The medical potential of AI and metabolites | Leila Pirhaji

翻译人员: Carol Wang 校对人员: Jiasi Hao

In 2003,

在 2003 年，

when we sequenced the human genome,

当我们测序人类基因组时，

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

我们以为会找到 治疗多种疾病的答案。

But the reality is far from that,

但是实际情况远非如此，

because in addition to our genes,

因为除了我们的基因，

our environment and lifestyle could have a significant role

我们的生存环境和生活方式

也可能导致多种重大疾病。

in developing many major diseases.

One example is fatty liver disease,

例如，影响全球

which is affecting over 20 percent of the population globally,

超过 20％ 人口的脂肪肝，

and it has no treatment and leads to liver cancer

没有任何有效的治疗方法， 而且最终可发展为肝癌

or liver failure.

或肝衰竭。

So sequencing DNA alone doesn't give us enough information

因此，单纯的 DNA 测序 无法提供足够信息，

帮助我们寻找有效治疗方法。

to find effective therapeutics.

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

好消息是，我们体内 还有许多其他分子。

In fact, there are over 100,000 metabolites.

实际上，有超过 10 万多种代谢物。

Metabolites are any molecule that is supersmall in their size.

代谢物是体积很小的分子，

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

像我们常听说的：

葡萄糖、果糖、脂肪、胆固醇等。

things we hear all the time.

Metabolites are involved in our metabolism.

代谢物参与我们的新陈代谢，

They are also downstream of DNA,

它们在 DNA 的下游，

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

因此，它们携带着来自基因 和我们生活方式的信息。

Understanding metabolites is essential to find treatments for many diseases.

了解代谢物对寻找许多疾病的 治疗方法至关重要。

I've always wanted to treat patients.

我一直想治病救人，

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

但尽管如此，在十五年前， 我因为喜欢数学

as I missed mathematics.

而离开了医学院。

Soon after, I found the coolest thing:

不久之后，我发现了最酷的东西：

I can use mathematics to study medicine.

我可以使用数学来研究医学。

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

从那时起，我一直在开发 用于分析生物学数据的算法。

So, it sounded easy:

这听起来很简单：

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

我们先收集体内所有代谢物，

然后，开发数学模型 描述疾病中的代谢物变化，

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

and intervene in those changes to treat them.

并通过干预这些变化来进行治疗。

Then I realized why no one has done this before:

然后，我终于明白 以前为何没人做这件事了：

it's extremely difficult.

这真是太困难了。

(Laughter)

（笑声）

There are many metabolites in our body.

我们体内有许多代谢产物，

Each one is different from the other one.

种类繁多。

For some metabolites, we can measure their molecular mass

对于某些代谢物，我们可以使用

质谱仪来检测其分子量。

using mass spectrometry instruments.

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

而质量完全相同的分子 可能有 10 种之多，

we don't know exactly what they are,

我们分不清谁是谁，

and if you want to clearly identify all of them,

如果想识别所有这些分子，

you have to do more experiments, which could take decades

则必须进行更多实验， 这可能需要几十年、

and billions of dollars.

数十亿美元。

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

为了做这件事，我们开发了 人工智能（AI）平台，

We leveraged the growth of biological data

我们利用生物数据的增长，

and built a database of any existing information about metabolites

建立了一个数据库， 包含代谢物现有信息

and their interactions with other molecules.

及与其它分子的相互作用的数据。

We combined all this data as a meganetwork.

我们将所有这些数据 组合成了一个大型网络，

Then, from tissues or blood of patients,

然后，在患者的组织或血液中，

we measure masses of metabolites

测量代谢物的质量，

and find the masses that are changed in a disease.

并寻找因疾病 而产生变化的代谢物的质量。

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

但是，正如我之前提到的， 我们并不知道是什么代谢物。

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

分子量为 180 的代谢物 可以是葡萄糖、半乳糖或果糖，

They all have the exact same mass

在我们体内，它们的质量完全相同，

but different functions in our body.

但功能不同。

Our AI algorithm considered all these ambiguities.

我们的 AI 算法 考虑了所有这些可能。

It then mined that meganetwork

然后，会挖掘那个巨型网络的数据，

to find how those metabolic masses are connected to each other

以发现那些代谢物

如何相互关联而导致疾病。

that result in disease.

And because of the way they are connected,

根据它们的关联方式，

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

我们就能推断出 每个代谢物的质量，

like that 180 could be glucose here,

如 180 分子量的 可能是葡萄糖，

and, more importantly, to discover

更重要的是，

发现葡萄糖和其他代谢物的变化

how changes in glucose and other metabolites

lead to a disease.

如何导致疾病。

This novel understanding of disease mechanisms

对疾病机制的这种新颖理解，

then enable us to discover effective therapeutics to target that.

使我们能够发现 针对该疾病的有效疗法。

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

凭借该技术，我们成立了 一家初创公司，

将该技术推向市场， 进而影响人们的生活。

and impact people's lives.

Now my team and I at ReviveMed are working to discover

现在，我们的 ReviveMed 团队

正努力寻找主要代谢疾病的疗法，

therapeutics for major diseases that metabolites are key drivers for,

like fatty liver disease,

例如脂肪肝，

because it is caused by accumulation of fats,

因为它由脂肪堆积造成，

which are types of metabolites in the liver.

而脂肪是肝脏中的代谢物。

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

如前所述，这种大型流行病 尚无有效疗法。

And fatty liver disease is just one example.

脂肪肝只是其中一个例子，

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

展望未来，我们将研究其它几百种

with no treatment.

尚无有效疗法的疾病。

And by collecting more and more data about metabolites

通过收集更多代谢物的数据，

and understanding how changes in metabolites

了解代谢物的变化

leads to developing diseases,

如何导致疾病发展，

our algorithms will get smarter and smarter

我们的算法会逐步完善，

to discover the right therapeutics for the right patients.

为某些患者找到合适的疗法。

And we will get closer to reach our vision

而且，我们将更加接近我们的愿景：

of saving lives with every line of code.

用程序代码拯救生命。

Thank you.

谢谢。

(Applause)

（掌声）