The medical potential of AI and metabolites | Leila Pirhaji

68,498 views ・ 2019-11-20

TED


μ•„λž˜ μ˜λ¬Έμžλ§‰μ„ λ”λΈ”ν΄λ¦­ν•˜μ‹œλ©΄ μ˜μƒμ΄ μž¬μƒλ©λ‹ˆλ‹€.

λ²ˆμ—­: Jungmin Hwang κ²€ν† : Taz B K
00:13
In 2003,
0
13507
1889
2003λ…„
00:15
when we sequenced the human genome,
1
15420
2913
인간 μœ μ „μ²΄ μ„œμ—΄μ˜ ν•΄λ…μœΌλ‘œ
00:18
we thought we would have the answer to treat many diseases.
2
18357
3922
λ§Žμ€ μ§ˆλ³‘μ˜ μΉ˜λ£Œμ— λŒ€ν•œ κΈ°λŒ€κ°€ μ»Έμ§€λ§Œ
00:22
But the reality is far from that,
3
22974
2657
ν˜„μ‹€κ³ΌλŠ” 거리가 λ©€μ—ˆμ–΄μš”.
00:26
because in addition to our genes,
4
26782
1921
μœ μ „μ²΄ 말고도
00:28
our environment and lifestyle could have a significant role
5
28727
4570
ν™˜κ²½κ³Ό μƒν™œ 방식이 λ§Žμ€ μ£Όμš” μ§ˆλ³‘λ“€μ—
00:33
in developing many major diseases.
6
33321
2548
μƒλ‹Ήν•œ 영ν–₯을 미치기 λ•Œλ¬Έμ΄μ£ .
00:35
One example is fatty liver disease,
7
35893
3580
ν•œ 예둜, 지방간 μ§ˆν™˜μœΌλ‘œ
00:39
which is affecting over 20 percent of the population globally,
8
39497
4083
μ „ 세계 20%κ°€ λ„˜λŠ” μ‚¬λžŒλ“€μ΄ κ³ μƒν•˜κ³  μžˆμ§€λ§Œ
00:43
and it has no treatment and leads to liver cancer
9
43604
3034
μΉ˜λ£Œλ²•μ΄ μ—†μ–΄ κ°„μ•”μ΄λ‚˜ κ°„λΆ€μ „μœΌλ‘œ μ§„ν–‰ν•˜κ²Œ λ©λ‹ˆλ‹€.
00:46
or liver failure.
10
46662
1423
00:49
So sequencing DNA alone doesn't give us enough information
11
49517
4744
DNA μ—ΌκΈ°μ„œμ—΄μ˜ ν•΄λ…λ§ŒμœΌλ‘œλŠ” 효과적인 μΉ˜λ£Œλ²•μ„
00:54
to find effective therapeutics.
12
54285
2232
찾을 수 μ—†λŠ” 이유죠.
00:56
On the bright side, there are many other molecules in our body.
13
56541
3756
λ‹€ν–‰νžˆ μ²΄λ‚΄μ—λŠ” λ‹€λ₯Έ λ§Žμ€ λΆ„μž ꡬ쑰듀이,
01:00
In fact, there are over 100,000 metabolites.
14
60321
3980
μ‹€μ œλ‘œ 10만개 μ΄μƒμ˜ λŒ€μ‚¬λ¬Όμ΄ μ‘΄μž¬ν•©λ‹ˆλ‹€.
01:04
Metabolites are any molecule that is supersmall in their size.
15
64325
4296
μƒλ‹Ήνžˆ μž‘μ€ 크기의 λΆ„μž ꡬ쑰λ₯Ό λŒ€μ‚¬λ¬Όμ΄λΌκ³  ν•˜λŠ”λ°
01:09
Known examples are glucose, fructose, fats, cholesterol --
16
69193
4972
포도당, κ³Όλ‹Ή, 지방, μ½œλ ˆμŠ€ν…Œλ‘€ λ“±
01:14
things we hear all the time.
17
74189
1510
항상 λ“£λŠ” 것듀이죠.
01:16
Metabolites are involved in our metabolism.
18
76273
2983
μ‹ μ§„λŒ€μ‚¬μ— 영ν–₯을 μ£ΌλŠ” λŒ€μ‚¬λ¬Όμ€
01:20
They are also downstream of DNA,
19
80066
4028
DNA의 μ•„λž˜ 뢀뢄에 μœ„μΉ˜ν•΄
01:24
so they carry information from both our genes as well as lifestyle.
20
84118
5082
μœ μ „μžμ™€ μƒν™œλ°©μ‹μ— λŒ€ν•œ λͺ¨λ“  정보λ₯Ό 가지고 μžˆμ–΄μ„œ
01:29
Understanding metabolites is essential to find treatments for many diseases.
21
89224
5649
이 λ¬Όμ§ˆμ— λŒ€ν•œ μ΄ν•΄λŠ” λ§Žμ€ μ§ˆλ³‘μ˜ μΉ˜λ£Œλ²•μ„ μ°ΎλŠ” 데에 ν•„μˆ˜μ μž…λ‹ˆλ‹€.
01:34
I've always wanted to treat patients.
22
94897
2212
늘 ν™˜μžλ“€μ„ μΉ˜λ£Œν•˜λŠ”κ²Œ κΏˆμ΄μ—ˆμ§€λ§Œ
01:37
Despite that, 15 years ago, I left medical school,
23
97934
3858
15λ…„ 전에 μ œκ°€ μ˜λŒ€λ₯Ό κ·Έλ§Œλ‘” 건
01:41
as I missed mathematics.
24
101816
1965
μˆ˜ν•™μ— λŒ€ν•œ λ―Έλ ¨ λ•Œλ¬Έμ΄μ—ˆμ–΄μš”.
01:45
Soon after, I found the coolest thing:
25
105019
2936
곧 μ•Œκ²Œ 된 μ•„μ£Ό μ‹ λ‚˜λŠ” 일은
01:48
I can use mathematics to study medicine.
26
108692
2763
μ˜ν•™μ— μˆ˜ν•™μ„ ν™œμš©ν•  수 μžˆλ‹€λŠ” μ‚¬μ‹€μ΄μ—ˆμ£ .
01:53
Since then, I've been developing algorithms to analyze biological data.
27
113026
5213
κ·Έλ•ŒλΆ€ν„° μ €λŠ” 생물학 자료의 뢄석 μ•Œκ³ λ¦¬μ¦˜μ„ κ°œλ°œν•˜κ³  μžˆμ–΄μš”.
01:59
So, it sounded easy:
28
119092
2283
μ²˜μŒμ—” μ•„μ£Ό μ‰½κ²Œ λ“€λ Έμ£ .
02:01
let's collect data from all the metabolites in our body,
29
121399
3601
μ²΄λ‚΄μ˜ λͺ¨λ“  λŒ€μ‚¬λ¬Όμ— λŒ€ν•œ μžλ£Œλ“€μ„ λͺ¨μ•„μ„œ
02:05
develop mathematical models to describe how they are changed in a disease
30
125024
5128
λ³‘μœΌλ‘œ λ°œμ „ν•˜λŠ” 과정을 λ³΄μ—¬μ£ΌλŠ” μˆ˜ν•™μ  λͺ¨ν˜•μ„ λ§Œλ“€κ³ 
02:10
and intervene in those changes to treat them.
31
130176
2988
κ·Έ 과정에 κ°œμž…ν•΄ μΉ˜λ£Œν•œλ‹€λŠ” 건데
02:14
Then I realized why no one has done this before:
32
134488
3472
μ™œ 아무도 이걸 해보지 μ•Šμ•˜λŠ”μ§€ 곧 μ•Œκ²Œ λμ–΄μš”.
02:19
it's extremely difficult.
33
139230
1687
μ™„μ „ μ–΄λ ΅λ”λΌκ³ μš”.
02:20
(Laughter)
34
140941
1087
(μ›ƒμŒ)
02:22
There are many metabolites in our body.
35
142052
2412
우리 λͺΈ μ•ˆμ˜ λ§Žμ€ λŒ€μ‚¬λ¬Όμ€
02:24
Each one is different from the other one.
36
144783
2500
각기 μ„±μ§ˆμ„ 달리 ν•©λ‹ˆλ‹€.
02:27
For some metabolites, we can measure their molecular mass
37
147307
3728
μ–΄λ–€ λŒ€μ‚¬λ¬Όμ€ μ§ˆλŸ‰ 뢄석기λ₯Ό 톡해
02:31
using mass spectrometry instruments.
38
151059
2593
λΆ„μžλŸ‰μ˜ 츑정이 κ°€λŠ₯ν•˜μ§€λ§Œ
02:33
But because there could be, like, 10 molecules with the exact same mass,
39
153676
4393
같은 μ§ˆλŸ‰μ˜ λΆ„μžκ°€ 10κ°œλ‚˜ 될 μˆ˜λ„ 있고
02:38
we don't know exactly what they are,
40
158093
1807
정체 μ‘°μ°¨ λͺ¨λ₯΄λŠ”
02:39
and if you want to clearly identify all of them,
41
159924
2774
κ·Έ λͺ¨λ“  λΆ„μžλ“€μ„ λͺ…ν™•νžˆ 식별해 λ‚΄λ €λ©΄
02:42
you have to do more experiments, which could take decades
42
162722
3104
더 λ§Žμ€ μ‹€ν—˜μ„ 거쳐야 ν•˜λŠ”λ°, μˆ˜μ‹­ λ…„ λ™μ•ˆ
02:45
and billions of dollars.
43
165850
1714
μˆ˜μ‹­ μ–΅μ˜ μ˜ˆμ‚°μ΄ λ“€κ² μ£ .
02:48
So we developed an artificial intelligence, or AI, platform, to do that.
44
168207
5563
κ·Έλž˜μ„œ 인곡지λŠ₯ ν”Œλž«νΌμ„ κ°œλ°œν–ˆμ–΄μš”.
02:53
We leveraged the growth of biological data
45
173794
2844
생물학 μžλ£Œλ“€μ˜ 증가λ₯Ό ν™œμš©ν•΄
02:56
and built a database of any existing information about metabolites
46
176662
4424
λŒ€μ‚¬λ¬Όκ³Ό λ‹€λ₯Έ λΆ„μžμ™€μ˜ μƒν˜Έμž‘μš©μ— λŒ€ν•œ
03:01
and their interactions with other molecules.
47
181110
3128
κΈ°μ‘΄ μ •λ³΄μ˜ λ°μ΄ν„°λ² μ΄μŠ€λ₯Ό λ§Œλ“€κ³ 
03:04
We combined all this data as a meganetwork.
48
184262
3424
이 λͺ¨λ“  μžλ£Œλ“€μ„ λͺ¨μ•„ λ©”κ°€λ„€νŠΈμ›Œν¬λ₯Ό κ΅¬μΆ•ν–ˆμ–΄μš”.
03:07
Then, from tissues or blood of patients,
49
187710
3396
그리고 ν™˜μžλ“€μ˜ μ‘°μ§μ΄λ‚˜ ν˜ˆμ•‘μ—μ„œ
03:11
we measure masses of metabolites
50
191130
2751
λŒ€μ‚¬λ¬Ό μ§ˆλŸ‰μ„ μΈ‘μ •ν•΄
03:13
and find the masses that are changed in a disease.
51
193905
3259
λ°œλ³‘ μ‹œ μ§ˆλŸ‰μ˜ λ³€ν™”λ₯Ό μ‚΄νŽ΄λ΄€μ–΄μš”.
03:17
But, as I mentioned earlier, we don't know exactly what they are.
52
197188
3190
ν•˜μ§€λ§Œ λ§μ”€λ“œλ Έλ“―μ΄, μ •ν™•νžˆ 그것이 뭔지 μ•Œμ§€ λͺ»ν•˜μ£ .
03:20
A molecular mass of 180 could be either the glucose, galactose or fructose.
53
200402
5135
포도당, κ°ˆλ½ν† μ˜€μŠ€, κ³Όλ‹Ή λͺ¨λ‘ λΆ„μžλŸ‰μ΄ 180μ΄λ‹ˆκΉŒμš”.
03:25
They all have the exact same mass
54
205561
2019
μ •ν™•ν•˜κ²Œ 같은 μ§ˆλŸ‰μ„ 가지고 μžˆμ§€λ§Œ
03:27
but different functions in our body.
55
207604
2087
μ²΄λ‚΄μ—μ„œμ˜ κΈ°λŠ₯은 각기 λ‹€λ₯΄μ£ .
03:29
Our AI algorithm considered all these ambiguities.
56
209715
3587
이런 λͺ¨ν˜Έμ„±κΉŒμ§€ κ°μ•ˆν•˜λŠ” 저희 인곡지λŠ₯ μ•Œκ³ λ¦¬μ¦˜μ€
03:33
It then mined that meganetwork
57
213326
2736
λ©”κ°€λ„€νŠΈμ›Œν¬μ— λ“€μ–΄κ°€
03:36
to find how those metabolic masses are connected to each other
58
216086
4353
μ–΄λ–€ μƒν˜Έμž‘μš©μ„ 거쳐 이 λŒ€μ‚¬λ¬Όλ“€μ΄ λ³‘μœΌλ‘œ λ°œμ „ν•˜λŠ”μ§€
03:40
that result in disease.
59
220463
1958
λ°ν˜€ λ‚΄λ €κ³  ν•˜κ³  μžˆμ–΄μš”.
03:42
And because of the way they are connected,
60
222445
2238
μ—°κ²°λ˜μ–΄ μžˆλŠ” 방식에 따라
03:44
then we are able to infer what each metabolite mass is,
61
224707
4323
각 λŒ€μ‚¬λ¬Όμ˜ λΆ„μžλŸ‰ μœ μΆ”κ°€ κ°€λŠ₯ν•œλ°,
03:49
like that 180 could be glucose here,
62
229054
2924
μ—¬κΈ° 180은 포도당을 가리킀겠죠.
03:52
and, more importantly, to discover
63
232002
2551
더 μ€‘μš”ν•œ 사싀은
03:54
how changes in glucose and other metabolites
64
234577
3367
포도당과 λ‹€λ₯Έ λŒ€μ‚¬λ¬Όλ“€μ΄ μ–΄λ–€ 변이λ₯Ό 거쳐 병이 λ˜λŠ”μ§€
03:57
lead to a disease.
65
237968
1505
μ•Œ 수 μžˆλ‹€λŠ” κ²ƒμž…λ‹ˆλ‹€.
03:59
This novel understanding of disease mechanisms
66
239497
2995
μ§ˆλ³‘ ꡬ쑰의 이해λ₯Ό μœ„ν•œ 이런 μ°Έμ‹ ν•œ μ‹œλ„λ‘œ
04:02
then enable us to discover effective therapeutics to target that.
67
242516
4492
효과적인 μΉ˜λ£Œλ²•μ„ λ°œκ²¬ν•  수 μžˆλŠ” κ±°μ£ .
04:07
So we formed a start-up company to bring this technology to the market
68
247601
3845
κ·Έλž˜μ„œ 이 κΈ°μˆ μ„ μƒμš©ν™”ν•˜κ³  μ‚¬λžŒλ“€μ„ 돕기 μœ„ν•΄
04:11
and impact people's lives.
69
251470
1805
μƒˆλ‘œ νšŒμ‚¬λ₯Ό μ°¨λ Έμ–΄μš”.
04:13
Now my team and I at ReviveMed are working to discover
70
253722
3545
ν˜„μž¬ ReviveMedμ—μ„œ μΌν•˜λŠ” 저희 νŒ€μ€
04:17
therapeutics for major diseases that metabolites are key drivers for,
71
257291
5105
λ°œλ³‘ 원인이 λŒ€μ‚¬λ¬ΌμΈ μ£Όμš” μ§ˆλ³‘μ˜ μΉ˜λ£Œλ²•μ„ μ—°κ΅¬ν•˜κ³  μžˆμ–΄μš”.
04:22
like fatty liver disease,
72
262420
1897
μ§€λ°©κ°„μ˜ κ²½μš°μ—λŠ”
04:24
because it is caused by accumulation of fats,
73
264341
2924
간에 μžˆλŠ” λŒ€μ‚¬λ¬Ό 쀑 ν•˜λ‚˜μΈ
04:27
which are types of metabolites in the liver.
74
267289
2473
μ§€λ°©μ˜ 좕적이 κ·Έ 원인인데
04:29
As I mentioned earlier, it's a huge epidemic with no treatment.
75
269786
3940
λ§μ”€λ“œλ¦° κ²ƒμ²˜λŸΌ, μΉ˜λ£Œλ²•λ„ μ—†λŠ”λ° κΈ‰μ†νžˆ μ¦κ°€ν•˜κ³  있죠.
04:33
And fatty liver disease is just one example.
76
273750
2724
지방간 μ§ˆν™˜μ€ ν•˜λ‚˜μ˜ μ˜ˆμ— λΆˆκ³Όν•΄μš”.
04:36
Moving forward, we are going to tackle hundreds of other diseases
77
276498
4178
κ³„μ†ν•΄μ„œ μ €ν¬λŠ” μˆ˜λ§Žμ€ λ³‘μ˜ μΉ˜λ£Œλ²•μ„ μ°ΎκΈ° μœ„ν•΄
04:40
with no treatment.
78
280700
1493
λ…Έλ ₯ν•  κ²ƒμž…λ‹ˆλ‹€.
04:42
And by collecting more and more data about metabolites
79
282217
4554
λŒ€μ‚¬λ¬Όμ— λŒ€ν•œ 자료λ₯Ό 더 많이 μˆ˜μ§‘ν•˜κ³ 
04:46
and understanding how changes in metabolites
80
286795
3544
λŒ€μ‚¬λ¬Όμ˜ 변이가 μ–΄λ–»κ²Œ λ³‘μœΌλ‘œ λ°œμ „ν•˜λŠ”μ§€
04:50
leads to developing diseases,
81
290363
2402
μ•Œμ•„κ°€λŠ” 과정을 톡해
04:52
our algorithms will get smarter and smarter
82
292789
3489
저희 μ•Œκ³ λ¦¬μ¦˜μ€ 점점 더 λ˜‘λ˜‘ν•΄μ Έ
04:56
to discover the right therapeutics for the right patients.
83
296302
4196
ν™˜μžμ—κ²Œ λ§žλŠ” μ˜¬λ°”λ₯Έ μΉ˜λ£Œλ²•μ„ λ°œκ²¬ν•˜κ²Œ λ κ±°μ—μš”.
05:00
And we will get closer to reach our vision
84
300522
3770
μ½”λ“œ ν•œ 쀄을 μ“Έ λ•Œ λ§ˆλ‹€
05:04
of saving lives with every line of code.
85
304316
3863
생λͺ…을 κ΅¬ν•œλ‹€λŠ” 비전에 μ’€ 더 λ‹€κ°€κ°ˆ κ²ƒμž…λ‹ˆλ‹€.
05:08
Thank you.
86
308203
1321
κ°μ‚¬ν•©λ‹ˆλ‹€.
05:09
(Applause)
87
309548
3827
(λ°•μˆ˜)
이 μ›Ήμ‚¬μ΄νŠΈ 정보

이 μ‚¬μ΄νŠΈλŠ” μ˜μ–΄ ν•™μŠ΅μ— μœ μš©ν•œ YouTube λ™μ˜μƒμ„ μ†Œκ°œν•©λ‹ˆλ‹€. μ „ 세계 졜고의 μ„ μƒλ‹˜λ“€μ΄ κ°€λ₯΄μΉ˜λŠ” μ˜μ–΄ μˆ˜μ—…μ„ 보게 될 κ²ƒμž…λ‹ˆλ‹€. 각 λ™μ˜μƒ νŽ˜μ΄μ§€μ— ν‘œμ‹œλ˜λŠ” μ˜μ–΄ μžλ§‰μ„ 더블 ν΄λ¦­ν•˜λ©΄ κ·Έκ³³μ—μ„œ λ™μ˜μƒμ΄ μž¬μƒλ©λ‹ˆλ‹€. λΉ„λ””μ˜€ μž¬μƒμ— 맞좰 μžλ§‰μ΄ μŠ€ν¬λ‘€λ©λ‹ˆλ‹€. μ˜κ²¬μ΄λ‚˜ μš”μ²­μ΄ μžˆλŠ” 경우 이 문의 양식을 μ‚¬μš©ν•˜μ—¬ λ¬Έμ˜ν•˜μ‹­μ‹œμ˜€.

https://forms.gle/WvT1wiN1qDtmnspy7