The medical potential of AI and metabolites | Leila Pirhaji

譯者: Lo Hsien Huang 審譯者: Yanyan Hong

In 2003,

在 2003 年，

when we sequenced the human genome,

當我們為人類的基因組定序時，

we thought we would have the answer to treat many diseases.

我們以為會找到許多疾病的治療方法。

But the reality is far from that,

但實際情形卻遠非如此，

because in addition to our genes,

因為除了我們的基因之外，

our environment and lifestyle could have a significant role

生活環境和生活作息

也是引發重大疾病的關鍵因素。

in developing many major diseases.

One example is fatty liver disease,

以脂肪肝疾病爲例，

which is affecting over 20 percent of the population globally,

全球超過 20% 的人口 受此疾病影響，

and it has no treatment and leads to liver cancer

目前沒有任何治療方法 而且最後可發展為肝癌，

or liver failure.

或是肝臟衰竭。

So sequencing DNA alone doesn't give us enough information

所以只靠基因定序 並不能給我們足夠的訊息，

to find effective therapeutics.

找出有效的治療方法。

On the bright side, there are many other molecules in our body.

好消息是，我們身體裡 還有許多其他的分子，

In fact, there are over 100,000 metabolites.

事實上，我們身體 有超過十萬的代謝物。

Metabolites are any molecule that is supersmall in their size.

代謝物是體積超級小的分子，

Known examples are glucose, fructose, fats, cholesterol --

已知的例子包括， 葡萄糖、果糖、脂肪、膽固醇——

things we hear all the time.

我們時常聽到的這些東西。

Metabolites are involved in our metabolism.

代謝物會參與新陳代謝活動，

They are also downstream of DNA,

它們也是 DNA 的後段，

so they carry information from both our genes as well as lifestyle.

所以它們帶著基因訊息 也透露出我們的生活作息。

Understanding metabolites is essential to find treatments for many diseases.

要找出許多疾病的治療方法 就有必要瞭解代謝物，

I've always wanted to treat patients.

我一直都想要醫治好病人，

Despite that, 15 years ago, I left medical school,

但是十五年前，

因爲傾心於數學而離開了醫學院。

as I missed mathematics.

Soon after, I found the coolest thing:

不久，我發現最酷的事情是：

I can use mathematics to study medicine.

我可以用數學來研究醫學，

Since then, I've been developing algorithms to analyze biological data.

從那時起，我就一直開發 演算法用來分析生物數據。

So, it sounded easy:

這聽起來很簡單：

let's collect data from all the metabolites in our body,

我們收集身體中所有代謝物的數據，

develop mathematical models to describe how they are changed in a disease

然後開發數學模型來描述 它們在疾病中如何變化，

and intervene in those changes to treat them.

並且干預這些變化來進行治療。

Then I realized why no one has done this before:

然後，我明白為什麼之前 沒有人做過這件事了：

it's extremely difficult.

因為這實在太困難了。

(Laughter)

（笑聲）

There are many metabolites in our body.

我們身體中有太多代謝物了，

Each one is different from the other one.

每一個都不盡相同。

For some metabolites, we can measure their molecular mass

針對一些代謝物，

我們能夠用質譜儀 來測量它們的分子量。

using mass spectrometry instruments.

But because there could be, like, 10 molecules with the exact same mass,

但是具有完全相同的 分子量可能有十種之多，

we don't know exactly what they are,

所以無法知道它們確切是什麼東西，

and if you want to clearly identify all of them,

假如要清楚辨識所有代謝物，

you have to do more experiments, which could take decades

必須要做更多的實驗， 那有可能要花上數十年的時間，

and billions of dollars.

還要耗費幾十億美元。

So we developed an artificial intelligence, or AI, platform, to do that.

因此，我們開發了一種 人工智慧來做這事。

We leveraged the growth of biological data

我們利用生物數據的增長，

and built a database of any existing information about metabolites

然後建立一個資料庫 裡面有代謝物的相關訊息，

and their interactions with other molecules.

包含代謝物與其他分子 相互作用的訊息，

We combined all this data as a meganetwork.

我們把所有數據組合成一個巨大網絡。

Then, from tissues or blood of patients,

接著，從患者的器官組織或是血液中，

we measure masses of metabolites

我們測量到代謝物的分子量，

and find the masses that are changed in a disease.

並且尋找因疾病 而產生變化的代謝物質量。

But, as I mentioned earlier, we don't know exactly what they are.

但是，正如我稍早提過， 我們無法確切知道它們是什麼，

A molecular mass of 180 could be either the glucose, galactose or fructose.

分子量為 180 可能是葡萄糖， 不然就是半乳糖或是果糖，

They all have the exact same mass

它們都擁有相同的質量，

but different functions in our body.

但在身體中有著不同的功能。

Our AI algorithm considered all these ambiguities.

我們的人工智慧演算考慮到 這些含糊不清的情形，

It then mined that meganetwork

它會在巨大網絡中挖掘數據，

to find how those metabolic masses are connected to each other

找出那些代謝物如何相互連結，

that result in disease.

才會導致疾病的發生。

And because of the way they are connected,

而且因為它們連接的方式，

then we are able to infer what each metabolite mass is,

我們得以推斷出 每個代謝物的分子量是多少。

like that 180 could be glucose here,

在這裡，分子量 180 的可能是葡萄糖。

and, more importantly, to discover

而且更重要的是，

how changes in glucose and other metabolites

發現葡萄糖和其他代謝物的變化

lead to a disease.

如何引發疾病。

This novel understanding of disease mechanisms

這種針對疾病機制的新穎理解，

then enable us to discover effective therapeutics to target that.

讓我們能夠針對疾病 找出有效的治療方法。

So we formed a start-up company to bring this technology to the market

所以我們成立了一家新創公司 將這項技術帶入市場，

and impact people's lives.

對大家的生活帶來正面影響，

Now my team and I at ReviveMed are working to discover

現在我和團隊 在 ReviveMed 生技公司

正利用代謝物 努力尋找重大疾病的療法，

therapeutics for major diseases that metabolites are key drivers for,

like fatty liver disease,

像是脂肪肝疾病，

because it is caused by accumulation of fats,

這是由於脂肪的堆積引起。

which are types of metabolites in the liver.

而脂肪是肝臟中 不同類型的代謝物組成，

As I mentioned earlier, it's a huge epidemic with no treatment.

我稍早提到這種重大疾病 目前沒有任何治療方式，

And fatty liver disease is just one example.

脂肪肝疾病只是其中一個例子，

Moving forward, we are going to tackle hundreds of other diseases

我們接著要解決

其他數百種目前尚無治療方式的疾病。

with no treatment.

And by collecting more and more data about metabolites

藉著搜集更多的代謝物數據，

and understanding how changes in metabolites

並且瞭解這些代謝物的變化

leads to developing diseases,

如何引發疾病。

our algorithms will get smarter and smarter

我們的演算法會變得愈來愈聰明，

to discover the right therapeutics for the right patients.

幫助病患找出正確的治療方法。

And we will get closer to reach our vision

而且我們能夠利用每條基因碼

of saving lives with every line of code.

一步步達成拯救生命的願景。

Thank you.

謝謝大家。

(Applause)

（掌聲）