It could really get out of hand.” Wang Jun agrees. It’s prudent to go slowly. 

“So of course, people are also talking about how to change, or rebuild, the whole life information, right, which I think is too dangerous for a species like humans,” he says. “We don’t even understand humans as a system, yet. Don’t just try to change it from the basic level. Maybe we can start with bacteria, or yeast. But it’s still too early to think about humans.

And then there’s the question of privacy.  In this brave new world of combining genomics and big data, how much privacy do you give up, and to whom, and for whose benefit? For the pharma companies to profit from your information? Do you get anything back from that? Do you lose anything when so many more people know the essence of what physiologically makes you, you?

Interestingly, the attitude of several presenters at the conference seemed to be — sharing is good. Get it all out there. They put their medical and genomic data up on Power Point slides.  Hundreds of people were in the audience. Some took notes. Some took photos on their mobile phones. 

Americans might understandably cringe. Even with the Affordable Care Act, preventing insurance companies from denying insurance to people with pre-existing conditions, there are still ways they could use such data that could be detrimental to the patient. And that’s not even going into stigma attached to certain conditions or genetic predispositions.

If your genome says you have a high risk of getting cancer, Alzheimer’s or becoming an alcoholic, how widely would you want to share that information? With a prospective employer?  With someone you’re dating? How would your family members feel about you sharing such data? After all, they might have some of the same genes. Few at the BGI conference seemed to have such qualms. Wang Jun put information from his own genomic sequencing up on PowerPoint.

“So I did my personal genetic testing many times, through Illumina Complete Genomics technology, using my research grant — which is great, because I get it for free,” he says. “So I probably have one of the most accurate genomes in the world.  And I’ve looked at it, many many times.”

Wang Jun points out that he has one gene that makes him more likely to have high uric acid levels, and get gout. He had his uric acid levels tested, and sure enough, they were twice what they should have been. He started digging around in Traditional Chinese Medicine books, found that burdock root tea has worked in preventing gout, and started drinking it. His uric acid levels went down.

He also talked about some of his genetic risk factors, like, a gene he said he and half the Chinese population carry, for a disease that can potentially be deadly. He points to this as an example of why it’s important to pool data, to better understand how such genes are expressed in a wider population. If millions of people have that same gene, and only some get sick, what made the difference? What could others learn from that?

Nikolas Rose, who heads the social science, health & medicine department at King’s College in London, says there is a compelling argument for sharing data — under the right conditions. Not everyone is going to want to be as public about their data as Wang Jun was, and Rose believes current methods of trying to ensure confidentiality and anonymity are insufficient.

“You cannot give meaningful consent to every future use of your data, because you have absolutely no idea what that is,” he says. “And even the most technologically advanced means for ensuring anonymity, stripping off your personal identifiers, can be cracked with one or two identifiers left. ... Experts assure me that quantum computing can defeat any technological measure to preserve anonymity."

Rose says rather than simply try to preserve privacy and anonymity, society should start to act against misuses of data, against genetic discrimination and stigma.

“I think the US is actually a good example in its Genetic Non-Discrimination Act, in making discrimination on the grounds of genetic information illegal, and I would argue that that act should be extended not only to genetic information, but to neurobiological information,” he says. "Because, of course, it’s not just genetic information that’s going up online. It’s clinical information, and it’s brain scan information and other scanning information, imaging information, that’s being digitized and going up online.”

Of course, regulations won’t stop malevolent hackers from getting into people’s data, and possibly reveling more than they want the world to know. 

“I just think we have to continually upgrade our anti-hacking software, to protect ourselves against those risks,” Rose says. 

And if that’s not entirely reassuring to those unconvinced they should be sharing and pooling their genomic data, Rose argues that this is the world in which we now live, and we do best to harness the best uses of new technologies, while protecting against the worst abuses of it.

“There’s a word in technology development called ‘path dependency.’ Once you’re down a certain path, it’s impossible to go back,” he says. “Twenty years ago, perhaps we should have said, ‘ah, the Internet.  It’s going to raise all these problems 20 years down the line, when all our personal information is going to be out there. Let’s not have the Internet. But that’s an impossible thing to uninvent, once we’re there. So we have to try to find realistic measures, as I say, to deal with the abuses, but not pretend that we can not be where we are.”

And here we are, in a world with ever-faster, cheaper ways of sequencing genomes, monitoring biometrics, pooling and analyzing data, editing genes. These are tools. Tools can be used in all kinds of ways. A hammer can be used to build or to destroy.

As the evolution of these tools accelerates, scientists and bioethicists are continuing an active conversation, along with a widespread willingness to tap the brakes and think — really think — about where certain roads might lead. 

Jennifer Doudna, a biochemist and one of the discoverers of the gene editing technique CRISPR, told the New Yorker she lies awake some nights and wonders if CRISPR will eventually do more harm than good. She said she once dreamt that she met Adolf Hitler, who asked her for more details on the uses and implications of what he called, "this amazing technology."

When it comes to understanding and harnessing all these new techniques and technology to understand life — the code of life, genomics, and how to alter and influence it — this will be a transformative century. But transformed how?  ... In some ways, in important ways — that’s up to us — to shape it and steer it as we go.