that is probably the biggest dream of humanity:

接下來的16分鐘， 我要帶各位進行一段冒險之旅，

to understand the code of life.

這大概是人類最大的夢想：

So for me, everything started many, many years ago

了解生命的密碼。

when I met the first 3D printer.

對我而言，這一切的開始， 要拉回到好幾好幾年前，

The concept was fascinating.

當我第一次遇上3D印表機時。

A 3D printer needs three elements:

它的概念真的很棒。

a bit of information, some raw material, some energy,

3D印表機需要三個元素：

and it can produce any object that was not there before.

少量的資訊、一些原物料、 再加上點能量，

I was doing physics, I was coming back home

這樣它就可以製造出 以前從未存在過的任何東西。

and I realized that I actually always knew a 3D printer.

我當時研究的是物理學， 有天回到家裡時，

And everyone does.

我突然意識到，我家裡 就有一台 3D 印表機。

It was my mom.

而且每個人家裡都有一台。

(Laughter)

那就是我媽嗎。

My mom takes three elements:

（笑聲）

a bit of information, which is between my father and my mom in this case,

我媽也有三個元素：

raw elements and energy in the same media, that is food,

少量的資訊：我這個例子， 指的是我媽跟我爸之間的投入，

and after several months, produces me.

食物就是原物料及能量的來源，

And I was not existent before.

然後，幾個月後，生下了我。

So apart from the shock of my mom discovering that she was a 3D printer,

而我以前也是不存的。

I immediately got mesmerized by that piece,

所以，除了我發現我媽 就是一台3D列印機之外，

the first one, the information.

我突然間也被 這個吸引注了，

What amount of information does it take

那邊的第一項，資訊。

to build and assemble a human?

要有多少這樣的資訊

Is it much? Is it little?

才能建構並組裝出一個人來呢？

How many thumb drives can you fill?

要很多嗎？還是只要一點點？

Well, I was studying physics at the beginning

要多少隨身碟存取這些資訊呢？

and I took this approximation of a human as a gigantic Lego piece.

我一開始是研究物理學的，

So, imagine that the building blocks are little atoms

我喜歡把人類比喻成 一個大型的樂高玩具，

and there is a hydrogen here, a carbon here, a nitrogen here.

你可以想像，每一個 樂高積木就是一個原子，

So in the first approximation,

氫原子在這，碳原子在這， 氮原子在這。

if I can list the number of atoms that compose a human being,

按照最初的估算想法，

I can build it.

如果我可以列出 人類的原子清單的數量，

Now, you can run some numbers

我就可以把它建造出來。

and that happens to be quite an astonishing number.

現在，請各位算一下，

So the number of atoms,

這想必是個驚人的數字。

the file that I will save in my thumb drive to assemble a little baby,

所以，存在這隨身碟裡面

will actually fill an entire Titanic of thumb drives --

可以組合出來一個小寶寶的檔案， 裡面的原子數數量，

multiplied 2,000 times.

實際上若用樂高玩具 組裝起一個人類，

This is the miracle of life.

它的大小足足有 2000台鐵達尼號這麼大。

Every time you see from now on a pregnant lady,

這就是生命的奇蹟啊！

she's assembling the biggest amount of information

從現在起，你每次 看到懷孕的婦女，

that you will ever encounter.

她就是那個正在組裝

Forget big data, forget anything you heard of.

你這輩子所遇到的最大量資訊。

This is the biggest amount of information that exists.

忘了大數據吧！ 忘了你曾聽過的。

(Applause)

這就是現存的 最大數據資料。

But nature, fortunately, is much smarter than a young physicist,

（笑聲）

and in four billion years, managed to pack this information

但...好在大自然比一位 年輕的物理學家還聰明，

in a small crystal we call DNA.

這40億年來，大自然中 負責管理包裹這個資訊的

We met it for the first time in 1950 when Rosalind Franklin,

小晶體--我們稱之為DNA。

an amazing scientist, a woman,

我們在1950年第一次認識了它，

took a picture of it.

當時有一位了不起的女科學家 --羅莎琳．富蘭克林--

But it took us more than 40 years to finally poke inside a human cell,

給 DNA 拍了張照。

take out this crystal,

但我們花了40年的時間， 最後才戳進人類細胞裡

unroll it, and read it for the first time.

取出這個晶體，

The code comes out to be a fairly simple alphabet,

才首次把它伸展開來閱讀。

four letters: A, T, C and G.

而密碼也就是大家所孰知的

And to build a human, you need three billion of them.

四個字母：A、T、C、G。

Three billion.

而建造一個人類， 你需要30億個字母。

How many are three billion?

30億。

It doesn't really make any sense as a number, right?

30億有多少？

So I was thinking how I could explain myself better

我們對這個數字 真的很沒有概念，對吧？

about how big and enormous this code is.

所以，我在想，這麼大的數字

But there is -- I mean, I'm going to have some help,

我要怎麼解釋 才讓人比較容易了解。

and the best person to help me introduce the code

但，我的意思是... 我最好找個人來幫忙，

is actually the first man to sequence it, Dr. Craig Venter.

而能幫我介紹基因密碼 的最佳人選，

So welcome onstage, Dr. Craig Venter.

想當然就是第一個定序的人， 克萊格．凡特博士。

(Applause)

所以，讓我們歡迎 克萊格．凡特博士上台。

Not the man in the flesh,

（掌聲）

but for the first time in history,

當然不是活生生的人，

this is the genome of a specific human,

但這是史上第一次

printed page-by-page, letter-by-letter:

特定人類的基因組被

262,000 pages of information,

一頁接著一頁，一個字 接著一個字地列印出來：

450 kilograms, shipped from the United States to Canada

262,000頁的資料，

thanks to Bruno Bowden, Lulu.com, a start-up, did everything.

450公斤、從美國運到加拿大，

It was an amazing feat.

感謝新創公司Lulu.com的布魯諾．鮑登， 他們幫我做的這一切。

But this is the visual perception of what is the code of life.

這是個很棒的饗宴。

And now, for the first time, I can do something fun.

但這只是對生命密碼 的視覺感受。

I can actually poke inside it and read.

現在，為了慶祝第一次， 我要做件有趣的事。

So let me take an interesting book ... like this one.

我真的可以從裡面 挑一段來讀一讀。

I have an annotation; it's a fairly big book.

所以，讓我來找一本有趣的.... 書兒，比如這本。

So just to let you see what is the code of life.

我做了個註記；這書太厚了。

Thousands and thousands and thousands

讓各位看一下甚麼是生命密碼。

and millions of letters.

數以百萬、千萬、

And they apparently make sense.

億個字母。

Let's get to a specific part.

它們當然都有意義。

Let me read it to you:

讓我來找一段特別的

(Laughter)

讀給各位聽：

"AAG, AAT, ATA."

（笑聲）

To you it sounds like mute letters,

"AAG, AAT, ATA."

but this sequence gives the color of the eyes to Craig.

你們可能覺得像是在聽天書，

I'll show you another part of the book.

但這段序列，決定了 克萊格的眼睛顏色。

This is actually a little more complicated.

我再展示另一段給各位看。

Chromosome 14, book 132:

這段實際上稍微複雜些。

(Laughter)

14 號染色體，第132 號書：

As you might expect.

（笑聲）

(Laughter)

如你所望！

"ATT, CTT, GATT."

（笑聲）

This human is lucky,

"ATT, CTT, GATT."

because if you miss just two letters in this position --

這個人很幸運，

two letters of our three billion --

因為如果你在這個位置 剛好漏掉兩個字母--

he will be condemned to a terrible disease:

30億個字母，只漏掉兩個--

cystic fibrosis.

你就等同於被宣判 得了一個恐佈的疾病：

We have no cure for it, we don't know how to solve it,

囊性纖維化。

and it's just two letters of difference from what we are.

目前我們沒有治療的方式， 我們不知道如何解決，

A wonderful book, a mighty book,

僅僅就這兩個字母上 的差異而已。

a mighty book that helped me understand

這本偉大的書，

and show you something quite remarkable.

這本偉大的書，

Every one of you -- what makes me, me and you, you --

可以幫助我了解，也能讓各位 看到一些嘆為觀止的事情。

is just about five million of these,

在場的每一個人， 成就你我不同的地方

half a book.

就這五百萬個 字母的差異，

For the rest,

半本書。

we are all absolutely identical.

剩下的，

Five hundred pages is the miracle of life that you are.

我們絕對都長一樣。

The rest, we all share it.

就是這 500 頁的字母， 行塑了你是甚麼樣的人，

So think about that again when we think that we are different.

剩下的，我們都一樣。

This is the amount that we share.

所以，當我們在討論彼此差異的時候， 讓我們再反思一下，

So now that I have your attention,

其實我們共同的地方 真的有這麼多。

the next question is:

所以，我問一下各位，

How do I read it?

接下來的問題：

How do I make sense out of it?

我要怎麼讀它？

Well, for however good you can be at assembling Swedish furniture,

我要怎麼搞懂它？

this instruction manual is nothing you can crack in your life.

其實，無論你多麼會 看說明書組裝瑞典的家具，

(Laughter)

這本安裝手冊也沒辦法 教你如何破解你的人生。

And so, in 2014, two famous TEDsters,

（笑聲）

Peter Diamandis and Craig Venter himself,

2014年，兩位出名的 TED 演講者，

decided to assemble a new company.

彼得．戴曼迪斯和 克雷格．文特爾本人，

Human Longevity was born,

他們決定創立一家新公司。

with one mission:

《人類長壽公司》誕生了，

trying everything we can try

並賦予一個使命：

and learning everything we can learn from these books,

竭盡所能的，

with one target --

從這些書上，嘗試每樣東西， 學習每樣東西，

making real the dream of personalized medicine,

就為了一個目標——

understanding what things should be done to have better health

讓個人化醫療的美夢可以成真，

and what are the secrets in these books.

了解需要做哪些事 才能更健康，

An amazing team, 40 data scientists and many, many more people,

以及了解這些書 裡面的秘密。

a pleasure to work with.

一個令人驚豔的團隊，40 個數據科學家， 還有其他很多、很多的人，

The concept is actually very simple.

一起為團隊努力。

We're going to use a technology called machine learning.

這概念其實很簡單。

On one side, we have genomes -- thousands of them.

我們將要使用一種叫 「機械自主學習」的概念。

On the other side, we collected the biggest database of human beings:

一方面，我們有 成千上萬的基因組——

phenotypes, 3D scan, NMR -- everything you can think of.

另一方面，我們收集了 人類最大的資料庫：

Inside there, on these two opposite sides,

生物特性、3D掃描、核磁共振—— 你能想到的每樣東西。

there is the secret of translation.

這兩方面的資料， 被自主翻譯出來後

And in the middle, we build a machine.

就可以解開很多的祕密。

We build a machine and we train a machine --

在這兩個中間， 我們建立了一台機器。

well, not exactly one machine, many, many machines --

我建立它，訓練它——

to try to understand and translate the genome in a phenotype.

當然，並不只一台機器啦！ 是很多很多台機器——

What are those letters, and what do they do?

嘗試去了解並翻譯 基因組的生物特徵表象。

It's an approach that can be used for everything,

這些字母代表甚麼？ 它們有甚麼作用？

but using it in genomics is particularly complicated.

這個方法可以運用在每件事上，

Little by little we grew and we wanted to build different challenges.

但用在基因學上， 它就特別複雜。

We started from the beginning, from common traits.

在一點一滴的慢慢累積後， 我們想建立不一樣的挑戰。

Common traits are comfortable because they are common,

我們從共同的特徵開始。

everyone has them.

談共同特徵比較輕鬆， 因為它們都很普遍。

So we started to ask our questions:

每個人都有。

Can we predict height?

我們從這個問題開始問：

Can we read the books and predict your height?

我們可以預測身高嗎？

Well, we actually can,

我們可以光看書 就可以知道你的身高嗎？

with five centimeters of precision.

沒錯，我們真的可以，

BMI is fairly connected to your lifestyle,

預測的誤差在五公分內。

but we still can, we get in the ballpark, eight kilograms of precision.

身體質量指數與 你的生活形式有關，

Can we predict eye color?

但我們仍然可以，相當精準地 將預測誤差控制在 8 公斤以內。

Yeah, we can.

那我們可以預測眼睛顏色嗎？

Eighty percent accuracy.

是的，我們可以。

Can we predict skin color?

精準度高達80%。

Yeah we can, 80 percent accuracy.

我們可以預測皮膚顏色嗎？

Can we predict age?

是的，可以，80%的準確率。

We can, because apparently, the code changes during your life.

年齡呢？

It gets shorter, you lose pieces, it gets insertions.

可以，因為隨著年紀， 你的基因碼也會更著改變。

We read the signals, and we make a model.

它會變短、消失或被插入。

Now, an interesting challenge:

我們可以讀到那個訊號， 並把它模擬出來。

Can we predict a human face?

現在，有一項有趣的挑戰：

It's a little complicated,

我們可以預測一個人的臉嗎？

because a human face is scattered among millions of these letters.

這有點複雜，

And a human face is not a very well-defined object.

因為人臉上散播了 上百萬個這種字母。

So, we had to build an entire tier of it

而人臉不太容易預測。

to learn and teach a machine what a face is,

所以，我們必須建立一個 完整的堆疊系統，

and embed and compress it.

去學習並教會機器 人臉是甚麼，

And if you're comfortable with machine learning,

然後把它嵌進去並壓縮。

you understand what the challenge is here.

如果你很懂機器自主學習，

Now, after 15 years -- 15 years after we read the first sequence --

你會懂得這邊的挑戰是甚麼。

this October, we started to see some signals.

15年後--整整15年後-- 我們讀取到第一個序列--

And it was a very emotional moment.

今年10月，我們開始看到一些訊號。

What you see here is a subject coming in our lab.

真的是令人感動的時刻。

This is a face for us.

你現在看到的是一個 進來我們實驗室的實驗對象。

So we take the real face of a subject, we reduce the complexity,

這是一個我們人類的臉。

because not everything is in your face --

所以我們拿一個真實的臉當作實驗對象， 我們減少了複雜度，

lots of features and defects and asymmetries come from your life.

因為不是每樣東西都會在 你的臉上原貌呈現出來--

We symmetrize the face, and we run our algorithm.

有很多的特徵、缺陷及不對稱 來自於你後天的生活方式。

The results that I show you right now,

我們把臉對稱好後， 拿去跑我們的演算法。

this is the prediction we have from the blood.

我現在展示給各位看的結果，

(Applause)

是由血液演算出來的預測結果。

Wait a second.

（掌聲）

In these seconds, your eyes are watching, left and right, left and right,

稍等一下。

and your brain wants those pictures to be identical.

在這短短的幾秒鐘，你的眼睛會 左看看、右看看做比較，

So I ask you to do another exercise, to be honest.

而你的大腦會希望 這些照片是一致的。

Please search for the differences,

所以，我要求各位做另一項活動， 這次要誠實。

which are many.

請找出他們不一樣的地方，

The biggest amount of signal comes from gender,

有很多喔。

then there is age, BMI, the ethnicity component of a human.

最多的訊號來自性別，

And scaling up over that signal is much more complicated.

然後是年齡、身體質量指數、 人類種族族群。

But what you see here, even in the differences,

把這些訊號擴大是相當複雜的。

lets you understand that we are in the right ballpark,

但即使你現在看到有點不同，

that we are getting closer.

還是要讓各位知道， 我們預測還算不錯，

And it's already giving you some emotions.

已經很接近了。

This is another subject that comes in place,

這已經讓你有點激動了。

and this is a prediction.

這裡有另外一個例子，

A little smaller face, we didn't get the complete cranial structure,

這是預測的結果。

but still, it's in the ballpark.

有點小的臉，我們雖然沒有 跑完整個頭蓋骨結構，

This is a subject that comes in our lab,

但，還是很精準。

and this is the prediction.

這是另一個實驗對象，

So these people have never been seen in the training of the machine.

這是預測結果。

These are the so-called "held-out" set.

這些人從未在我們 訓練的機器裡面出現過。

But these are people that you will probably never believe.

也就是說這些從 外面隨機取樣的。

We're publishing everything in a scientific publication,

但也許各位不相信。

you can read it.

我們已經在科學期刊上 發表這一切了，

But since we are onstage, Chris challenged me.

你可以找到。

I probably exposed myself and tried to predict

但自從知道我們要上台後， 克里斯就挑戰我說，

someone that you might recognize.

我也許可以自己上陣

So, in this vial of blood -- and believe me, you have no idea

並嘗試預測你們可能認識的人。

what we had to do to have this blood now, here --

所以，在這一瓶血液裡面-- 相信我，你們絕對不知道

in this vial of blood is the amount of biological information

我們去哪裡搞來這一瓶血的，

that we need to do a full genome sequence.

這瓶血就擁有 全部的生物資訊，

We just need this amount.

夠我們跑完全部的基因組定序。

We ran this sequence, and I'm going to do it with you.

我們只需要這麼多。

And we start to layer up all the understanding we have.

我們已經把它拿去定序， 下次再做給大家看。

In the vial of blood, we predicted he's a male.

然後開始堆疊出 所有我們知道的東西，

And the subject is a male.

從這瓶血液裡， 我們預測出他是位男士。

We predict that he's a meter and 76 cm.

而實驗對象是男士。

The subject is a meter and 77 cm.

我們預測他身高176公分。

So, we predicted that he's 76; the subject is 82.

實際上他身高177公分。

We predict his age, 38.

我們預測他的體重是76公斤； 實際上是82公斤。

The subject is 35.

我們預測他的年齡是38歲。

We predict his eye color.

實際上是35歲。

Too dark.

我們預測眼睛的顏色是這樣。

We predict his skin color.

太暗了。

We are almost there.

我們預測他的皮膚顏色。

That's his face.

幾乎很接近了。

Now, the reveal moment:

這是他的臉。

the subject is this person.

現在，真相要大白的時刻了：