English-Video.net comment policy

The comment field is common to all languages

Let's write in your language and use "Google Translate" together

Please refer to informative community guidelines on TED.com

TED2017

Jun Wang: How digital DNA could help you make better health choices

Filmed:
1,209,741 views

What if you could know exactly how food or medication would impact your health -- before you put it in your body? Genomics researcher Jun Wang is working to develop digital doppelgangers for real people; they start with genetic code, but they'll also factor in other kinds of data as well, from food intake to sleep to data collected by a "smart toilet." With all of this valuable information, Wang hopes to create an engine that will change the way we think about health, both on an individual level and as a collective.

- Genomics researcher
At iCarbonX, Jun Wang aims to establish a big data platform for health management. Full bio

Today I'm here, actually,
to pose you a question.
00:12
What is life?
00:16
It has been really puzzling me
for more than 25 years,
00:17
and will probably continue doing so
for the next 25 years.
00:21
This is the thesis I did
when I was still in undergraduate school.
00:25
While my colleagues still treated
computers as big calculators,
00:31
I started to teach computers to learn.
00:38
I built digital lady beetles
00:41
and tried to learn from real lady beetles,
just to do one thing:
00:44
search for food.
00:49
And after very simple neural network --
00:51
genetic algorithms and so on --
00:54
look at the pattern.
00:56
They're almost identical to real life.
00:57
A very striking learning experience
for a twenty-year-old.
01:01
Life is a learning program.
01:07
When you look
at all of this wonderful world,
01:12
every species has
its own learning program.
01:15
The learning program is genome,
01:19
and the code of that program is DNA.
01:22
The different genomes of each species
represent different survival strategies.
01:27
They represent hundreds of millions
of years of evolution.
01:33
The interaction between
every species' ancestor
01:38
and the environment.
01:42
I was really fascinated about the world,
01:46
about the DNA,
01:48
about, you know, the language of life,
01:49
the program of learning.
01:52
So I decided to co-found
the institute to read them.
01:54
I read many of them.
01:59
We probably read more than half
of the prior animal genomes in the world.
02:01
I mean, up to date.
02:06
We did learn a lot.
02:09
We did sequence, also,
one species many, many times ...
02:11
human genome.
02:15
We sequenced the first Asian.
02:16
I sequenced it myself many, many times,
02:18
just to take advantage of that platform.
02:21
Look at all those repeating base pairs:
02:24
ATCG.
02:27
You don't understand anything there.
02:29
But look at that one base pair.
02:31
Those five letters, the AGGAA.
02:32
These five SNPs represent
a very specific haplotype
02:35
in the Tibetan population
02:39
around the gene called EPAS1.
02:41
That gene has been proved --
02:43
it's highly selective --
02:45
it's the most significant signature
of positive selection of Tibetans
02:46
for the higher altitude adaptation.
02:50
You know what?
02:53
These five SNPs were the result
of integration of Denisovans,
02:54
or Denisovan-like individuals into humans.
03:00
This is the reason
why we need to read those genomes.
03:04
To understand history,
03:06
to understand what kind
of learning process
03:08
the genome has been through
for the millions of years.
03:12
By reading a genome,
it can give you a lot of information --
03:17
tells you the bugs in the genome --
03:20
I mean, birth defects,
monogenetic disorders.
03:22
Reading a drop of blood
03:25
could tell you why you got a fever,
03:26
or it tells you which medicine
and dosage needs to be used
03:28
when you're sick, especially for cancer.
03:31
A lot of things could be studied,
but look at that:
03:35
30 years ago, we were still poor in China.
03:38
Only .67 percent of the Chinese
adult population had diabetes.
03:43
Look at now: 11 percent.
03:47
Genetics cannot change over 30 years --
03:49
only one generation.
03:53
It must be something different.
03:54
Diet?
03:56
The environment?
03:57
Lifestyle?
03:59
Even identical twins
could develop totally differently.
04:01
It could be one becomes
very obese, the other is not.
04:07
One develops a cancer
and the other does not.
04:11
Not mentioning living
in a very stressed environment.
04:13
I moved to Shenzhen 10 years ago ...
04:19
for some reason, people may know.
04:22
If the gene's under stress,
04:25
it behaves totally differently.
04:27
Life is a journey.
04:30
A gene is just a starting point,
04:32
not the end.
04:35
You have this statistical risk
of certain diseases when you are born.
04:37
But every day you make different choices,
04:42
and those choices will increase
or decrease the risk of certain diseases.
04:45
But do you know
where you are on the curve?
04:51
What's the past curve look like?
04:54
What kind of decisions
are you facing every day?
04:56
And what kind of decision is the right one
04:59
to make your own right curve
over your life journey?
05:02
What's that?
05:07
The only thing you cannot change,
05:09
you cannot reverse back,
05:11
is time.
05:13
Probably not yet; maybe in the future.
05:14
(Laughter)
05:16
Well, you cannot change
the decision you've made,
05:17
but can we do something there?
05:20
Can we actually try to run
multiple options on me,
05:22
and try to predict right
on the consequence,
05:27
and be able to make the right choice?
05:31
After all,
05:34
we are our choices.
05:35
These lady beetles came to me afterwards.
05:38
25 years ago, I made
the digital lady beetles
05:41
to try to simulate real lady beetles.
05:45
Can I make a digital me ...
05:47
to simulate me?
05:49
I understand the neural
network could become
05:51
much more sophisticated
and complicated there.
05:54
Can I make that one,
05:57
and try to run multiple options
on that digital me --
05:59
to compute that?
06:03
Then I could live in different universes,
06:05
in parallel, at the same time.
06:08
Then I would choose
whatever is good for me.
06:11
I probably have the most comprehensive
digital me on the planet.
06:14
I've spent a lot of dollars
on me, on myself.
06:18
And the digital me told me
I have a genetic risk of gout
06:21
by all of those things there.
06:27
You need different technology to do that.
06:29
You need the proteins, genes,
06:31
you need metabolized antibodies,
06:32
you need to screen all your body
06:35
about the bacterias and viruses
covering you, or in you.
06:38
You need to have
all the smart devices there --
06:41
smart cars, smart house, smart tables,
06:44
smart watch, smart phone
to track all of your activities there.
06:47
The environment is important --
06:51
everything's important --
06:52
and don't forget the smart toilet.
06:54
(Laughter)
06:55
It's such a waste, right?
06:56
Every day, so much invaluable information
just has been flushed into the water.
06:58
And you need them.
07:04
You need to measure all of them.
07:06
You need to be able to measure
everything around you
07:07
and compute them.
07:10
And the digital me told me
I have a genetic defect.
07:12
I have a very high risk of gout.
07:16
I don't feel anything now,
07:19
I'm still healthy.
07:21
But look at my uric acid level.
07:22
It's double the normal range.
07:24
And the digital me searched
all the medicine books,
07:26
and it tells me, "OK, you could
drink burdock tea" --
07:29
I cannot even pronounce it right --
07:33
(Laughter)
07:35
That is from old Chinese wisdom.
07:36
And I drank that tea for three months.
07:39
My uric acid has now gone back to normal.
07:41
I mean, it worked for me.
07:45
All those thousands of years
of wisdom worked for me.
07:46
I was lucky.
07:49
But I'm probably not lucky for you.
07:50
All of this existing
knowledge in the world
07:55
cannot possibly be efficient enough
or personalized enough for yourself.
07:57
The only way to make
that digital me work ...
08:03
is to learn from yourself.
08:07
You have to ask a lot
of questions about yourself:
08:11
"What if?" --
08:13
I'm being jet-lagged now here.
08:15
You don't probably see it, but I do.
08:17
What if I eat less?
08:20
When I took metformin,
supposedly to live longer?
08:21
What if I climb Mt. Everest?
08:25
It's not that easy.
08:26
Or run a marathon?
08:28
What if I drink a bottle of mao-tai,
08:30
which is a Chinese liquor,
08:32
and I get really drunk?
08:33
I was doing a video rehearsal last time
with the folks here,
08:35
when I was drunk,
08:39
and I totally delivered
a different speech.
08:40
(Laughter)
08:42
What if I work less, right?
08:45
I have been less stressed, right?
08:48
So that probably never happened to me,
08:50
I was really stressed every day,
08:51
but I hope I could be less stressed.
08:53
These early studies told us,
08:56
even with the same banana,
08:58
we have totally different
glucose-level reactions
09:00
over different individuals.
09:03
How about me?
09:04
What is the right breakfast for me?
09:06
I need to do two weeks
of controlled experiments,
09:08
of testing all kinds of different
food ingredients on me,
09:11
and check my body's reaction.
09:15
And I don't know
the precise nutrition for me,
09:17
for myself.
09:20
Then I wanted to search
all the Chinese old wisdom
09:23
about how I can live longer,
and healthier.
09:27
I did it.
09:30
Some of them are really unachievable.
09:32
I did this once last October,
09:34
by not eating for seven days.
09:37
I did a fast for seven days
with six partners of mine.
09:40
Look at those people.
09:44
One smile.
09:46
You know why he smiled?
09:47
He cheated.
09:48
(Laughter)
09:49
He drank one cup of coffee at night,
09:50
and we caught it from the data.
09:53
(Laughter)
09:55
We measured everything from the data.
09:56
We were able to track them,
09:58
and we could really see --
10:01
for example, my immune system,
10:02
just to give you a little hint there.
10:04
My immune system changed
dramatically over 24 hours there.
10:06
And my antibody regulates my proteins
10:11
for that dramatic change.
10:15
And everybody was doing that.
10:16
Even if we're essentially
totally different at the very beginning.
10:18
And that probably will be
an interesting treatment in the future
10:21
for cancer and things like that.
10:24
It becomes very, very interesting.
10:26
But something you probably
don't want to try,
10:28
like drinking fecal water
from a healthier individual,
10:31
which will make you feel healthier.
10:34
This is from old Chinese wisdom.
10:36
Look at that, right?
10:38
Like 1,700 years ago,
10:39
it's already there, in the book.
10:41
But I still hate the smell.
10:44
(Laughter)
10:46
I want to find out the true way to do it,
10:47
maybe find a combination of cocktails
of bacterias and drink it,
10:49
it probably will make me better.
10:54
So I'm trying to do that.
10:55
Even though I'm trying this hard,
10:56
it's so difficult to test out
all possible conditions.
11:00
It's not possible to do
all kinds of experiments at all ...
11:05
but we do have seven billion
learning programs on this planet.
11:11
Seven billion.
11:15
And every program
is running in different conditions
11:16
and doing different experiments.
11:20
Can we all measure them?
11:21
Seven years ago,
I wrote an essay in "Science"
11:24
to celebrate the human genome's
10-year anniversary.
11:28
I said, "Sequence yourself,
11:32
for one and for all."
11:33
But now I'm going to say,
11:35
"Digitalize yourself for one and for all."
11:37
When we make this digital me
into a digital we,
11:42
when we try to form an internet of life,
11:47
when people can learn from each other,
11:51
when people can learn
from their experience,
11:54
their data,
11:57
when people can really form
a digital me by themselves
11:59
and we learn from it,
12:02
the digital we will be
totally different with a digital me.
12:05
But it can only come from the digital me.
12:11
And this is what I try to propose here.
12:16
Join me --
12:20
become we,
12:21
and everybody should build up
their own digital me,
12:23
because only by that
will you learn more about you,
12:28
about me,
12:33
about us ...
12:34
about the question I just posed
at the very beginning:
12:36
"What is life?"
12:40
Thank you.
12:42
(Applause)
12:43
Chris Anderson:
One quick question for you.
12:49
I mean, the work is amazing.
12:52
I suspect one question people have is,
12:54
as we look forward to these amazing
technical possibilities
12:58
of personalized medicine,
13:01
in the near-term it feels like
they're only going to be affordable
13:02
for a few people, right?
13:06
It costs many dollars to do
all the sequencing and so forth.
13:07
Is this going to lead to a kind of,
13:10
you know, increasing inequality?
13:13
Or do you have this vision
that the knowledge that you get
13:16
from the pioneers
13:20
can actually be
pretty quickly disseminated
13:21
to help a broader set of recipients?
13:23
Jun Wang: Well, good question.
13:27
I'll tell you that seven years ago,
when I co-founded BGI,
13:29
and served as the CEO
of the company there,
13:32
the only goal there for me to do
13:36
was to drive the sequencing cost down.
13:38
It started from 100 million dollars
per human genome.
13:41
Now, it's a couple hundred dollars
for a human genome.
13:43
The only reason to do it
is to get more people to benefit from it.
13:46
So for the digital me,
it's the same thing.
13:50
Now, you probably need,
13:52
you know, one million dollars
to digitize a person.
13:54
I think it has to be 100 dollars.
13:57
It has to be free for many of those people
that urgently need that.
13:59
So this is our goal.
14:04
And it seems that with all
this merging of the technology,
14:05
I'm thinking that in the very near future,
14:09
let's say three to five years,
14:12
it will come to reality.
14:14
And this is the whole idea
of why I founded iCarbonX,
14:15
my second company.
14:19
It's really trying to get the cost down
14:21
to a level where every individual
could have the benefit.
14:24
CA: All right, so the dream is not
elite health services for few,
14:27
it's to really try
14:30
and actually make overall health care
much more cost effective --
14:31
JW: But we started
from some early adopters,
14:34
people believing ideas and so on,
14:37
but eventually, it will become
everybody's benefit.
14:39
CA: Well, Jun, I think
it's got to be true to say
14:44
you're one of the most amazing
scientific minds on the planet,
14:46
and it's an honor to have you.
14:49
JW: Thank you.
14:51
(Applause)
14:52

▲Back to top

About the speaker:

Jun Wang - Genomics researcher
At iCarbonX, Jun Wang aims to establish a big data platform for health management.

Why you should listen

In 1999, Jun Wang founded the Bioinformatics Department of Beijing Genomics Institute (BGI, now known as BGI Shenzhen), one of China’s premier research facilities. Until July 2015, Wang led the institution of 5,000+ people engaged in studies of genomics and its informatics, including genome assembly, annotation, expression, comparative genomics, molecular evolution, transcriptional regulation, genome variation analysis, database construction as well as methodology development such as the sequence assembler and alignment tools. He also focuses on interpretation of the definition of "gene" by expression and conservation study. In 2003, Wang was also involved in the SARS genome analysis and the silkworm genome assembly and analysis in cooperation with Chinese Southeast Agricultural University. The Pig Genome Project was completed at BGI under his leadership, as well as the chicken genome variation map and the TreeFam in collaboration with the Sanger Institute. In 2007, he and his group finished the first Asian diploid genome, the 1000 genome project, and many more projects. He initiated the "million genomes project" which seeks to better understand health based on human, plant, animal and micro-ecosystem genomes.

In late 2015, Wang founded a new institute/company, iCarbonX, aiming to develop an artificial intelligence engine to interpret and mine multiple health-related data and help people better manage their health and defeat disease.

More profile about the speaker
Jun Wang | Speaker | TED.com