back to index

Manolis Kellis: Human Genome and Evolutionary Dynamics | Lex Fridman Podcast #113


small model | large model

link |
00:00:00.000
The following is a conversation with Manolis Kellis.
link |
00:00:03.040
He's a professor at MIT and head of the MIT Computational Biology Group.
link |
00:00:08.560
He's interested in understanding the human genome from a computational,
link |
00:00:12.400
evolutionary, biological, and other cross disciplinary perspectives.
link |
00:00:16.960
He has more big impactful papers and awards than I can list.
link |
00:00:21.360
But most importantly, he's a kind, curious, brilliant
link |
00:00:25.040
human being and just someone I really enjoy talking to.
link |
00:00:28.640
His passion for science and life in general is contagious.
link |
00:00:32.800
The hours honestly flew by and I'm sure we'll talk again on this podcast soon.
link |
00:00:37.600
Quick summary of the ads. Three sponsors.
link |
00:00:40.240
Blinkist, Aidsleep, and Masterclass.
link |
00:00:43.040
Please consider supporting this podcast by going to blinkist.com slash lex,
link |
00:00:47.680
aidsleep.com slash lex, and signing up at masterclass.com slash lex.
link |
00:00:53.200
Click the links, buy the stuff, get the discount.
link |
00:00:56.000
It's the best way to support this podcast.
link |
00:00:58.720
If you enjoy this thing, subscribe on YouTube,
link |
00:01:01.120
review it with five stars on Apple Podcasts, support it on Patreon,
link |
00:01:04.560
or connect with me on Twitter at Lex Freedman.
link |
00:01:08.080
As usual, I'll do a few minutes of ads now and never any ads in the middle
link |
00:01:11.760
that can break the flow of the conversation.
link |
00:01:14.400
This episode is supported by Blinkist, my favorite app for learning new things.
link |
00:01:19.760
Get it at blinkist.com slash lex for a seven day free trial
link |
00:01:23.920
and 25% off afterwards.
link |
00:01:26.800
Blinkist takes the key ideas from thousands of nonfiction books
link |
00:01:30.240
and condenses them down into just 15 minutes that you can read or listen to.
link |
00:01:35.200
I'm a big believer in reading at least an hour every day.
link |
00:01:38.800
As part of that, I use blinkist every day to try out a book
link |
00:01:43.040
I may otherwise never have a chance to read.
link |
00:01:45.360
And in general, it's a great way to broaden your view of the idea landscape out there
link |
00:01:50.000
and find books that you may want to read more deeply.
link |
00:01:53.360
With Blinkist, you get unlimited access to read or listen to a massive library
link |
00:01:58.560
of condensed nonfiction books.
link |
00:02:00.720
Go to blinkist.com slash lex to try it free for seven days
link |
00:02:05.200
and save 25% off your new subscription.
link |
00:02:08.240
That's blinkist.com slash lex.
link |
00:02:10.720
Blinkist spelled B L I N K I S T.
link |
00:02:15.920
This show is also sponsored by ASleep and its PodPro mattress
link |
00:02:19.680
that you can check out at ASleep.com slash lex to get $200 off.
link |
00:02:24.560
It controls temperature with an app.
link |
00:02:26.640
It can cool down to as low as 55 degrees on each side of the bed separately.
link |
00:02:31.440
Research shows that temperature has a big impact on the quality of our sleep.
link |
00:02:35.600
Anecdotally, it's been true for me.
link |
00:02:37.760
It's truly been a game changer.
link |
00:02:39.760
I love it.
link |
00:02:40.720
The PodPro is packed with sensors that track heart rate,
link |
00:02:43.360
heart rate variability and respiratory rate showing it all in their app.
link |
00:02:48.000
The app's health metrics are amazing, but the cooling alone is honestly worth the money.
link |
00:02:52.800
Check it out at ASleep.com slash lex to get $200 off.
link |
00:02:57.280
This show is also sponsored by masterclass.
link |
00:03:00.240
Sign up at masterclass.com slash lex to get a discount and to support this podcast.
link |
00:03:05.360
When I first heard about masterclass, I thought it was too good to be true.
link |
00:03:08.960
For 180 bucks a year, you get an all access pass to watch courses from
link |
00:03:13.360
to list some of my favorites, Chris Hadfield on space exploration,
link |
00:03:17.520
Neil deGrasse Tyson on scientific thinking and communication.
link |
00:03:20.880
Will Wright, one of my favorite game designers, Carlos Santana,
link |
00:03:24.400
one of my favorite guitar players, Gary Kasparov, of course,
link |
00:03:27.840
the greatest chess player of all time.
link |
00:03:29.520
I'm not biased.
link |
00:03:30.640
Daniel Negrano on poker and many more.
link |
00:03:33.600
Chris Hadfield explaining how rockets work and the experience of being
link |
00:03:36.800
launched into space alone is worth the money.
link |
00:03:39.920
By the way, you can watch it on basically any device.
link |
00:03:42.960
Once again, sign up at masterclass.com slash lex to get a discount
link |
00:03:47.200
and to support this podcast.
link |
00:03:49.520
And now here's my conversation with Manolis Kellis.
link |
00:03:54.480
What to you is the most beautiful aspect of the human genome?
link |
00:03:58.560
Don't get me started.
link |
00:04:01.920
So we got time.
link |
00:04:04.160
The first answer is that the beauty of genomes transcends humanity.
link |
00:04:07.520
So it's not just about the human genome.
link |
00:04:09.440
Genomes in general are amazingly beautiful.
link |
00:04:12.560
And again, I'm obviously biased.
link |
00:04:14.000
So in my view, the way that I like to introduce the human genome
link |
00:04:19.360
and the way that I like to introduce genomics to my class
link |
00:04:22.160
is by telling them, you know, we're not the inventors
link |
00:04:24.880
of the first digital computer.
link |
00:04:26.480
We are the descendants of the first digital computer.
link |
00:04:30.560
Basically, life is digital.
link |
00:04:32.240
And that's absolutely beautiful about life.
link |
00:04:34.480
The fact that at every replication step,
link |
00:04:37.040
you don't lose any information because that information is digital.
link |
00:04:40.000
If it was analog, if it was just protein concentrations,
link |
00:04:43.040
you'd lose it after a few generations.
link |
00:04:44.640
It would just dissolve away.
link |
00:04:46.240
And that's what the ancients didn't understand about inheritance.
link |
00:04:50.000
The first person to understand digital inheritance
link |
00:04:52.080
was Mendel, of course.
link |
00:04:54.400
And his theory, in fact, stayed in a bookshelf for like 50 years
link |
00:04:58.880
while Darwin was getting famous about natural selection.
link |
00:05:02.320
But the missing component was this digital inheritance,
link |
00:05:05.440
the mechanism of evolution that Mendel had discovered.
link |
00:05:09.600
So that aspect, in my view, is the most beautiful aspect,
link |
00:05:12.960
but it transcends all of life.
link |
00:05:14.720
And can you elaborate maybe the inheritance part?
link |
00:05:17.920
What was the key thing that the ancients didn't understand?
link |
00:05:22.400
So the very theory of inheritance as discrete units
link |
00:05:28.400
throughout the life of Mendel and well after his writing,
link |
00:05:32.720
people thought that his P experiments were just a little fluke,
link |
00:05:36.400
that they were just a little exception
link |
00:05:38.560
that would normally not even apply to humans.
link |
00:05:41.760
That basically what they saw is this continuum of eye color,
link |
00:05:48.000
this continuum of skin color, this continuum of hair color,
link |
00:05:50.960
this continuum of height.
link |
00:05:52.560
And all of these continuums did not fit
link |
00:05:54.960
with a discrete type of inheritance that Mendel was describing.
link |
00:05:58.720
But what's unique about genomics
link |
00:06:00.320
and what's unique about the genome
link |
00:06:01.520
is really that there are two copies
link |
00:06:03.760
and that you get a combination of these.
link |
00:06:05.680
But for every trait, there are dozens of contributing variables.
link |
00:06:10.000
And it was only Ronald Fisher in the 20th century
link |
00:06:13.360
that basically recognized that even five Mendelian traits
link |
00:06:19.520
would add up to a continuum like inheritance pattern.
link |
00:06:24.400
And he wrote a series of papers
link |
00:06:26.880
that still are very relevant today
link |
00:06:29.360
about this Mendelian inheritance of continuum like traits.
link |
00:06:34.400
And I think that that was the missing step in inheritance.
link |
00:06:37.760
So well before the discovery of the structure of DNA,
link |
00:06:40.800
which is again another amazingly beautiful aspect,
link |
00:06:43.600
the double helix, what I like to call
link |
00:06:45.280
the most noble molecule of our time,
link |
00:06:47.680
holds within it the secret of that discrete inheritance.
link |
00:06:53.360
But the conceptualization of discrete elements
link |
00:06:57.600
is something that precedes that.
link |
00:06:58.880
So even though it's discrete,
link |
00:07:00.400
when it materializes itself into actual traits that we see,
link |
00:07:05.840
it can be continuous and basically arbitrarily rich and complex.
link |
00:07:10.000
So if you have five genes that contribute to human height,
link |
00:07:14.240
and there aren't five, there's a thousand.
link |
00:07:16.160
If there's only five genes,
link |
00:07:17.680
and you inherit some combination of them,
link |
00:07:20.000
and everyone makes you two inches taller or two inches shorter,
link |
00:07:23.280
it'll look like a continuum trait, a continuous trait.
link |
00:07:26.640
But instead of five, there are thousands,
link |
00:07:29.360
and every one of them contributes to less than one millimeter.
link |
00:07:32.560
We change in height more during the day
link |
00:07:35.440
than each of these genetic variants contributes.
link |
00:07:38.640
So by the evening, you're shorter than you were, you walk up with.
link |
00:07:42.640
Isn't it weird then that we're not more different than we are?
link |
00:07:46.640
Why are we all so similar
link |
00:07:48.480
if there's so much possibility to be different?
link |
00:07:51.280
Yeah, so there are selective advantages
link |
00:07:54.640
to being medium.
link |
00:07:57.440
If you're extremely tall or extremely short,
link |
00:07:59.840
you run into selective disadvantages.
link |
00:08:02.160
So you have trouble breathing, you have trouble running,
link |
00:08:04.080
you have trouble sitting.
link |
00:08:04.960
If you're too tall, if you're too short,
link |
00:08:06.960
you might have other selective pressures acting against that.
link |
00:08:10.880
If you look at natural history of human population,
link |
00:08:13.520
there's actually selection for height in northern Europe
link |
00:08:16.880
and selection against height in southern Europe.
link |
00:08:19.840
So there might actually be advantages
link |
00:08:21.680
to actually being not super tall.
link |
00:08:24.800
And if you look across the entire human population,
link |
00:08:28.560
for many, many traits, there's a lot of push towards the middle.
link |
00:08:31.920
Balancing selection is the usual term for selection
link |
00:08:36.480
that seeks to not be extreme
link |
00:08:39.600
and to have a combination of alleles that keep recombining.
link |
00:08:45.840
And if you look at mate selection,
link |
00:08:48.480
super, super tall people will not tend to sort of marry
link |
00:08:51.760
super, super tall people.
link |
00:08:53.200
Very often you see these couples
link |
00:08:55.040
that are kind of compensating for each other.
link |
00:08:57.840
And the best predictor of the kid's age
link |
00:09:00.320
is very often just take the average of the two parents
link |
00:09:03.360
and then adjust for sex and boom, you get it.
link |
00:09:06.880
It's extremely heritable.
link |
00:09:08.400
Let me ask, you kind of took a step back to the genome
link |
00:09:12.640
outside of just humans,
link |
00:09:13.760
but is there something that you find beautiful
link |
00:09:15.760
about the human genome specifically?
link |
00:09:18.720
So I think the genome, if more people understood
link |
00:09:23.040
the beauty of the human genome,
link |
00:09:24.400
they would be so many fewer wars,
link |
00:09:26.560
so much less anger in the world.
link |
00:09:28.880
I mean, what's really beautiful about the human genome
link |
00:09:31.360
is really the variation
link |
00:09:33.840
that teaches us both about individuality
link |
00:09:36.400
and about similarity.
link |
00:09:38.160
So any two people on the planet are 99.9% identical.
link |
00:09:42.160
How can you fight with someone who's 99.9% identical to you?
link |
00:09:46.160
It's just counterintuitive.
link |
00:09:48.560
And yet any two siblings of the same parents
link |
00:09:52.400
differ in millions of locations.
link |
00:09:55.600
So every one of them is basically two to the million
link |
00:09:58.800
unique from any pair of parents,
link |
00:10:01.680
let alone any two random parents on the planet.
link |
00:10:04.160
So that's, I think, something that teaches us
link |
00:10:07.200
about sort of the nature of humanity in many ways
link |
00:10:09.840
that every one of us is as unique as any star
link |
00:10:13.200
and way more unique in actually many ways.
link |
00:10:16.000
And yet, we're all brothers and sisters.
link |
00:10:20.320
Yeah, just like stars, most of it is just fusion reactions.
link |
00:10:24.720
Yeah, you only have a few parameters to describe stars.
link |
00:10:27.040
Yeah, exactly.
link |
00:10:27.920
You know, mass size, initial size, and stage of life.
link |
00:10:31.600
Whereas for humans, thousands of parameters
link |
00:10:34.960
scattered across our genome.
link |
00:10:36.400
Scattered across our genome. So the other thing that makes humans unique,
link |
00:10:41.360
the other things that makes inheritance unique in humans is that
link |
00:10:47.200
most species inherit things vertically.
link |
00:10:50.320
Basically, instinct is a huge part of their behavior.
link |
00:10:54.480
The way that, you know, I mean, with my kids,
link |
00:10:57.440
we've been watching this nest of birds with two little eggs,
link |
00:11:02.160
you know, outside our window for the last few months,
link |
00:11:04.320
for the last few weeks as they've been growing.
link |
00:11:07.440
And there's so much behavior that's hard coded.
link |
00:11:12.240
Birds don't just learn as they grow.
link |
00:11:15.200
They don't, you know, there's no culture.
link |
00:11:17.120
Like a bird that's born in Boston will be the same
link |
00:11:20.080
as a bird that's born in California.
link |
00:11:22.000
So there's not as much inheritance of ideas, of customs.
link |
00:11:28.080
A lot of it is hard coding in their genome.
link |
00:11:30.400
What's really beautiful about the human genome is that
link |
00:11:32.800
if you take a person from today
link |
00:11:34.880
and you place them back in ancient Egypt,
link |
00:11:36.960
or if you take a person from ancient Egypt
link |
00:11:39.040
and you place them here today,
link |
00:11:41.040
they will grow up to be completely normal.
link |
00:11:44.960
That is not genetics.
link |
00:11:47.600
This is the other type of inheritance in humans.
link |
00:11:51.680
So on one hand, we have genetic inheritance,
link |
00:11:53.760
which is vertical from your parents down.
link |
00:11:56.000
On the other hand, we have horizontal inheritance,
link |
00:11:58.320
which is the ideas that are built up at every generation
link |
00:12:02.320
are horizontally transmitted.
link |
00:12:04.480
And the huge amount of time that we spend in educating ourselves,
link |
00:12:09.360
a concept known as neoteny,
link |
00:12:11.840
neo for newborn and then teni for holding.
link |
00:12:15.120
So if you look at humans,
link |
00:12:17.200
I mean the little birds that were, you know, eggs two weeks ago,
link |
00:12:20.240
and now one of them has already flown off.
link |
00:12:22.480
The other one's ready to fly off.
link |
00:12:24.400
In two weeks, they're ready to just fend for themselves.
link |
00:12:27.280
Humans, 16 years.
link |
00:12:29.440
18 years, 24, getting out of college.
link |
00:12:32.880
I'm still learning.
link |
00:12:34.320
So that's so fascinating, this picture of vertical and horizontal.
link |
00:12:38.640
When you talk about the horizontal is in the realm of ideas.
link |
00:12:41.920
Exactly.
link |
00:12:42.560
Okay.
link |
00:12:42.800
So it's the actual social interactions.
link |
00:12:45.120
That's exactly right.
link |
00:12:46.000
That's exactly right.
link |
00:12:46.960
So basically the concept of neoteny
link |
00:12:49.120
is that you spend acquiring characteristics from your environment
link |
00:12:54.080
in an extremely malleable state of your brain
link |
00:12:56.800
and the wiring of your brain
link |
00:12:58.400
for a long period of your life.
link |
00:13:00.560
Compared to primates, we are useless.
link |
00:13:03.360
You take any primate at seven weeks
link |
00:13:05.200
and in human at seven weeks, we lose the battle.
link |
00:13:08.240
But at 18 years, you know, all better off.
link |
00:13:11.520
Like we basically, our brain continues to develop
link |
00:13:14.720
in an extremely malleable form till very late.
link |
00:13:17.840
And this is what allows education.
link |
00:13:20.240
This is what allows the person from Egypt
link |
00:13:22.320
to do extremely well now.
link |
00:13:24.560
And the reason for that is that
link |
00:13:26.640
the wiring of our brain
link |
00:13:30.160
and the development of that wiring is actually delayed.
link |
00:13:32.880
So, you know, the longer you delay that,
link |
00:13:35.920
the more opportunity you have to pass on knowledge,
link |
00:13:39.360
to pass on concepts, ideals, ideas from the parents to the child.
link |
00:13:44.400
And what's really absolutely beautiful about humans today
link |
00:13:47.360
is that that lateral transfer of ideas and culture
link |
00:13:50.320
is not just from uncles and aunts and teachers at school,
link |
00:13:53.840
but it's from Wikipedia and review articles on the web
link |
00:13:58.320
and thousands of journals
link |
00:14:00.640
that are sort of putting out information for free
link |
00:14:03.600
and podcasts and video casts and all of that stuff,
link |
00:14:06.880
where you can basically learn about any topic
link |
00:14:10.640
pretty much everything that would be
link |
00:14:13.600
in any super advanced textbook in a matter of days,
link |
00:14:17.200
instead of having to go to the library of Alexandria
link |
00:14:21.040
and sail there to read three books and then sail for another few days
link |
00:14:24.240
to get to Athens and et cetera, et cetera, et cetera.
link |
00:14:26.720
So, the democratization of knowledge
link |
00:14:29.280
and the spread, the speed of spread of knowledge,
link |
00:14:32.320
is what defines, I think, the human inheritance pattern.
link |
00:14:36.800
So, you sound excited about it.
link |
00:14:39.440
Are you also a little bit afraid,
link |
00:14:41.760
or are you more excited by the power of this kind of
link |
00:14:45.360
distributed spread of information?
link |
00:14:47.920
So, you put a very kind,
link |
00:14:49.360
that most people are kind of using the internet
link |
00:14:53.360
in looking Wikipedia, reading articles, reading papers, and so on,
link |
00:14:58.080
but if we're honest, most people online,
link |
00:15:01.920
especially when they're younger,
link |
00:15:03.200
probably looking at five second clips on TikTok
link |
00:15:05.760
or whatever the new social network is,
link |
00:15:08.480
are you, given this power of horizontal inheritance,
link |
00:15:12.480
are you optimistic or a little bit pessimistic
link |
00:15:16.480
about this new effect of the internet
link |
00:15:22.400
and democratization of knowledge on our,
link |
00:15:26.880
what would you call this, this genome?
link |
00:15:28.640
Like, would you use the term genome, by the way, for this?
link |
00:15:31.200
Yeah, yeah.
link |
00:15:32.320
I think, you know, we use the genome to talk about DNA,
link |
00:15:36.080
but very often we say, you know, I'm Greek,
link |
00:15:38.800
so people ask me, hey, what's in the Greek genome?
link |
00:15:40.800
And I'm like, well, yeah, what's in the Greek genome
link |
00:15:42.720
is both our genes and also our ideas and our ideals
link |
00:15:45.680
and our culture, so.
link |
00:15:46.800
The poetic meaning of the word.
link |
00:15:48.160
Exactly, exactly, yeah.
link |
00:15:50.000
So I think that there's a beauty
link |
00:15:55.840
to the democratization of knowledge,
link |
00:15:57.600
the fact that you can reach as many people
link |
00:16:00.160
as, you know, any other person on the planet
link |
00:16:02.720
and it's not who you are,
link |
00:16:04.160
it's really your ideas that matter,
link |
00:16:06.640
is a beautiful aspect of the internet.
link |
00:16:09.680
The, I think there's, of course,
link |
00:16:13.200
a danger of my ignorance is as important as your expertise.
link |
00:16:18.160
The fact that with this democratization
link |
00:16:21.280
comes the abolishment of respecting expertise.
link |
00:16:25.040
Just because you've spent, you know, 10,000 hours of your life
link |
00:16:28.880
studying, I don't know, human brain circuitry,
link |
00:16:33.200
why should I trust you?
link |
00:16:34.000
I'm just going to make up my own theories
link |
00:16:35.520
and they'll be just as good as yours,
link |
00:16:37.120
is an attitude that sort of counteracts
link |
00:16:39.520
the beauty of the different democratization.
link |
00:16:41.760
And I think that within our educational system
link |
00:16:46.720
and within the upbringing of our children,
link |
00:16:48.960
we have to not only teach them knowledge,
link |
00:16:51.520
but we have to teach them the means to get to knowledge.
link |
00:16:54.560
And that, you know, it's very similar to sort of,
link |
00:16:57.440
you fish, you catch a fish for a man for one day,
link |
00:17:00.640
you fed them for one day, you teach them how to fish,
link |
00:17:03.040
you fed them for the rest of their life.
link |
00:17:04.640
So instead of just gathering the knowledge they need
link |
00:17:07.600
for any one task, we can just tell them,
link |
00:17:09.520
all right, here's how you Google it.
link |
00:17:12.400
Here's how to figure out what's real and what's not.
link |
00:17:14.480
Here's how you check the sources.
link |
00:17:16.320
Here's how you form a basic opinion for yourself.
link |
00:17:19.200
And I think that inquisitive nature is paramount
link |
00:17:24.640
to being able to sort through this huge wealth of knowledge.
link |
00:17:29.200
So you need a basic educational foundation
link |
00:17:32.480
based on which you can then add on
link |
00:17:35.440
the sort of domain specific knowledge.
link |
00:17:38.000
But that basic educational foundation should just,
link |
00:17:40.640
not just be knowledge, but it should also be epistemology,
link |
00:17:45.120
the way to acquire knowledge.
link |
00:17:47.120
I'm not sure any of us know how to do that in this modern day.
link |
00:17:50.480
We're actually learning.
link |
00:17:51.520
One of the big surprising things to me about the coronavirus,
link |
00:17:56.400
for example, is that Twitter has been
link |
00:17:59.600
one of the best sources of information.
link |
00:18:02.720
Basically, like building your own network of experts,
link |
00:18:05.520
of, as opposed to the traditional centralized expertise
link |
00:18:10.960
of the WHO and the CDC or maybe any one particular
link |
00:18:17.600
respectable person at the top of a department
link |
00:18:20.560
and some kind of institution.
link |
00:18:21.680
You instead look at 10, 20 hundreds of people,
link |
00:18:26.480
some of whom are young kids with just that are incredibly
link |
00:18:32.320
good at aggregating data and plotting
link |
00:18:34.160
and visualizing that data.
link |
00:18:35.680
That's been really surprising to me.
link |
00:18:37.120
I don't know what to make of it.
link |
00:18:39.760
I don't know how that matures into something stable.
link |
00:18:45.440
I don't know if you have ideas.
link |
00:18:46.400
Like what if you were to try to explain to your kids
link |
00:18:49.840
of how, where should you go to learn about coronavirus?
link |
00:18:55.840
What would you say?
link |
00:18:56.720
It's such a beautiful example.
link |
00:18:57.920
And I think the current pandemic and the speed at which
link |
00:19:01.520
the scientific community has moved
link |
00:19:03.040
in the current pandemic, I think exemplifies
link |
00:19:06.320
this horizontal transfer and the speed of horizontal
link |
00:19:08.960
transfer of information.
link |
00:19:10.720
The fact that the genome was first sequenced in early January.
link |
00:19:16.240
The first sample was obtained December 29, 2019,
link |
00:19:20.240
a week after the publication of the first genome sequence
link |
00:19:24.320
Moderna had already finalized its vaccine design
link |
00:19:27.520
and was moving to production.
link |
00:19:29.920
This is phenomenal.
link |
00:19:31.680
The fact that we go from not knowing what the heck
link |
00:19:35.200
is killing people in Wuhan to, wow, it's SARS CoV2
link |
00:19:40.160
and here's the set of genes, here's the genome,
link |
00:19:42.560
here's the sequence, here are the polymorphisms, etc.
link |
00:19:45.520
In the matter of weeks is phenomenal.
link |
00:19:48.080
In that incredible pace of transfer of knowledge,
link |
00:19:52.720
there have been many mistakes.
link |
00:19:54.320
So, you know, some of those mistakes may have
link |
00:19:56.720
been politically motivated.
link |
00:19:57.840
Our other mistakes may have just been innocuous errors.
link |
00:20:00.720
Others may have been misleading the public for the greater good,
link |
00:20:03.840
such as don't wear masks because we don't want the mask to run out.
link |
00:20:07.040
I mean, that was very silly in my view and a very big mistake.
link |
00:20:11.280
But the spread of knowledge from the scientific community
link |
00:20:16.240
was phenomenal and some people will point out to bogus articles
link |
00:20:20.560
that snuck in and made the front page.
link |
00:20:22.720
Yeah, they did, but within 24 hours they were debunked
link |
00:20:26.080
and went out of the front page.
link |
00:20:27.360
And I think that's the beauty of science today.
link |
00:20:30.000
The fact that it's not, oh, knowledge is fixed.
link |
00:20:33.040
It's the ability to embrace that nothing is permanent
link |
00:20:36.800
when it comes to knowledge,
link |
00:20:37.760
that everything is the current best hypothesis
link |
00:20:40.000
and the current best model that best fits the current data
link |
00:20:42.960
and the willingness to be wrong.
link |
00:20:45.680
The expectation that we're going to be wrong
link |
00:20:48.160
and the celebration of success based on
link |
00:20:50.560
how long was I not proven wrong for
link |
00:20:52.640
rather than, wow, I was exactly right
link |
00:20:55.440
because no one is going to be exactly right
link |
00:20:56.880
with partial knowledge, but the arc towards perfection,
link |
00:21:02.960
I think is so much more important than how far you are
link |
00:21:07.200
in your first step.
link |
00:21:08.480
And I think that's what sort of the current pandemic has taught us,
link |
00:21:13.200
the fact that, yeah, no, of course we're going to make mistakes,
link |
00:21:15.920
but at least we're going to learn from those mistakes
link |
00:21:18.240
and become better and learn better
link |
00:21:20.160
and spread information better.
link |
00:21:21.200
So if I were to answer the question of where would you go
link |
00:21:24.080
to learn about coronavirus, first textbook,
link |
00:21:28.720
it all starts with a textbook.
link |
00:21:29.840
Just open up a chapter on virology
link |
00:21:32.480
and how coronavirus is work.
link |
00:21:34.400
Then some basic epidemiology
link |
00:21:36.800
and sort of how pandemics have worked in the past.
link |
00:21:39.680
What are the basic principles surrounding
link |
00:21:41.840
these first wave, second wave?
link |
00:21:43.360
Why do they even exist?
link |
00:21:45.280
Then understanding about growth,
link |
00:21:47.120
understanding about the R0 and RT at various time points.
link |
00:21:52.560
And then understanding the means of spread,
link |
00:21:55.280
how it spreads from person to person,
link |
00:21:57.280
then how does it get into your cells?
link |
00:22:00.080
From when it gets into the cells,
link |
00:22:01.440
what are the paths that it takes?
link |
00:22:03.120
What are the cell types that express
link |
00:22:05.040
the particular ACE2 receptor?
link |
00:22:07.200
How is your immune system interacting with the virus?
link |
00:22:09.840
And once your immune system launches it defense,
link |
00:22:12.240
how is that helping or actually hurting your health?
link |
00:22:15.360
What about the cytokine storm?
link |
00:22:16.800
What are most people dying from?
link |
00:22:18.480
Why are the comorbidities and these risk factors even applying?
link |
00:22:23.840
What makes obese people respond more
link |
00:22:25.840
or elderly people respond more to the virus
link |
00:22:28.800
while kids are completely, very often,
link |
00:22:33.760
not even aware that they're spreading it?
link |
00:22:36.320
So I think there's some basic questions
link |
00:22:41.040
that you would start from.
link |
00:22:42.640
And then I'm sorry to say,
link |
00:22:44.240
but Wikipedia is pretty awesome.
link |
00:22:45.760
Google is pretty awesome.
link |
00:22:46.720
It used to be a time maybe five years ago.
link |
00:22:50.400
I forget when,
link |
00:22:51.920
but people kind of made fun of Wikipedia
link |
00:22:54.160
for being an unreliable source.
link |
00:22:57.120
I never quite understood it.
link |
00:22:58.400
I thought from the early days it was pretty reliable.
link |
00:23:01.040
They're better than a lot of the alternatives.
link |
00:23:03.520
But at this point,
link |
00:23:04.640
it's kind of like a solid accessible survey paper
link |
00:23:08.240
on every subject ever.
link |
00:23:11.200
There's an ascertainment bias and a writing bias.
link |
00:23:14.560
So I think this is related to people saying,
link |
00:23:17.680
oh, so many nature papers are wrong.
link |
00:23:20.640
And they're like, why would you publish in nature?
link |
00:23:22.400
So many nature papers are wrong.
link |
00:23:23.600
And my answer is, no, no, no.
link |
00:23:26.000
So many nature papers are scrutinized.
link |
00:23:29.360
And just because more of them are being proven wrong
link |
00:23:31.840
than in other articles is actually evidence
link |
00:23:35.200
that they're actually better paper overall
link |
00:23:36.960
because they're being scrutinized at a rate
link |
00:23:39.200
much higher than any other journal.
link |
00:23:40.960
So if you basically judge Wikipedia
link |
00:23:45.600
by not the initial content,
link |
00:23:49.840
but by the number of revisions,
link |
00:23:52.320
then of course it's going to be
link |
00:23:53.360
the best source of knowledge eventually.
link |
00:23:55.200
It's still very superficial.
link |
00:23:56.960
You then have to go into the review papers,
link |
00:23:58.640
et cetera, et cetera, et cetera.
link |
00:24:00.160
But I mean, for most scientific topics,
link |
00:24:03.280
it's extremely superficial.
link |
00:24:05.040
But it is quite authoritative
link |
00:24:07.680
because it is the place
link |
00:24:09.120
that everybody likes to criticize as being wrong.
link |
00:24:11.600
You say that it's superficial.
link |
00:24:13.520
And a lot of topics that I've studied a lot of,
link |
00:24:18.240
I find it, I don't know if superficial is the right word
link |
00:24:24.160
because superficial kind of implies
link |
00:24:25.600
that it's not correct.
link |
00:24:27.680
No, no, I don't mean any implication
link |
00:24:30.320
of it not being correct.
link |
00:24:31.520
It's just superficial.
link |
00:24:32.720
It's basically only scratching the surface.
link |
00:24:35.440
For depth, you don't go to Wikipedia
link |
00:24:36.960
and you go to the review articles.
link |
00:24:38.240
But it can be profound in the way that articles rarely,
link |
00:24:41.680
one of the frustrating things to me about
link |
00:24:44.640
like certain computer science,
link |
00:24:46.400
like in the machine learning world,
link |
00:24:48.320
articles, they don't as often take the bigger picture view.
link |
00:24:54.720
There's a kind of data set and you show that it works
link |
00:24:57.280
and you kind of show that here's an architectural thing
link |
00:24:59.520
that creates an improvement and so on and so forth.
link |
00:25:02.160
But you don't say, well, what does this mean
link |
00:25:05.200
for the nature of intelligence,
link |
00:25:07.200
for future data sets we haven't even thought about
link |
00:25:09.920
or if you were trying to implement this,
link |
00:25:11.760
like if we took this data set of 100,000 examples
link |
00:25:15.760
and scale it to 100 billion examples with this method,
link |
00:25:19.120
like look at the bigger picture,
link |
00:25:21.040
which is what a Wikipedia article would actually try to do,
link |
00:25:25.360
which is like, what does this mean
link |
00:25:27.440
in the context of computer,
link |
00:25:29.920
the broad field of computer vision or something like that?
link |
00:25:32.240
Yeah, yeah, and no, I agree with you completely,
link |
00:25:34.640
but it depends on the topic.
link |
00:25:35.920
I mean, for some topics, there's been a huge amount of work.
link |
00:25:38.320
For other topics, it's just a stub.
link |
00:25:40.240
So, you know.
link |
00:25:41.440
I got it.
link |
00:25:41.920
Yeah.
link |
00:25:42.400
Well, yeah, actually, which we'll talk on, genomics was not great.
link |
00:25:47.920
Yeah, it's very shallow.
link |
00:25:49.200
Yeah, it's not wrong, it's just shallow.
link |
00:25:51.680
It's shallow.
link |
00:25:52.240
Yeah.
link |
00:25:52.880
Every time I criticize something,
link |
00:25:54.560
I should feel partly responsible.
link |
00:25:56.160
Basically, if more people from my community went there
link |
00:25:58.640
and edited, it would not be shallow.
link |
00:26:00.960
It's just that there's different modes of communication
link |
00:26:03.840
in different fields.
link |
00:26:05.120
And in some fields, the experts have embraced Wikipedia.
link |
00:26:08.800
In other fields, it's relegated.
link |
00:26:11.040
And perhaps the reason is that if it was any better to start with,
link |
00:26:16.400
people would invest more time.
link |
00:26:17.920
But if it's not great to start with,
link |
00:26:19.840
then you need a few initial pioneers who will basically go in
link |
00:26:23.040
and say, enough, we're just going to fix that.
link |
00:26:26.480
And then I think it'll catch on much more.
link |
00:26:28.960
So, if it's okay before we go on to genomics,
link |
00:26:32.080
can we linger a little bit longer on the beauty of the human genome?
link |
00:26:36.960
You've given me a few notes.
link |
00:26:38.480
What else do you find beautiful about the human genome?
link |
00:26:41.520
So, the last aspect of what makes the human genome unique,
link |
00:26:44.800
in addition to the similarity and the differences
link |
00:26:49.840
and the individuality, is that very early on,
link |
00:26:56.160
people would basically say, oh, you don't do that experiment in human.
link |
00:26:59.040
You have to learn about that in fly.
link |
00:27:01.040
Or you have to learn about that in yeast first,
link |
00:27:02.960
or in mouse first, or in a primate first.
link |
00:27:05.760
And the human genome was in fact relegated to all the last place
link |
00:27:09.760
that you're going to go to learn something new.
link |
00:27:12.480
That has dramatically changed.
link |
00:27:14.080
And the reason that changed is human genetics.
link |
00:27:18.560
We are these species in the planet that's the most studied right now.
link |
00:27:24.480
It's embarrassing to say that.
link |
00:27:26.080
But this was not the case a few years ago.
link |
00:27:28.160
It used to be first viruses, then bacteria, then yeast,
link |
00:27:35.120
then the fruit fly and the worm, then the mouse.
link |
00:27:38.800
And eventually, human was very far last.
link |
00:27:42.240
So, it's embarrassing that it took us this long to focus on it?
link |
00:27:46.480
It's embarrassing that the model organisms have been taken over
link |
00:27:50.320
because of the power of human genetics.
link |
00:27:52.560
That right now, it's actually simpler to figure out the phenotype of something
link |
00:27:57.200
by mining this massive amount of human data
link |
00:28:01.200
than by going back to any of the other species.
link |
00:28:03.840
And the reason for that is that if you look at the natural variation
link |
00:28:06.640
that happens in a population of 7 billion,
link |
00:28:09.520
you basically have a mutation in almost every nucleotide.
link |
00:28:13.280
So, every nucleotide you want to perturb,
link |
00:28:15.520
you can go find a living, breathing human being
link |
00:28:18.640
and go test the function of that nucleotide
link |
00:28:20.160
by sort of searching the database and finding that person.
link |
00:28:22.400
Wait, why is that embarrassing?
link |
00:28:23.440
It's a beautiful data set.
link |
00:28:24.400
It's huge for humans.
link |
00:28:26.240
It's embarrassing for the model organism.
link |
00:28:29.200
For the flies.
link |
00:28:30.000
Yeah, exactly.
link |
00:28:30.960
I mean, do you feel on a small tangent,
link |
00:28:34.880
is there something of value in the genome of a fly
link |
00:28:39.920
and other of these model organisms that you miss
link |
00:28:43.600
that we wish we would be looking at deeper?
link |
00:28:47.360
So, directed perturbation, of course.
link |
00:28:49.760
So, I think the place where humans are still lagging
link |
00:28:54.000
is the fact that in an animal model, you can go and say,
link |
00:28:56.320
well, let me knock out this gene completely.
link |
00:28:58.480
And let me knock out these three genes completely.
link |
00:29:00.480
And the moment you get into combinatorics,
link |
00:29:02.640
it's something you can't do in the human
link |
00:29:04.080
because there just simply aren't enough humans on the planet.
link |
00:29:06.960
And again, let me be honest,
link |
00:29:08.800
we haven't sequenced all 7 billion people.
link |
00:29:11.040
It's not like we have every mutation,
link |
00:29:12.640
but we know that there's a carrier out there.
link |
00:29:14.880
So, if you look at the trend and the speed
link |
00:29:17.360
with which human genetics has progressed,
link |
00:29:19.360
we can now find thousands of genes involved
link |
00:29:23.200
in human cognition, in human psychology,
link |
00:29:27.040
in the emotions and the feelings
link |
00:29:28.960
that we used to think are uniquely learned.
link |
00:29:31.600
Turns out there's a genetic basis to a lot of that.
link |
00:29:34.320
So, the human genome has continued to elucidate
link |
00:29:42.480
through these studies of genetic variation,
link |
00:29:44.800
so many different processes
link |
00:29:46.160
that we previously thought were something like free will.
link |
00:29:51.680
Free will is this beautiful concept
link |
00:29:54.240
that humans have had for a long time.
link |
00:29:57.760
You know, in the end,
link |
00:29:58.480
it's just a bunch of chemical reactions
link |
00:29:59.840
happening in your brain.
link |
00:30:00.720
And the particular abundance of receptors
link |
00:30:03.120
that you have this day based on what you ate yesterday
link |
00:30:06.080
or that you have been wired with
link |
00:30:08.400
based on your parents and your upbringing, etc.
link |
00:30:12.720
Determines a lot of that, quote unquote,
link |
00:30:14.400
free will component to sort of narrower slices.
link |
00:30:21.680
So, how much on that point,
link |
00:30:23.280
how much freedom do you think we have
link |
00:30:25.600
to escape the constraints of our genome?
link |
00:30:30.480
You're making it sound like more and more
link |
00:30:31.840
we're discovering that our genome
link |
00:30:33.280
is actually has a lot of the story already encoded into it.
link |
00:30:37.600
How much freedom do we have?
link |
00:30:38.800
So, let me describe what that freedom would look like.
link |
00:30:44.960
That freedom would be my saying,
link |
00:30:47.520
oh, I'm going to resist the urge to eat that apple
link |
00:30:51.520
because I choose not to.
link |
00:30:54.480
But there are chemical receptors
link |
00:30:56.320
that made me not resist the urge
link |
00:30:59.280
to prove my individuality and my free will
link |
00:31:02.160
by resisting the apple.
link |
00:31:03.920
So then the next question is,
link |
00:31:05.520
well, maybe now I'll resist the urge to resist the apple
link |
00:31:08.080
and I'll go for the chocolate instead
link |
00:31:09.440
to prove my individuality.
link |
00:31:10.640
But then what about those other receptors that, you know...
link |
00:31:15.840
That might be all encoded in there.
link |
00:31:17.680
So, it's kicking the bucket down the road
link |
00:31:19.360
and basically saying, well, your choice
link |
00:31:22.000
will may have actually been driven
link |
00:31:23.680
by other things that you actually are not choosing.
link |
00:31:27.680
So, that's why it's very hard to answer that question.
link |
00:31:29.920
It's hard to know what to do with that.
link |
00:31:31.280
I mean, if the genome has...
link |
00:31:32.960
If there's not much freedom, it's the butterfly effect.
link |
00:31:40.400
It's basically that in the short term,
link |
00:31:42.880
you can predict something extremely well
link |
00:31:45.600
by knowing the current state of the system.
link |
00:31:48.000
But a few steps down,
link |
00:31:49.520
it's very hard to predict based on the current knowledge.
link |
00:31:52.320
Is that because the system is truly free?
link |
00:31:55.040
When I look at weather patterns,
link |
00:31:56.240
I can predict the next 10 days.
link |
00:31:57.680
Is it because the weather has a lot of freedom
link |
00:32:00.080
and after 10 days,
link |
00:32:01.280
it chooses to do something else?
link |
00:32:03.280
Or is it because, in fact, the system is fully deterministic?
link |
00:32:07.280
And there's just a slightly different magnetic field of the earth,
link |
00:32:10.480
slightly more energy arriving from the sun,
link |
00:32:12.240
a slightly different spin of the gravitational pull of Jupiter
link |
00:32:15.920
that is now causing all kinds of tides
link |
00:32:18.720
and slight deviation of the moon, etc.
link |
00:32:20.720
Maybe all of that can be fully modeled.
link |
00:32:22.880
Maybe the fact that China is emitting a little more carbon today
link |
00:32:27.120
is actually going to affect the weather in Egypt in three weeks.
link |
00:32:31.360
And all of that could be fully modeled.
link |
00:32:33.760
In the same way,
link |
00:32:34.960
if you take a complete view of a human being now,
link |
00:32:39.440
I model everything about you,
link |
00:32:42.720
the question is, can I predict your next step?
link |
00:32:44.800
Probably.
link |
00:32:46.320
But at how far?
link |
00:32:47.680
And if it's a little further,
link |
00:32:49.360
is that because of stochasticity
link |
00:32:51.280
and sort of chaos properties of unpredictability
link |
00:32:54.480
of beyond a certain level?
link |
00:32:56.000
Or was that actually true free will?
link |
00:32:58.240
Yeah, so the number of variables might be so.
link |
00:33:01.280
You might need to build an entire universe to be able to model.
link |
00:33:05.120
To simulate a human.
link |
00:33:06.560
And then maybe that human will be fully simulatable.
link |
00:33:09.440
But maybe aspects of free will will exist.
link |
00:33:12.080
And where's that free will coming from?
link |
00:33:13.280
It's still coming from the same neurons,
link |
00:33:14.800
or maybe from a spirit inhabiting these neurons.
link |
00:33:17.440
But again, it's very difficult empirically
link |
00:33:19.600
to sort of evaluate where does free will begin
link |
00:33:22.400
and sort of chemical reactions and electric signals and, you know, and...
link |
00:33:26.720
So on that topic, let me ask the most absurd question
link |
00:33:31.120
that most MIT faculty roll their eyes on.
link |
00:33:33.840
But what do you think about the simulation hypothesis
link |
00:33:38.160
and the idea that we live in a simulation?
link |
00:33:40.080
I think it's completely BS.
link |
00:33:43.840
Okay.
link |
00:33:44.320
There's no empirical evidence.
link |
00:33:45.600
No, it's not.
link |
00:33:46.080
Absolutely not.
link |
00:33:47.120
Not in terms of empirical evidence,
link |
00:33:48.640
not but in terms of thought experiment.
link |
00:33:52.240
Does it help you think about the universe?
link |
00:33:54.720
I mean, so if you look at the genome,
link |
00:33:57.360
it's encoding a lot of the information
link |
00:33:59.040
that is required to create some of the beautiful human complexity
link |
00:34:02.320
that we see around this.
link |
00:34:04.080
It's an interesting thought experiment.
link |
00:34:05.920
How much, you know, parameters do we need to have
link |
00:34:11.200
in order to model some, you know, this full human experience?
link |
00:34:15.120
Like if we were to build a video game,
link |
00:34:16.720
yeah, how hard it would be to build a video game
link |
00:34:19.840
that's like convincing enough and fun enough and, you know,
link |
00:34:24.880
it has consistent laws of physics, all that stuff.
link |
00:34:28.160
It's not interesting to use a thought experiment.
link |
00:34:31.200
I mean, it's cute, but, you know, it's Occam's razor.
link |
00:34:34.880
I mean, what's more realistic,
link |
00:34:36.640
the fact that you're actually a machine
link |
00:34:38.160
or that you're, you know, a person?
link |
00:34:39.600
What's, you know, the fact that all of my experiences exist
link |
00:34:43.200
inside the chemical molecules that I have
link |
00:34:45.440
or that somebody's actually, you know, same lading all that.
link |
00:34:49.360
Well, you did refer to humans as a digital computer earlier.
link |
00:34:52.080
Of course, of course, but that does not...
link |
00:34:54.000
It's kind of a machine, right?
link |
00:34:55.200
I know, I know.
link |
00:34:56.240
But I think the probability of all that is nil
link |
00:35:01.680
and let the machines wake me up
link |
00:35:03.360
and just terminate me now if it's not.
link |
00:35:07.360
I challenge you machines.
link |
00:35:08.640
They're gonna wait a little bit to see what you're gonna do next.
link |
00:35:12.320
It's fun.
link |
00:35:12.880
It's fun to watch,
link |
00:35:13.680
especially the clever humans.
link |
00:35:17.280
What's the difference to you between the way
link |
00:35:19.440
a computer stores information
link |
00:35:21.280
and the human genome stores information?
link |
00:35:23.840
So you also have roots and your work.
link |
00:35:26.800
Would you say you're...
link |
00:35:28.640
When you introduce yourself at a bar...
link |
00:35:31.760
It depends who I'm talking to.
link |
00:35:33.840
Would you say it's computation biology?
link |
00:35:36.000
Do you reveal your expertise in computers?
link |
00:35:42.480
It depends who I'm talking to, truly.
link |
00:35:45.200
I mean, basically, if I meet someone who's in computers,
link |
00:35:47.600
I'll say, oh, I mean, professor in computer science.
link |
00:35:50.960
If I meet someone who's in engineering,
link |
00:35:52.320
I say computer science and electrical engineering.
link |
00:35:54.640
If I meet someone in biology, I'll say, hey, I work in genomics.
link |
00:35:57.040
If I meet someone in medicine, I'm like, hey, I work on genetics.
link |
00:36:00.640
You're a fun person to meet at a bar.
link |
00:36:02.080
I got you, but so...
link |
00:36:03.760
No, no, but what I'm trying to say is that I don't...
link |
00:36:07.520
There's no single attribute that I will define myself as.
link |
00:36:10.320
There's a few things I know.
link |
00:36:11.920
There's a few things I study.
link |
00:36:12.960
There's a few of the things I have degrees on,
link |
00:36:14.880
and there's a few things that I grant degrees in.
link |
00:36:17.920
And I publish papers across the whole gamut,
link |
00:36:22.720
the whole spectrum of computation to biology, et cetera.
link |
00:36:26.240
I mean, the complete answer is that I use computer science
link |
00:36:31.360
to understand biology.
link |
00:36:34.000
So I'm a developed methods in AI and machine learning statistics
link |
00:36:39.840
and algorithms, et cetera.
link |
00:36:41.440
But the ultimate goal of my career is to really understand biology.
link |
00:36:45.520
If these things don't advance our understanding of biology,
link |
00:36:49.440
I'm not as fascinated by them.
link |
00:36:51.840
Although there are some beautiful computational problems by themselves,
link |
00:36:56.320
I've sort of made it my mission to apply the power of computer science
link |
00:37:01.440
to truly understand the human genome, health, disease,
link |
00:37:06.000
you know, and the whole gamut of how our brain works,
link |
00:37:09.120
our body works, and all of that, which is so fascinating.
link |
00:37:12.960
So there's not an equivalent sort of complementary dream
link |
00:37:18.800
of understanding human biology in order to create an artificial life,
link |
00:37:22.400
an artificial brain, an artificial intelligence
link |
00:37:25.120
that supersedes the intelligence and the capabilities of us humans.
link |
00:37:29.680
It's an interesting question.
link |
00:37:31.120
It's a fascinating question.
link |
00:37:32.240
So understanding the human brain is undoubtedly coupled
link |
00:37:39.600
to how do we make better AI?
link |
00:37:42.080
Because so much of AI has, in fact, been inspired by the brain.
link |
00:37:47.120
It may have taken 50 years since the early days of neural networks
link |
00:37:50.960
till we have, you know, all of these amazing progress
link |
00:37:54.720
that we've seen with, you know, deep belief networks and, you know,
link |
00:38:02.800
all of these advances in go and chess, in image synthesis,
link |
00:38:08.400
in deep fakes, in you name it.
link |
00:38:11.840
And but the underlying architecture
link |
00:38:15.280
is very much inspired by the human brain,
link |
00:38:17.920
which actually posits a very, very interesting question.
link |
00:38:20.560
Why are neural networks performing so well?
link |
00:38:27.040
And they perform amazingly well.
link |
00:38:28.880
Is it because they can simulate any possible function?
link |
00:38:32.400
And the answer is no, no, they simulate a very small number of functions.
link |
00:38:37.040
Is it because they can simulate every functional function
link |
00:38:39.360
in the universe?
link |
00:38:40.400
And that's where it gets interesting.
link |
00:38:41.440
The answer is actually, yeah, a little closer to that.
link |
00:38:44.640
And here's where it gets really fun.
link |
00:38:46.160
If you look at human brain and human cognition,
link |
00:38:51.600
it didn't evolve in a vacuum.
link |
00:38:53.840
It evolved in a world with physical constraints,
link |
00:38:58.720
like the world that inhabits us.
link |
00:39:00.560
It is the world that we inhabit.
link |
00:39:03.120
And if you look at our senses, what do they perceive?
link |
00:39:08.080
They perceive different, you know, parts of the electromagnetic spectrum.
link |
00:39:11.760
You know, the hearing is just different movements in air, the touch, et cetera.
link |
00:39:17.760
I mean, all of these things, we've built intuitions
link |
00:39:20.720
for the physical world that we inhabit.
link |
00:39:23.120
And our brains and the brains of all animals evolved for that world.
link |
00:39:27.680
And the AI systems that we have built happen to work well with images
link |
00:39:33.520
of the type that we encounter in the physical world that we inhabit.
link |
00:39:36.240
Whereas if you just take noise and you add random signal
link |
00:39:40.560
that doesn't match anything in our world, neural networks will not do as well.
link |
00:39:44.960
And that actually basically has this whole loop around this,
link |
00:39:50.800
which is this was designed by studying our own brain,
link |
00:39:55.360
which was evolved for our own world,
link |
00:39:57.680
and they happen to do well in our own world.
link |
00:39:59.920
And they happen to make the same types of mistakes
link |
00:40:01.920
that humans make many, many times.
link |
00:40:04.640
And of course, you can engineer images by adding just the right amount of,
link |
00:40:08.240
you know, sort of pixel deviations to make a zebra look like a bamboo
link |
00:40:11.920
and stuff like that, or like a table.
link |
00:40:16.080
But ultimately, the undocked images at least are very often, you know, mistaken,
link |
00:40:23.040
I don't know, between muffins and dogs, for example,
link |
00:40:26.160
in the same way that humans make those mistakes.
link |
00:40:28.720
So it's, you know, there's no doubt in my view that the more we understand
link |
00:40:34.640
about the tricks that our human brain has evolved
link |
00:40:37.440
to understand the physical world around us,
link |
00:40:39.600
the more we will be able to bring new computational primitives in our AI systems
link |
00:40:45.520
to, again, better understand not just the world around us,
link |
00:40:49.040
but maybe even the world inside us,
link |
00:40:51.440
and maybe even the computational problems that arise from new types of data
link |
00:40:55.360
that we haven't seen in the past.
link |
00:40:57.040
It's from new types of data that we haven't been exposed to,
link |
00:41:00.240
but are yet inhabiting the same universe that we live in
link |
00:41:03.360
with the very tiny little subset of functions
link |
00:41:06.000
from all possible mathematical functions.
link |
00:41:08.080
Yeah, and that small subset of functions, all that matters to us humans, really.
link |
00:41:11.600
That's what makes...
link |
00:41:12.800
It's all that has mattered so far, and even within our scientific realm,
link |
00:41:17.120
it's all that seems to continue to matter.
link |
00:41:19.760
But, I mean, I always like to think about our senses
link |
00:41:24.000
and how much of the physical world around us we perceive.
link |
00:41:28.960
And if you look at the LIGO experiment of the last year and a half,
link |
00:41:35.920
has been all over the news, what did LIGO do?
link |
00:41:38.720
It created a new sense for human beings,
link |
00:41:42.160
a sense that has never been sensed in the history of our planet.
link |
00:41:48.160
Gravitational waves have been traversing the Earth
link |
00:41:50.800
since its creation a few billion years ago.
link |
00:41:54.240
Life has evolved senses to sense things that were never before sensed.
link |
00:42:01.200
Light was not perceived by early life.
link |
00:42:04.400
No one cared.
link |
00:42:06.400
And eventually, photoreceptors evolved,
link |
00:42:10.080
and the ability to sense colors by sort of catching different parts
link |
00:42:14.800
of that electromagnetic spectrum.
link |
00:42:17.040
And hearing evolved, and touch evolved, et cetera.
link |
00:42:21.040
But no organism evolved a way to sense neutrinos
link |
00:42:24.720
floating through Earth, or gravitational waves
link |
00:42:27.040
flowing through Earth, et cetera.
link |
00:42:28.560
And I find it so beautiful in the history of not just humanity,
link |
00:42:32.000
but life on the planet,
link |
00:42:33.840
that we are now able to capture additional signals
link |
00:42:37.280
from the physical world than we ever knew before.
link |
00:42:40.560
And axioms, for example, have been all over the news in the last few weeks.
link |
00:42:44.480
The concept that we can capture and perceive more of that physical world
link |
00:42:52.000
is as exciting as the fact that we were blind to it is traumatizing before.
link |
00:42:59.680
Because that also tells us, we're in 2020.
link |
00:43:05.040
Picture yourself in 30, 20, or in 20, you know.
link |
00:43:08.560
What new senses might we have?
link |
00:43:10.560
Is it, you know, could it be that we're missing nine tenths of physics?
link |
00:43:16.480
Yeah.
link |
00:43:16.960
That there's a lot of physics out there that we're just blind to,
link |
00:43:20.320
completely oblivious to it, and yet they're permeating us all the time.
link |
00:43:24.160
Yeah, so it might be right in front of us.
link |
00:43:25.840
So when you're thinking about premonitions,
link |
00:43:29.840
yeah, a lot of that is ascertainment bias.
link |
00:43:32.320
Like, yeah, every now and then you're like,
link |
00:43:34.080
oh, I remember my friend,
link |
00:43:35.680
and you're like, oh, I remember my friend,
link |
00:43:37.760
and you're like, oh, I remember my friend,
link |
00:43:40.160
and then my friend doesn't appear,
link |
00:43:42.000
and I'll forget that I remembered my friend.
link |
00:43:43.760
But every now and then my friend will actually appear,
link |
00:43:45.360
and I'm like, oh, my God, I thought about you a minute ago,
link |
00:43:47.600
you just called me, that's amazing.
link |
00:43:49.120
So, you know, some of that is this,
link |
00:43:51.200
but some of that might be that there are, within our brain,
link |
00:43:56.160
sensors for waves that we emit that we're not even aware of.
link |
00:44:02.560
And this whole concept of when I hug my children,
link |
00:44:06.160
there's such an emotional transfer there
link |
00:44:10.320
that we don't comprehend.
link |
00:44:12.080
I mean, sure, yeah, of course,
link |
00:44:13.440
we're all like hard wire for all kinds of touchy feely things
link |
00:44:16.560
between parents and kids, it's beautiful,
link |
00:44:18.080
between partners, it's beautiful, et cetera.
link |
00:44:20.560
But then there are intangible aspects of human communication
link |
00:44:27.280
that I don't think it's unfathomable
link |
00:44:29.840
that our brain has actually evolved ways and sensors for it
link |
00:44:32.640
that we just don't capture.
link |
00:44:33.840
We don't understand the function
link |
00:44:35.120
of the vast majority of our neurons.
link |
00:44:37.440
And maybe our brain is already sensing it,
link |
00:44:40.000
but even worse, maybe our brain is not sensing it at all.
link |
00:44:43.840
And we're oblivious to this until we build a machine
link |
00:44:46.480
that suddenly is able to sort of capture
link |
00:44:48.240
so much more of what's happening in the natural world.
link |
00:44:50.160
So what you're saying is we're going,
link |
00:44:52.160
physics is going to discover a sensor for love.
link |
00:44:57.120
And maybe dogs are off scale for that.
link |
00:45:01.360
And we've been, you know,
link |
00:45:02.480
we've been oblivious to it the whole time
link |
00:45:03.920
because we didn't have the right sensor.
link |
00:45:05.600
And now you're going to have a little wrist that says,
link |
00:45:07.280
oh my God, I feel all this love in the house.
link |
00:45:09.520
I sense these turbines in the forest.
link |
00:45:12.640
It's all around us.
link |
00:45:13.520
And dogs and cats will have zero.
link |
00:45:15.680
None, none.
link |
00:45:16.480
None, none.
link |
00:45:17.040
It's just, nothing lost.
link |
00:45:20.000
But let's take a step back to our unfortunate place.
link |
00:45:24.640
One of the 400 topics that we had actually planned for.
link |
00:45:28.880
Yeah, but to our sad time in 2020
link |
00:45:31.680
when we only have just a few sensors and
link |
00:45:34.640
very primitive early computers.
link |
00:45:37.520
So in your, you have a foot in computer science
link |
00:45:41.680
and a foot in biology.
link |
00:45:43.360
In your sense, how do computers represent information
link |
00:45:48.160
differently than like the genome or biological systems?
link |
00:45:52.160
So first of all, let me correct that,
link |
00:45:55.760
no, we're in an amazing time in 2020.
link |
00:45:57.920
Computer science is totally awesome.
link |
00:46:02.400
And physics is totally awesome.
link |
00:46:03.840
And we have understood so much of the natural world
link |
00:46:06.800
than ever before.
link |
00:46:08.400
So I am extremely grateful and feeling extremely lucky
link |
00:46:13.040
to be living the time that we are.
link |
00:46:16.000
Because, you know, first of all,
link |
00:46:17.440
who knows when the asteroid will hit?
link |
00:46:21.760
And second, you know, of all times in humanity,
link |
00:46:26.000
this is probably the best time to be a human being.
link |
00:46:29.280
And this might actually be the best place
link |
00:46:30.960
to be a human being.
link |
00:46:31.680
So anyway, you know, for anyone who loves science,
link |
00:46:34.320
this is, this is it.
link |
00:46:35.120
This is awesome.
link |
00:46:35.680
It's a great time.
link |
00:46:36.800
At the same time, just a quick comment.
link |
00:46:39.280
All I meant is that if we look several hundred years from now
link |
00:46:43.520
and we end up somehow not destroying ourselves,
link |
00:46:48.400
people will probably look back at this time
link |
00:46:50.240
in computer science and at your work of Manos at MIT.
link |
00:46:54.960
In fontile.
link |
00:46:56.480
In fontile and silly and how ignorant it all was.
link |
00:46:59.520
I like to joke very often with my students
link |
00:47:02.400
that, you know, we've written so many papers.
link |
00:47:04.080
We've published so much.
link |
00:47:05.120
We've been citing so much.
link |
00:47:06.400
And every single time I tell my students,
link |
00:47:08.000
you know, the best is ahead of us.
link |
00:47:09.600
What we're working on now is the most exciting thing
link |
00:47:12.320
I've ever worked on.
link |
00:47:13.760
So in a way, I do have this sense of,
link |
00:47:16.080
yeah, even the papers I wrote 10 years ago,
link |
00:47:18.400
they were awesome at the time.
link |
00:47:20.160
But I'm so much more excited about where we're heading now.
link |
00:47:22.320
And I don't mean to minimize
link |
00:47:23.680
any of the stuff we've done in the past,
link |
00:47:25.360
but, you know, there's just this sense of excitement
link |
00:47:28.880
about what you're working on now
link |
00:47:30.960
that as soon as a paper is submitted,
link |
00:47:33.200
it's like, oh, it's old.
link |
00:47:35.040
Like, you know, I can't talk about that anymore.
link |
00:47:36.960
At the same time, you're not,
link |
00:47:38.560
you probably are not going to be able to predict
link |
00:47:41.360
what are the most impactful papers and ideas.
link |
00:47:45.360
When people look back 200 years from now at your work,
link |
00:47:47.760
what would be the most exciting papers?
link |
00:47:50.560
And it may very well be not the thing that you expected, or the things you got awards for,
link |
00:47:57.120
or, you know.
link |
00:47:57.920
This might be true in some fields.
link |
00:47:59.840
I don't know.
link |
00:48:00.240
I feel slightly differently about it in our field.
link |
00:48:02.240
I feel that I kind of know what are the important ones.
link |
00:48:05.600
And there's a very big difference
link |
00:48:07.200
between what the press picks up on
link |
00:48:09.120
and what's actually fundamentally important for the field.
link |
00:48:11.520
And I think for the fundamentally important ones,
link |
00:48:13.280
we kind of have a pretty good idea what they are.
link |
00:48:15.520
And it's hard to sometimes get the press excited
link |
00:48:18.080
about the fundamental advances.
link |
00:48:19.520
But, you know, we take what we get and celebrate what we get.
link |
00:48:24.720
And sometimes, you know, one of our papers,
link |
00:48:27.120
which was in a minor journal, made the front page of Reddit
link |
00:48:30.160
and suddenly had like hundreds of thousands of views.
link |
00:48:33.360
Even though it was in a minor journal,
link |
00:48:34.960
because, you know, somebody pitched it the right way
link |
00:48:36.960
that it suddenly caught everybody's attention.
link |
00:48:39.200
Whereas other papers that are sort of truly fundamental,
link |
00:48:42.160
you know, we have a hard time getting the editors even excited about them.
link |
00:48:46.080
When so many hundreds of people are already using the results
link |
00:48:49.120
and building upon them.
link |
00:48:50.880
So I do appreciate that there's a discrepancy
link |
00:48:54.320
between the perception and the perceived success
link |
00:48:57.360
and the awards that you get for various papers.
link |
00:48:59.360
But I think that fundamentally I know that, you know, some paper.
link |
00:49:02.960
I'm so, so when you're right, you're most proud.
link |
00:49:06.720
See now you just, you trapped yourself.
link |
00:49:09.280
No, no, no, no, I mean, is there a line of work that you've,
link |
00:49:12.400
you have a sense is really powerful that you've done the day.
link |
00:49:16.880
You've done so much work in so many directions,
link |
00:49:19.440
which is interesting.
link |
00:49:21.120
Is there something where you think is quite special?
link |
00:49:24.640
I mean, it's like asking me to say which of my three children I love best.
link |
00:49:29.520
I mean,
link |
00:49:33.520
Exactly.
link |
00:49:34.000
So, I mean, and it's such a gimme question that is so,
link |
00:49:39.760
so difficult not to brag about the awesome work
link |
00:49:43.360
that my team and my students have done.
link |
00:49:45.200
And I'll just mention a few off the top of my head.
link |
00:49:48.080
I mean, basically there's a few landmark papers
link |
00:49:51.120
that I think have shaped my scientific path.
link |
00:49:54.480
And, you know, I like to somehow describe it as a linear continuation
link |
00:50:00.160
of one thing led to another and led to another led to another.
link |
00:50:03.280
And, you know, it kind of all started with skip, skip, skip, skip, skip.
link |
00:50:10.320
Let me try to start somewhere in the middle.
link |
00:50:12.080
So, my first PhD paper was the first comparative analysis
link |
00:50:17.600
of multiple species, so multiple complete genomes.
link |
00:50:20.640
So for the first time, we basically developed a concept
link |
00:50:24.400
of genome wide evolutionary signatures,
link |
00:50:26.960
the fact that you could look across the entire genome
link |
00:50:29.600
and understand how things evolve.
link |
00:50:32.160
And from these signatures of evolution, you could go back
link |
00:50:36.400
and study any one region and say,
link |
00:50:38.960
that's a protein coding gene, that's an RNA gene,
link |
00:50:42.080
that's a regulatory motif, that's a, you know, binding site,
link |
00:50:45.680
and so on and so forth.
link |
00:50:46.880
So, I'm sorry, so comparing different species of the same.
link |
00:50:51.440
So, so the human mouse, rat and dog, you know,
link |
00:50:54.800
they're all animals, they're all mammals,
link |
00:50:56.640
they're all performing similar functions with their heart,
link |
00:50:59.520
with their brain, with their lungs, et cetera, et cetera.
link |
00:51:02.560
So, there's many functional elements that make us
link |
00:51:05.760
uniquely mammalian, and those mammalian elements
link |
00:51:09.280
are actually conserved.
link |
00:51:11.280
99% of our genome does not code for protein.
link |
00:51:15.440
1% codes for protein.
link |
00:51:17.440
The other 99%, we frankly didn't know what it does
link |
00:51:21.920
until we started doing these comparative genomic studies.
link |
00:51:24.880
So basically, these series of papers in my career
link |
00:51:28.880
have basically first developed that concept
link |
00:51:31.440
of evolutionary signatures,
link |
00:51:32.960
and then apply them to yeast, apply them to flies,
link |
00:51:36.000
apply them to four mammals, apply them to 17 fungi,
link |
00:51:38.880
apply them to 12 susophila species,
link |
00:51:40.880
apply them to then 29 mammals, and now 200 mammals.
link |
00:51:43.840
So, sorry, so can we, so the evolutionary signatures,
link |
00:51:47.840
it seems like a such a fascinating idea,
link |
00:51:50.640
and we're probably gonna linger in your early PhD work
link |
00:51:54.640
for two hours, but what is, how can you reveal something
link |
00:52:00.640
interesting about the genome by looking at the multiple
link |
00:52:04.640
species and looking at the evolutionary signatures?
link |
00:52:08.640
Yeah, so you basically align the matching regions.
link |
00:52:14.640
So, everything evolved from a common ancestor,
link |
00:52:16.640
way, way back, and mammals evolved from a common ancestor
link |
00:52:19.680
about 60 million years back.
link |
00:52:21.680
So, after, you know, the evolution of the evolution
link |
00:52:26.640
of the dinosaurs about 50 million years back,
link |
00:52:30.640
so after, you know, the meteor that killed off the dinosaurs,
link |
00:52:36.640
landed near Machu Picchu, we know the crater,
link |
00:52:40.640
it didn't allegedly land.
link |
00:52:42.640
That was the aliens, okay.
link |
00:52:44.640
No, just slightly north of Machu Picchu,
link |
00:52:46.640
in the Gulf of Mexico, there's a giant hole that,
link |
00:52:50.640
that meteor impact.
link |
00:52:52.640
By the way, is that definitive?
link |
00:52:54.640
really figured out what killed the dinosaurs?
link |
00:52:58.160
I think so.
link |
00:52:59.160
So it was a meteor?
link |
00:53:00.760
Well, you know, volcanic activity, all kinds of other stuff is coinciding, but the meteor
link |
00:53:08.400
is pretty unique and we now have...
link |
00:53:09.880
That's also terrifying.
link |
00:53:13.360
We still have a lot of 2020 left, so if anything...
link |
00:53:15.720
No, no, but think about it this way.
link |
00:53:17.400
So the dinosaurs ruled the Earth for 175 million years.
link |
00:53:24.640
We humans have been around for, what, less than one million years, if you're super generous
link |
00:53:30.920
about what you call humans, and you include chimps basically.
link |
00:53:35.560
So we are just getting warmed up, and you know, we've ruled the planet much more ruthlessly
link |
00:53:42.520
than Tyrannosaurus rex.
link |
00:53:46.320
Tyrannosaurus rex had much less of an environmental impact than we did.
link |
00:53:50.040
And if you give us another 174 million years, you know, humans will look very different
link |
00:53:56.760
if we make it that far.
link |
00:53:58.480
So I think dinosaurs basically are much more of life history on Earth than we are in all
link |
00:54:05.400
respects.
link |
00:54:06.400
But look at the bright side, when they were killed off, another life form emerged, mammals.
link |
00:54:10.960
And that's that whole evolutionary branching that's happened.
link |
00:54:15.160
So you kind of have, when you have these evolutionary signatures, is there basically a map of how
link |
00:54:21.400
the genome changed throughout?
link |
00:54:23.640
So now you can go back to this early mammal that was hiding in caves, and you can basically
link |
00:54:28.840
ask what happened after the dinosaurs were wiped out.
link |
00:54:31.200
A ton of evolutionary niches opened up, and the mammals started populating all of these
link |
00:54:35.880
niches.
link |
00:54:37.800
And in that diversification, there was room for expansion of new types of functions.
link |
00:54:44.920
So some of them populated the air with bats flying, a new evolution of light.
link |
00:54:53.600
Some populated the oceans with dolphins and whales going off to swim, et cetera.
link |
00:54:58.840
But we all are fundamentally mammals.
link |
00:55:01.560
So you can take the genomes of all these species and align them on top of each other.
link |
00:55:06.440
And basically create nucleotide resolution correspondences.
link |
00:55:12.040
What my PhD work showed is that when you do that, when you line up species on top of each
link |
00:55:16.160
other, you can see that within protein coding genes, there's a particular pattern of evolution
link |
00:55:21.960
that is dictated by the level at which evolutionary selection acts.
link |
00:55:28.760
If I'm coding for a protein, and I change the third code on position of a triplet that
link |
00:55:35.320
codes for that amino acid, the same amino acid will be encoded.
link |
00:55:39.880
So that basically means that any kind of mutation that preserves that translation that is invariant
link |
00:55:46.920
to that ultimate functional assessment that evolution will give is tolerated.
link |
00:55:53.320
So for any function that you're trying to achieve, there's a set of sequences that
link |
00:55:57.160
encode it.
link |
00:55:58.640
You can now look at the mapping, the graph isomorphism, if you wish, between all of the
link |
00:56:06.640
possible DNA encodings of a particular function and that function.
link |
00:56:10.560
And instead of having just that exact sequence at the protein level, you can think of the
link |
00:56:15.200
set of protein sequences that all fulfill the same function.
link |
00:56:18.840
What's evolution doing?
link |
00:56:20.440
Evolution has two components.
link |
00:56:21.600
One component is random, blind, and stupid mutation.
link |
00:56:26.080
The other component is super smart, ruthless selection.
link |
00:56:32.840
That's my mom calling from Greece.
link |
00:56:35.000
Yes, I might be a fully grown man, but I am a Greek.
link |
00:56:40.720
Did you just cancel the call?
link |
00:56:41.720
Wow, you're in trouble.
link |
00:56:42.720
I know, I'm in trouble.
link |
00:56:43.720
She's going to be calling the cops.
link |
00:56:44.720
I'm going to edit this clip out and send it to her.
link |
00:56:51.720
So there's a lot of encoding for the same kind of function.
link |
00:56:54.960
So you now have this mapping between all of the set of functions that could all encode
link |
00:57:00.160
the same, all of the set of sequences that can all encode the same function.
link |
00:57:04.400
What evolutionary signatures does is that it basically looks at the shape of that distribution
link |
00:57:10.360
of sequences that all encode the same thing.
link |
00:57:13.520
And based on that shape, you can basically say, ooh, proteins have a very different
link |
00:57:17.320
shape than RNA structures, than regulatory motifs, et cetera.
link |
00:57:21.400
So just by scanning a sequence, ignoring the sequence, and just looking at the patterns
link |
00:57:25.560
of change, I'm like, wow, this thing is evolving like a protein, and that thing is evolving
link |
00:57:30.840
like a motif, and that thing is evolving.
link |
00:57:33.260
So that's exactly what we just did for COVID.
link |
00:57:35.720
So our paper that we posted in a bioarchive about coronavirus basically took this concept
link |
00:57:40.200
of evolutionary signatures and applied it on the SARS CoV2 genome that is responsible
link |
00:57:46.480
of the COVID 19 pandemic.
link |
00:57:48.760
And comparing it to 44 Sarbacovirus species.
link |
00:57:52.520
So this is the beta.
link |
00:57:53.520
What word did you just use, Sarbacovirus?
link |
00:57:56.360
Sarbacovirus, the SARS related beta coronavirus.
link |
00:58:00.600
It's a portmanteau.
link |
00:58:01.600
That's a family of viruses.
link |
00:58:03.400
Yeah.
link |
00:58:04.400
How big is that family, by the way?
link |
00:58:05.400
We have 44 species that...
link |
00:58:07.160
It's 44 species in the family.
link |
00:58:09.880
Virus is a clever bunch.
link |
00:58:10.880
No, no, but there's just 44, and again, we don't call them species in viruses.
link |
00:58:15.720
We call them strains.
link |
00:58:16.720
But anyway, there's 44 strains, and that's a tiny little subset of maybe another 50 strains
link |
00:58:22.280
that are just far too distantly related.
link |
00:58:24.800
Most of those only infect bats as the host, and a subset of only four or five have ever
link |
00:58:32.320
infected humans.
link |
00:58:34.320
And we basically took all of those, and we aligned them in the same exact way that we've
link |
00:58:37.960
aligned mammals.
link |
00:58:38.960
And then we looked at what proteins are...
link |
00:58:42.520
Which of the currently hypothesized genes for the coronavirus genome are in fact evolving
link |
00:58:48.440
like proteins, and which ones are not.
link |
00:58:50.440
And what we found is that ORF10, the last little open reading frame, the last little
link |
00:58:55.040
gene in the genome, is bogus.
link |
00:58:57.160
That's not a protein at all.
link |
00:58:58.840
What is it?
link |
00:59:00.160
It's an RNA structure.
link |
00:59:02.080
That doesn't have a...
link |
00:59:03.680
It doesn't get translated into amino acids.
link |
00:59:05.840
And that's...
link |
00:59:06.840
So it's important to narrow down to basically discover what's useful and what's not.
link |
00:59:11.080
Exactly.
link |
00:59:12.080
Basically, what is even the set of genes?
link |
00:59:13.720
The other thing that these evolutionary signatures showed is that within ORF3A, like a tiny little
link |
00:59:19.520
additional gene encoded within the other gene.
link |
00:59:22.840
So you can translate a DNA sequence in three different reading frames.
link |
00:59:26.960
If you start in the first one, it's ATG, et cetera.
link |
00:59:30.160
If you start on the second one, it's TGC, et cetera.
link |
00:59:33.400
And there's a gene within a gene.
link |
00:59:36.760
So there's a whole other protein that we didn't know about that might be super important.
link |
00:59:41.360
So we don't even know the building blocks of SARS CoV2.
link |
00:59:45.800
So if we want to understand coronavirus biology and eventually find it successfully, we need
link |
00:59:50.800
to even have the set of genes.
link |
00:59:52.240
And these evolutionary signatures that are developed in my PhD work, we just recently
link |
00:59:57.000
used.
link |
00:59:58.000
You know what?
link |
00:59:59.000
Let's run with that tangent for a little bit if it's okay.
link |
01:00:01.640
Can we talk about the COVID 19 a little bit more?
link |
01:00:08.840
What's your sense about the genome, the proteins, the functions that we understand about COVID
link |
01:00:15.760
19?
link |
01:00:16.760
Where do we stand in your sense?
link |
01:00:19.200
What are the big open problems?
link |
01:00:21.600
And also, you kind of said it's important to understand what are the important proteins
link |
01:00:29.760
and why is that important?
link |
01:00:34.240
So what else does the comparison of these species tell us?
link |
01:00:39.520
What it tells us is how fast are things evolving?
link |
01:00:43.080
It tells us about at what level is the acceleration or deceleration pedal set for every one of
link |
01:00:49.720
these proteins.
link |
01:00:50.880
So the genome has 30 some genes.
link |
01:00:54.320
Some genes evolve super, super fast.
link |
01:00:56.760
Others evolve super, super slow.
link |
01:00:59.080
If you look at the polymerase gene that basically replicates the genome, that's a super slow
link |
01:01:02.520
evolving one.
link |
01:01:04.280
If you look at the nucleocapsid protein, that's also super slow evolving.
link |
01:01:09.480
If you look at the spike one protein, this is the part of the spike protein that actually
link |
01:01:13.720
touches the ACE2 receptor and then enables the virus to attach to your cells.
link |
01:01:21.440
That's the thing that gives it that visual.
link |
01:01:23.920
The corona look basically.
link |
01:01:26.120
So basically the spike protein sticks out of the virus and there's a first part of the
link |
01:01:29.800
protein, S1, which basically attaches to the ACE2 receptor and then S2 is the latch that
link |
01:01:37.640
sort of pushes and channels the fusion of the membranes and then the incorporation of
link |
01:01:42.840
the viral RNA inside our cells, which then gets translated into all of these 30 proteins.
link |
01:01:50.640
So the S1 protein is evolving ridiculously fast.
link |
01:01:56.600
So if you look at the stop versus gas pedal, the gas pedal is all the way down.
link |
01:02:03.800
Orph 8 is also evolving super fast and Orph 6 is evolving super fast.
link |
01:02:07.640
We have no idea what they do.
link |
01:02:09.040
We have some idea, but nowhere near what S1 is.
link |
01:02:12.400
So what the...
link |
01:02:13.400
Isn't that terrifying?
link |
01:02:14.400
That means that's a really useful function and if it's evolving fast, doesn't that mean
link |
01:02:20.240
new strains can be created where it does something?
link |
01:02:22.440
That means that it's searching for how to match, how to best match the host.
link |
01:02:26.960
So basically anything in general in evolution, if you look at genomes, anything that's contacting
link |
01:02:31.400
the environment is evolving much faster than anything that's internal.
link |
01:02:35.280
And the reason is that the environment changes.
link |
01:02:37.420
So if you look at the evolution of the cervical viruses, the S1 protein has evolved very rapidly
link |
01:02:44.480
because it's attaching to different hosts each time.
link |
01:02:47.680
We think of them as bats, but there's thousands of species of bats and to go from one species
link |
01:02:51.800
of bat to another species of bat, you have to adjust S1 to the new ACE2 receptor that
link |
01:02:56.240
you're going to be facing in that new species.
link |
01:02:58.360
So quick tangent.
link |
01:03:00.000
Is it fascinating to you that viruses are doing this?
link |
01:03:03.320
I mean, it feels like they're this intelligent organism.
link |
01:03:06.960
I mean, does that give you pause how incredible it is that the evolutionary dynamics that
link |
01:03:15.760
you're describing is actually happening and they're figuring out how to jump from bats
link |
01:03:21.200
to humans all in this distributed fashion.
link |
01:03:24.640
And then most of us don't even say they're alive or intelligent or whatever.
link |
01:03:27.920
So intelligence is in the eye of the beholder.
link |
01:03:31.360
You know, stupid is a stupid does, as Forrest Gump would say.
link |
01:03:35.280
And intelligent is as intelligent does.
link |
01:03:36.920
So basically if the virus is finding solutions that we think of as intelligent, yeah, it's
link |
01:03:41.360
probably intelligent, but that's again in the eye of the beholder.
link |
01:03:44.200
Do you think viruses are intelligent?
link |
01:03:46.080
Of course not.
link |
01:03:47.080
Really?
link |
01:03:48.080
No.
link |
01:03:49.080
It's so incredible.
link |
01:03:50.400
So remember, remember when I was talking about the two components of evolution.
link |
01:03:53.840
One is the stupid mutation, which is completely blind, and the other one is the super smart
link |
01:03:59.400
selection, which is ruthless.
link |
01:04:02.040
So it's not viruses who are smart.
link |
01:04:04.960
It's this component of evolution that's smart.
link |
01:04:06.960
So it's evolution that sort of appears smart.
link |
01:04:10.560
And how is that happening?
link |
01:04:12.200
By huge parallel search across thousands of, you know, parallel infections throughout the
link |
01:04:20.800
world right now.
link |
01:04:21.800
Yes.
link |
01:04:22.800
Let's go back on that.
link |
01:04:24.080
So yes.
link |
01:04:25.080
So then the intelligence is in the mechanism.
link |
01:04:28.240
But then by that argument, viruses would be more intelligent because there's just more
link |
01:04:33.640
of them.
link |
01:04:34.800
So the search, they're basically the brute force search that's happening with viruses
link |
01:04:40.200
because there's so many more of them than humans, then they're taken as a whole or more
link |
01:04:46.920
intelligent.
link |
01:04:47.920
So you don't think it's possible that, I mean, who runs, would we even be here if viruses
link |
01:04:55.200
weren't, I mean, who runs this thing?
link |
01:04:58.560
So let me answer your question.
link |
01:05:03.200
So we would not be here if it wasn't for viruses.
link |
01:05:10.600
And part of the reason is that if you look at mammalian evolution early on in this mammalian
link |
01:05:15.040
radiation that basically happened after the death of the dinosaurs is that some of the
link |
01:05:19.560
viruses that we had in our genome spread throughout our genome and created binding sites for new
link |
01:05:27.720
classes of regulatory proteins.
link |
01:05:30.520
And these binding sites that landed all over our genome are now control elements that basically
link |
01:05:35.560
control our genes and sort of help the complexity of the circuitry of mammalian genomes.
link |
01:05:42.360
So everything's co evolution and we're working together and yet you say they're dumb.
link |
01:05:49.240
No, I never said they're dumb.
link |
01:05:51.640
They just don't care.
link |
01:05:53.720
They don't care.
link |
01:05:55.240
Another thing, oh, is the virus trying to kill us?
link |
01:05:56.920
No, it's not.
link |
01:05:58.080
The virus is not trying to kill you.
link |
01:06:00.080
It's actually actively trying to not kill you.
link |
01:06:02.920
So when you get infected, if you die, Palmer, I killed him.
link |
01:06:07.480
Is what the reaction of the virus will be.
link |
01:06:09.280
Why?
link |
01:06:10.280
Because that virus won't spread.
link |
01:06:12.280
I think people have a misconception of viruses are smart or viruses are mean.
link |
01:06:16.480
They don't care.
link |
01:06:19.360
You have to clean yourself of any kind of anthropomorphism out there.
link |
01:06:23.200
I don't know.
link |
01:06:24.200
Oh, yes.
link |
01:06:25.200
So there's a sense when taken as a whole that there's a...
link |
01:06:30.200
Tim, you have to be holder.
link |
01:06:33.320
Stupid is a stupid does.
link |
01:06:34.560
Intelligent is a stupid.
link |
01:06:35.560
Intelligent does.
link |
01:06:36.560
So if you want to call them intelligent, that's fine because the end result is that they're
link |
01:06:40.520
finding amazing solutions.
link |
01:06:42.440
Right.
link |
01:06:43.440
I mean, I mean...
link |
01:06:44.440
But they're so dumb about it.
link |
01:06:45.720
They're just doing dumb.
link |
01:06:46.720
They don't care.
link |
01:06:47.720
They're not dumb and they're not...
link |
01:06:48.720
They just don't care.
link |
01:06:49.720
They don't care.
link |
01:06:50.720
Exactly.
link |
01:06:51.720
The care word is really interesting.
link |
01:06:52.720
Exactly.
link |
01:06:53.720
I mean, there could be an argument that they're conscious.
link |
01:06:54.720
They're just dividing.
link |
01:06:55.720
They're not.
link |
01:06:56.720
They're just dividing.
link |
01:06:57.720
They're just a little entity which happens to be dividing and spreading.
link |
01:07:02.760
It doesn't want to kill us.
link |
01:07:04.560
In fact, it prefers not to kill us.
link |
01:07:06.480
It just wants to spread.
link |
01:07:07.920
And when I say wants, again, I'm anthropomorphizing, but it's just that if you have two versions
link |
01:07:14.440
of a virus, one acquires a mutation that spreads more, that's going to spread more.
link |
01:07:18.800
One acquires a mutation that spreads less, that's going to be lost.
link |
01:07:21.920
One acquires a mutation that enters faster, that's going to be kept.
link |
01:07:25.240
One acquires a mutation that kills you right away, it's going to be lost.
link |
01:07:28.560
So over evolutionary time, the viruses that spread super well but don't kill the host
link |
01:07:33.920
are the ones that are going to survive.
link |
01:07:35.560
Yeah.
link |
01:07:36.560
And so you're brilliantly described the basic mechanisms of how it all happens, but when
link |
01:07:41.280
you zoom out and you see the entirety of viruses, maybe across different strains of viruses,
link |
01:07:50.120
it seems like a living organism.
link |
01:07:52.360
I am in awe of biology.
link |
01:07:55.000
I find biology amazingly beautiful.
link |
01:07:58.240
I find the design of the current coronavirus, however lethal it is, amazingly beautiful.
link |
01:08:04.360
The way that it is encoded, the way that it tricks your cells into making 30 proteins
link |
01:08:10.440
from a single RNA.
link |
01:08:12.640
Human cells don't do that.
link |
01:08:14.640
Human cells make one protein from each RNA molecule.
link |
01:08:18.280
They don't make two, they make one.
link |
01:08:20.360
We are hardwired to make only one protein from every RNA molecule.
link |
01:08:23.960
And yet this virus goes in, throws in a single messenger RNA.
link |
01:08:28.840
Just like any messenger RNA, we have tens of thousands of messenger RNAs in our cells
link |
01:08:32.640
in any one time.
link |
01:08:34.240
In every one of our cells.
link |
01:08:35.960
It throws in one RNA and that RNA is so, I'm going to use your word here, not my word,
link |
01:08:43.160
intelligent, that it hijacks the entire machinery of your human cell.
link |
01:08:49.440
It basically has at the beginning a giant open reading frame.
link |
01:08:54.640
That's a giant protein that gets translated.
link |
01:08:57.240
Two thirds of that RNA make a single giant protein.
link |
01:09:01.840
That single protein is basically what a human cell would make.
link |
01:09:04.640
It's like, oh, here's a start code.
link |
01:09:06.320
I'm going to start translating here.
link |
01:09:07.920
Human cells are kind of dumb.
link |
01:09:08.920
I'm sorry.
link |
01:09:09.920
Again, this is not the word that we'd normally use, but the human cell basically says, oh,
link |
01:09:13.840
this is an RNA.
link |
01:09:14.840
It must be mine.
link |
01:09:15.840
Let me translate.
link |
01:09:16.840
And it starts translating it.
link |
01:09:17.840
And then you're in trouble.
link |
01:09:18.840
Why?
link |
01:09:19.840
Because that one protein, as it's growing, gets cleaved into about 20 different peptides.
link |
01:09:27.200
The first peptide and the second peptide start interacting and the third one and the fourth
link |
01:09:32.240
one.
link |
01:09:33.240
And they shut off the ribosome of the whole cell to not translate human RNAs anymore.
link |
01:09:42.980
So the virus basically hijacks your cells and it cuts, it cleaves every one of your
link |
01:09:49.240
human RNAs to basically say to the ribosome, don't translate this one junk.
link |
01:09:53.640
Don't look at this one junk.
link |
01:09:55.440
And it only spares its own RNAs because they have a particular mark that it spares.
link |
01:10:01.500
Then all of the ribosomes that normally make protein in your human cells are now only able
link |
01:10:06.800
to translate viral RNAs.
link |
01:10:08.800
They're more and more and more and more of them.
link |
01:10:11.680
That's the first 20 proteins.
link |
01:10:13.200
In fact, halfway down about protein 11, between 11 and 12, you basically have a translational
link |
01:10:19.800
slippage where the ribosome skips reading frame.
link |
01:10:23.560
And it translates from one reading frame to another reading frame.
link |
01:10:25.800
That means that about half of them are going to be translated from one to 11.
link |
01:10:29.560
And the other half are going to be translated from 12 to 16.
link |
01:10:32.800
It's gorgeous.
link |
01:10:34.480
And then you're done.
link |
01:10:37.520
Then that mRNA will never translate the last 10 proteins, but spike is the one right after
link |
01:10:42.040
that one.
link |
01:10:43.040
So how does spike even get translated?
link |
01:10:45.240
This positive strand RNA virus has a reverse transcriptase, which is an RNA based reverse
link |
01:10:51.560
transcriptase.
link |
01:10:52.560
So from the RNA on the positive strand, it makes an RNA on the negative strand.
link |
01:10:57.240
And in between every single one of these genes, these open reading frames, there's a little
link |
01:11:01.600
signal, AACGCA or something like that, that basically loops over to the beginning of the
link |
01:11:08.960
RNA.
link |
01:11:10.120
And basically instead of sort of having a single full negative strand RNA, it basically has
link |
01:11:15.160
a partial negative strand RNA that ends right before the beginning of that gene.
link |
01:11:19.800
And another one that ends right before the beginning of that gene.
link |
01:11:22.120
These negative strand RNAs now make positive strand RNAs that then look to the human host
link |
01:11:26.880
cell, just like any other human mRNA.
link |
01:11:29.560
It's like, oh, great.
link |
01:11:30.560
I'm going to translate that one because it doesn't have the cleaving that the virus has
link |
01:11:33.720
now put on all your human genes.
link |
01:11:36.520
And then you've lost the battle.
link |
01:11:38.600
That cell is now only making proteins for the virus that will then create the spike protein,
link |
01:11:45.600
the envelope protein, the membrane protein, the nucleocapsid protein that will package
link |
01:11:49.280
up the RNA and then sort of create new viral envelopes.
link |
01:11:54.000
And these will then be secreted out of that cell in new little packages that will then
link |
01:11:59.640
infect the rest of the cells.
link |
01:12:01.040
Repeat the whole process again.
link |
01:12:02.040
It's beautiful, right?
link |
01:12:03.040
It's mind blowing.
link |
01:12:04.040
It's hard not to anthropomorphize it.
link |
01:12:06.040
I know, but it's so gorgeous.
link |
01:12:08.200
So there is a beauty to it.
link |
01:12:10.000
Of course.
link |
01:12:11.000
Is it terrifying to you?
link |
01:12:14.120
So this is something that has happened throughout history.
link |
01:12:17.320
Humans have been nearly wiped out over and over and over again and yet never fully wiped
link |
01:12:22.400
out.
link |
01:12:23.400
So I'm not concerned about the human race.
link |
01:12:25.920
I'm not even concerned about, you know, the impact on sort of our survival as a species.
link |
01:12:34.040
This is absolutely something, I mean, you know, human life is so invaluable and every
link |
01:12:38.880
one of us is so invaluable.
link |
01:12:40.120
But if you think of it as sort of, is this the end of our species?
link |
01:12:44.720
By no means, basically.
link |
01:12:46.560
So let me explain.
link |
01:12:48.200
The Black Death killed what, 30% of Europe?
link |
01:12:52.520
That has left a tremendous imprint, you know, a huge hole, a horrendous hole in the genetic
link |
01:13:02.200
makeup of humans.
link |
01:13:04.840
There's been series of wiping out of huge fractions of entire species or just entire
link |
01:13:11.520
species altogether, and that has a consequence on the human immune repertoire.
link |
01:13:19.680
If you look at how Europe was shaped and how Africa was shaped by malaria, for example,
link |
01:13:27.320
all the individuals that carry a mutation that protects you from malaria were able to
link |
01:13:32.240
survive much more.
link |
01:13:33.760
And if you look at the frequency of sickle cell disease and the frequency of malaria,
link |
01:13:38.280
the maps are actually showing the same pattern, the same imprint on Africa.
link |
01:13:43.120
And that basically led people to hypothesize that the reason why sickle cell disease is
link |
01:13:46.280
so much more frequent in Americans of African descent is because there was selection in
link |
01:13:51.280
Africa against malaria, leading to sickle cell, because when the cell sickle, malaria
link |
01:13:58.120
is not able to replicate inside your cells as well, and therefore you protect against
link |
01:14:02.440
that.
link |
01:14:03.440
So if you look at human disease, all of the genetic associations that we do with human
link |
01:14:07.920
disease, you basically see the imprint of these waves of selection killing off gazillions
link |
01:14:17.040
of humans.
link |
01:14:18.720
And there's so many immune processes that are coming up as associated with so many different
link |
01:14:24.760
diseases.
link |
01:14:26.000
The reason for that is similar to what I was describing earlier, where the outward facing
link |
01:14:30.920
proteins evolve much more rapidly because the environment is always changing.
link |
01:14:36.000
But what's really interesting, the human genome is that we have coopted many of these immune
link |
01:14:39.840
genes to carry out nonimmune functions.
link |
01:14:42.520
For example, in our brain, we use immune cells to cleave off neuronal connections that don't
link |
01:14:49.080
get used.
link |
01:14:50.240
This whole use it or lose it, we know the mechanism.
link |
01:14:52.960
It's microglia that cleave off neuronal synaptic connections that are just not utilized.
link |
01:15:00.040
When you utilize them, you mark them in a particular way that basically when the microglia
link |
01:15:03.920
come, tell it, don't kill this one, it's used now.
link |
01:15:07.920
And the microglia will go off and kill it once you don't use.
link |
01:15:10.560
This is an immune function, which is coopted to do nonimmune things.
link |
01:15:15.040
If you look at our adipocytes, M1 versus M2 macrophages inside our fat will basically
link |
01:15:20.320
determine whether you're obese or not.
link |
01:15:22.720
And these are, again, immune cells that are resident and living within these tissues.
link |
01:15:27.140
So many disease associations.
link |
01:15:30.280
That we coopt these kinds of things for incredibly complicated functions.
link |
01:15:36.800
Exactly.
link |
01:15:37.800
Evolution works in so many different ways, which are all beautiful and mysterious at the
link |
01:15:42.160
same time.
link |
01:15:43.160
But not intelligent.
link |
01:15:44.160
Not intelligent.
link |
01:15:45.160
It's in the eye of the beholder.
link |
01:15:48.160
But the point that I'm trying to make is that if you look at the imprint that COVID will
link |
01:15:53.640
have, hopefully it will not be big.
link |
01:15:56.320
Hopefully the U.S. will get attacked together and stop the virus from spreading further.
link |
01:16:00.720
But if it doesn't, it's having an imprint on individuals who have particular genetic
link |
01:16:06.080
repertoires.
link |
01:16:07.080
So if you look at now the genetic associations of blood type and immune function cells,
link |
01:16:12.920
et cetera, there's actually association, genetic variation that basically says how
link |
01:16:16.440
much more likely am I or you to die if we contact the virus.
link |
01:16:20.360
And it's through these rounds of shaping the human genome that humans have basically
link |
01:16:25.720
made it so far.
link |
01:16:27.840
And selection is ruthless and it's brutal and it only comes with a lot of killing.
link |
01:16:34.480
But this is the way that viruses and environments have shaped the human genome.
link |
01:16:39.440
Basically, when you go through periods of famine, you select for particular genes.
link |
01:16:43.840
And what's left is not necessarily better, it's just whatever survived.
link |
01:16:49.200
And it may have been the surviving one back then, not because it was better, maybe the
link |
01:16:53.400
ones that ran slower survived.
link |
01:16:54.960
I mean, you know, again, not necessarily better.
link |
01:16:57.520
But the surviving ones are basically the ones that then are shaped for any kind of subsequent
link |
01:17:03.480
evolutionary condition and environmental condition.
link |
01:17:07.360
But if you look at, for example, obesity, obesity was selected for basically the genes
link |
01:17:12.400
that now predisposes to obesity were at 2% frequency in Africa.
link |
01:17:16.760
They rose to 44% frequency in Europe because you basically went through the ice ages and
link |
01:17:23.040
there was a scarcity of food, so you know, there was a selection to being able to store
link |
01:17:27.120
every single calorie you consume.
link |
01:17:30.000
Eventually, environment changes.
link |
01:17:33.440
So the better allele, which was the fat storing allele, became the worse allele because it's
link |
01:17:38.680
the fat storing allele.
link |
01:17:40.440
It still has the same function.
link |
01:17:42.580
So if you look at my genome, speaking of mom calling, mom gave me a bad copy of that gene,
link |
01:17:48.320
this FTO locus.
link |
01:17:49.320
Basically makes me...
link |
01:17:50.320
The one that has to do with obesity.
link |
01:17:53.120
Yeah.
link |
01:17:54.120
Basically now have a bad copy from mom that makes me more likely to be obese.
link |
01:17:58.120
And I also have a bad copy from dad that makes me more likely to be obese.
link |
01:18:01.600
I'm homozygous.
link |
01:18:03.600
And that's the allele.
link |
01:18:06.600
It's still the minor allele, but it's at 44% frequency in Southeast Asia, 42% frequency
link |
01:18:11.600
in Europe, even though it started at 2%.
link |
01:18:14.360
It was an awesome allele to have 100 years ago.
link |
01:18:17.720
Right now, it's a pretty terrible allele.
link |
01:18:19.520
So the other concept is that diversity matters.
link |
01:18:23.720
If we had 100 million nuclear physicists living the earth right now, we'd be in trouble.
link |
01:18:30.720
You need diversity, you need artists and you need musicians and you need mathematicians
link |
01:18:34.880
and you need politicians, yes, even those.
link |
01:18:38.280
And you need like...
link |
01:18:39.280
Well, let's not get crazy now.
link |
01:18:41.600
Because then if a virus comes along or whatever...
link |
01:18:44.440
Exactly.
link |
01:18:45.440
Exactly.
link |
01:18:46.440
So no, there's two reasons.
link |
01:18:47.440
One, you want diversity in the immune repertoire and we have built in diversity.
link |
01:18:52.000
So basically, they are the most diverse...
link |
01:18:54.600
Basically if you look at our immune system, there's layers and layers of diversity.
link |
01:18:57.680
Like, the way that you create your cells generates diversity because of the selection for the
link |
01:19:04.360
VDGA recombination that basically eventually leads to a huge number of repertoires.
link |
01:19:09.000
But that's only one small component of diversity.
link |
01:19:10.960
The blood type is another one.
link |
01:19:12.320
The major histocopatibility complex, the HLA alleles are another source of diversity.
link |
01:19:18.840
So the immune system of humans is by nature incredibly diverse and that basically leads
link |
01:19:25.520
to resilience.
link |
01:19:26.520
So basically what I'm saying that I don't worry for the human species because we are
link |
01:19:31.160
so diverse immunologically, we are likely to be very resilient against so many different
link |
01:19:37.600
attacks like this current virus.
link |
01:19:40.040
So you're saying natural pandemics may not be something that you're really afraid of
link |
01:19:44.600
because of the diversity in our genetic makeup.
link |
01:19:49.560
What about engineered pandemics?
link |
01:19:50.880
Do you have fears of us messing with the makeup of viruses or...
link |
01:19:56.960
Well, yeah, let's say with the makeup of viruses to create something that we can't control
link |
01:20:01.680
and would be much more destructive than it would come about naturally.
link |
01:20:06.560
Remember how we were talking about how smart evolution is, humans are much dumber.
link |
01:20:10.760
You mean like human scientists, engineers?
link |
01:20:13.800
Humans just like...
link |
01:20:14.800
Humans overall.
link |
01:20:15.800
Yeah, humans overall.
link |
01:20:17.300
But I mean, even the sort of synthetic biologists, basically if you were to create virus like
link |
01:20:28.320
SARS that will kill a lot of people, you would probably start with SARS.
link |
01:20:34.160
So whoever would like to design such a thing would basically start with SARS tree or at
link |
01:20:41.320
least some relative of SARS.
link |
01:20:43.720
The source genome for the current virus was something completely different.
link |
01:20:48.360
It was something that has never infected humans.
link |
01:20:50.740
No one in their right mind would have started there.
link |
01:20:52.920
But when you say source, it's like the nearest...
link |
01:20:55.000
The nearest relative is in a whole other branch, no species of which has ever infected humans
link |
01:21:00.840
in that branch.
link |
01:21:02.640
So let's put this to rest.
link |
01:21:05.400
This was not designed by someone to kill off the human race.
link |
01:21:09.320
You don't believe it was engineered?
link |
01:21:11.320
Well, likely.
link |
01:21:12.640
Yeah, the path to engineering a deadly virus would not come from this strain that was used.
link |
01:21:21.160
Moreover, there's been various claims of, ha, ha, this was mixed and matched in the
link |
01:21:28.500
lab because the S1 protein has three different components, each of which has a different
link |
01:21:33.360
evolutionary tree.
link |
01:21:34.720
So a lot of popular press basically said, ha, this came from pangolin and this came
link |
01:21:39.520
from all kinds of other species.
link |
01:21:43.000
This is what has been happening throughout the coronavirus tree.
link |
01:21:46.920
So basically the S1 protein has been recombining across species all the time.
link |
01:21:50.640
Remember when I was talking about the positive strand, the negative strand, subgenomic RNAs,
link |
01:21:54.560
these can actually recombine.
link |
01:21:55.560
And if you have two different viruses infecting the same cell, they can actually mix and match
link |
01:21:59.560
between the positive strand and the negative strand and basically create a new hybrid
link |
01:22:03.040
virus with recombination that now has the S1 from one and the rest of the genome from
link |
01:22:07.960
another.
link |
01:22:08.960
And this is something that happens a lot in S1, in Orphe, et cetera.
link |
01:22:12.080
And that's something that's true of the whole tree.
link |
01:22:13.960
For the whole family of coronavirus.
link |
01:22:15.480
Exactly.
link |
01:22:16.480
So it's not like someone has been messing with this for millions of years and, you know,
link |
01:22:20.600
changing the situation.
link |
01:22:21.600
So that's, again, beautiful that that somehow happens, that they recombine.
link |
01:22:25.960
So two different strands can infect the body and then recombine.
link |
01:22:30.560
So all of this actually magic happens inside hosts.
link |
01:22:35.200
Like all, like...
link |
01:22:36.200
Yeah.
link |
01:22:37.200
That's why, that's why classification wise, virus is not thought to be alive because it
link |
01:22:40.960
doesn't self replicate.
link |
01:22:41.960
It's not autonomous.
link |
01:22:43.120
It's something that enters a living cell and then coops it to basically make it its own.
link |
01:22:48.920
But by itself, people ask me, how do we kill this bastard?
link |
01:22:51.600
I'm like, you stop it from replicating.
link |
01:22:54.320
It's not like a bacterium that will just live in a, you know, puddle or something.
link |
01:23:01.320
It's a virus.
link |
01:23:02.840
Viruses don't live without their hosts.
link |
01:23:04.760
And they only live within their hosts for very little time.
link |
01:23:07.440
So if you stop it from replicating, it'll stop from spreading.
link |
01:23:10.440
I mean, it's not like HIV, which can stay dormant for a long time.
link |
01:23:14.080
Basically coronavirus is just don't do that.
link |
01:23:15.640
They're not integrating genomes.
link |
01:23:16.840
They're RNA genomes.
link |
01:23:17.840
So if it's not expressed, it degrades.
link |
01:23:20.280
RNA degrades.
link |
01:23:21.280
It doesn't just stick around.
link |
01:23:22.680
Well, let me ask also about the immune system you mentioned.
link |
01:23:27.280
A lot of people kind of ask, you know, how can we strengthen the immune system to respond
link |
01:23:34.680
to this particular virus, when the virus is in general?
link |
01:23:37.480
Do you have from a biological perspective thoughts on what we can do as humans to strengthen
link |
01:23:43.800
our immune system?
link |
01:23:44.800
If you look at the death rates across different countries.
link |
01:23:46.880
People with less vaccination have been dying more.
link |
01:23:49.800
If you look at North Italy, the vaccination rates are abysmal there, and a lot of people
link |
01:23:55.320
have been dying.
link |
01:23:56.320
If you look at Greece, very good vaccination rates.
link |
01:23:59.080
Almost no one has been dying.
link |
01:24:00.440
So yes, there's a policy component.
link |
01:24:03.680
So Italy reacted very slowly.
link |
01:24:06.200
Greece reacted very fast.
link |
01:24:07.600
So yeah, many fewer people died in Greece.
link |
01:24:09.880
But there might actually be a component of a genetic immune repertoire, basically how
link |
01:24:14.520
did people die off, you know, in the history of the Greek population versus the Italian
link |
01:24:20.160
population?
link |
01:24:21.160
Wow.
link |
01:24:22.160
That's interesting to think about.
link |
01:24:24.880
And then there's a component of what vaccinations did you have as a kid and what are the off
link |
01:24:29.800
target effects of those vaccinations?
link |
01:24:32.560
So basically a vaccination can have two components.
link |
01:24:35.120
One is training your immune system against that specific insult.
link |
01:24:39.520
The second one is boosting up your immune system for all kinds of other things.
link |
01:24:44.680
If you look at allergies, Northern Europe, super clean environments, tons of allergies.
link |
01:24:51.480
Southern Europe, my kids grew up eating dirt.
link |
01:24:55.040
No allergies.
link |
01:24:56.040
So growing up, I never had even heard of what allergies are, like really allergies.
link |
01:25:02.200
And the reason is that I was playing in the garden.
link |
01:25:03.520
I was putting all kinds of stuff in my mouth from, you know, all kinds of dirt and stuff.
link |
01:25:07.520
Tons of viruses, there are tons of bacteria there.
link |
01:25:09.680
You know, my immune system was built up.
link |
01:25:11.600
So the more you protect your immune system from exposure, the less opportunity it has
link |
01:25:18.280
to learn about non self repertoire in a way that prepares it for the next insult.
link |
01:25:24.120
So that's the horizontal thing too, like the, so it's throughout your lifetime and the
link |
01:25:28.240
lifetime of the, of the people that your ancestors, that kind of thing.
link |
01:25:34.080
Yeah.
link |
01:25:35.080
Absolutely.
link |
01:25:36.080
So it returns against free will on the free will side of things.
link |
01:25:39.660
Is there something we could do to strengthen our immune system in 2020?
link |
01:25:44.880
Is there like, you know, exercise, diet, all that kind of stuff.
link |
01:25:50.840
So it's kind of funny.
link |
01:25:53.080
There's a cartoon that basically shows two windows with a teller in each window.
link |
01:25:58.480
One has a humongous line and the other one has no one.
link |
01:26:02.400
The one that has no one above says health, no, it says exercise and diet.
link |
01:26:07.320
And the other one says pill and there's a huge line for pill.
link |
01:26:12.240
So we're looking basically for magic bullets for sort of ways that we can, you know, beat
link |
01:26:17.040
cancer and beat coronavirus and beat this and beat that and it turns out that the window
link |
01:26:20.600
with like just diet and exercise is the best way to boost every aspect of your health.
link |
01:26:26.200
If you look at Alzheimer's, exercise and nutrition, I mean, you're like, really?
link |
01:26:32.360
For my brain neurodegeneration?
link |
01:26:34.640
Absolutely.
link |
01:26:35.640
If you look at cancer, exercise and nutrition, if you look at coronavirus, exercise and nutrition,
link |
01:26:44.060
every single aspect of human health gets improved.
link |
01:26:47.400
And one of the studies we're doing now is basically looking at what are the effects
link |
01:26:51.200
of diet and exercise?
link |
01:26:53.120
How similar are they to each other?
link |
01:26:55.400
We basically take in diet intervention and exercise intervention in human and in mice
link |
01:27:01.440
and we're basically doing single cell profiling of a bunch of different tissues to basically
link |
01:27:05.320
understand how are the cells, both the stromal cells and the immune cells of each of these
link |
01:27:11.440
tissues responding to the effect of exercise?
link |
01:27:15.280
What are the communication networks between different cells where the muscle that exercises
link |
01:27:22.240
sends signals through the bloodstream, through the lymphatic system, through all kinds of
link |
01:27:26.440
other systems that give signals to other cells that I have exercised and you should change
link |
01:27:32.240
in this particular way, which basically reconfigure those receptor cells with the effect of exercise?
link |
01:27:40.200
How well understood is those reconfigurations?
link |
01:27:44.160
Very little.
link |
01:27:45.160
We're just starting now, basically.
link |
01:27:47.480
Is the hope there to understand the effect on the immune system?
link |
01:27:54.240
On the immune system, the effect on brain, the effect on your liver, on your digestive
link |
01:27:58.240
system, on your adipocytes, adipose, you know, the most misunderstood organ.
link |
01:28:03.960
Everybody thinks, oh, fat, terrible, no, fat is awesome.
link |
01:28:07.600
Your fat cells is what's keeping you alive because if you didn't have your fat cells,
link |
01:28:11.680
all those lipids and all those calories would be floating around in your blood and you'd
link |
01:28:15.560
be dead by now.
link |
01:28:17.040
Your adipocytes are your best friend.
link |
01:28:18.520
They're basically storing all these excess calories so that they don't hurt all of the
link |
01:28:23.560
rest of the body and they're also fat burning in many ways.
link |
01:28:29.080
So, you know, again, when you don't have the homozygous version that I have, your cells
link |
01:28:33.880
are able to burn calories much more easily by sort of flipping a master metabolic switch
link |
01:28:40.000
that involves these FTO locus that I mentioned earlier and its target genes, IRX3 and RX5,
link |
01:28:45.120
that basically switch your adipocytes during their three first days of differentiation
link |
01:28:50.680
as they're becoming mature adipocytes to basically become either fat burning or fat storing
link |
01:28:55.880
fat cells.
link |
01:28:57.240
And the fat burning fat cells are your best friends.
link |
01:28:59.080
They're much closer to muscle than they are to white adipocytes.
link |
01:29:03.120
Is there a lot of difference between people like that you could give, science could eventually
link |
01:29:08.560
give advice that is very generalizable or is our differences in our genetic makeup, like
link |
01:29:16.240
you mentioned, is that going to be basically something we have to be very specialized individuals?
link |
01:29:23.040
Any advice we'd give in terms of diet, like what we're just talking about?
link |
01:29:26.120
Believe it or not, the most personalized advice that you give for nutrition don't have to
link |
01:29:30.040
do with your genome.
link |
01:29:31.040
They have to do with your gut microbiome, with the bacteria that live inside you.
link |
01:29:36.000
So most of your digestion is actually happening by species that are not human inside you.
link |
01:29:40.480
You have more nonhuman cells than you have human cells.
link |
01:29:43.160
They're basically a giant bag of bacteria with a few human cells alone.
link |
01:29:50.600
And those do not necessarily have to do with your genetic makeup?
link |
01:29:55.040
They interact with your genetic makeup.
link |
01:29:56.840
They interact with your epigenome.
link |
01:29:58.040
They interact with your nutrition.
link |
01:29:59.640
They interact with your environment.
link |
01:30:01.640
They're basically an additional source of variation.
link |
01:30:07.080
So when you're thinking about personalized nutritional advice, part of that is actually
link |
01:30:11.400
how do you match your microbiome?
link |
01:30:13.760
And part of that is how do we match your genetics?
link |
01:30:17.240
But again, this is a very diverse set of contributors.
link |
01:30:22.440
And the effect sizes are not enormous, so I think the science for that is not fully developed
link |
01:30:27.160
yet.
link |
01:30:28.160
Speaking of diets, because I've wrestled in combat sports, but sports my whole life
link |
01:30:31.480
were weight matters, so you have to cut and all that stuff.
link |
01:30:35.520
One thing I've learned a lot about my body, and it seems to be, I think, true about other
link |
01:30:40.560
people's bodies is that you can adjust to a lot of things.
link |
01:30:45.280
That's the miraculous thing about this biological system is I fast often.
link |
01:30:52.320
I used to eat five, six times a day, and thought that was absolutely necessary.
link |
01:30:57.120
How could you not eat that often?
link |
01:30:59.280
And then when I started fasting, your body adjusted that, and you learned how to not
link |
01:31:03.600
eat.
link |
01:31:04.600
And if you just give it a chance for a few weeks, actually, over a period of a few weeks,
link |
01:31:10.360
your body can adjust to anything.
link |
01:31:11.880
And that's a miracle.
link |
01:31:12.880
That's such a beautiful thing.
link |
01:31:14.240
So I'm a computer scientist, and I've basically gone through periods of 24 hours without eating
link |
01:31:18.680
or stopping, and then I'm like, oh, must eat, and I eat a ton.
link |
01:31:23.080
I used to order two pizzas just with my brother.
link |
01:31:27.560
So I've gone through these extremes as well, and I've gone through the whole intermittent
link |
01:31:31.040
fasting thing, so I can sympathize with you both on the seven meals a day to the zero
link |
01:31:35.920
meals a day.
link |
01:31:37.680
So I think when I say everything with moderation, I actually think your body responds interestingly
link |
01:31:44.280
to these different changes in diet.
link |
01:31:47.320
I think part of the reason why we lose weight with pretty much every kind of change in behavior
link |
01:31:52.040
is because our epigenome and the set of proteins and enzymes that are expressed and our microbiome
link |
01:31:58.600
are not well suited to that nutritional source, and therefore, they will not be able to sort
link |
01:32:04.080
of catch everything that you give them, and then a lot of that will go undigested.
link |
01:32:09.280
And that basically means that your body can then lose weight in the short term, but very
link |
01:32:13.680
quickly will adjust to that new normal, and then we'll be able to sort of perhaps gain
link |
01:32:18.120
a lot of weight from the diet.
link |
01:32:20.480
So anyway, I mean, there's also studies in factories where basically people dim the
link |
01:32:26.400
lights, and then suddenly everybody started working better.
link |
01:32:28.720
It was like, wow, that's amazing.
link |
01:32:30.320
Three weeks later, they made the lights a little brighter.
link |
01:32:32.640
Everybody started working better.
link |
01:32:35.440
So any kind of intervention has a placebo effect of, wow, now I'm healthier, and I'm
link |
01:32:40.720
going to be running more often, et cetera.
link |
01:32:42.120
So it's very hard to uncouple the placebo effect of, wow, I'm doing something to intervene
link |
01:32:46.040
on my diet from the, wow, this is actually the right thing for me.
link |
01:32:50.360
So you know.
link |
01:32:51.360
Yeah, from the perspective from nutrition science, psychology, both things I'm interested
link |
01:32:55.560
in, especially psychology, it seems that it's extremely difficult to do good science because
link |
01:33:02.600
there's so many variables involved, it's so difficult to control the variables, so difficult
link |
01:33:07.080
to do sufficiently large scale experiments, both sort of in terms of number of subjects
link |
01:33:12.800
and temporal, like how long you do the study for, that it just seems like it's not even
link |
01:33:19.800
a real science for now, like nutrition science.
link |
01:33:22.640
I want to jump into the whole placebo effect for a little bit here, and basically talk
link |
01:33:28.440
about the implications of that.
link |
01:33:30.320
If I give you a sugar pill and I tell you it's a sugar pill, you won't get any better.
link |
01:33:35.360
But if I tell you a sugar pill and I tell you, wow, this is an amazing drug, it actually
link |
01:33:40.400
will stop your cancer, your cancer will actually stop with much higher probability.
link |
01:33:46.440
What does that mean?
link |
01:33:47.440
That's so amazing.
link |
01:33:48.440
That means that if I can trick your brain into thinking that I'm healing you, your brain
link |
01:33:52.360
will basically figure out a way to heal itself, to heal the body.
link |
01:33:56.240
And that tells us that there's so much that we don't understand in the interplay between
link |
01:34:01.920
our cognition and our biology that if we were able to better harvest the power of our brain
link |
01:34:10.080
to sort of impact the body through the placebo effect, we would be so much better in so many
link |
01:34:16.080
different things.
link |
01:34:17.520
Just by tricking yourself into thinking that you're doing better, you're actually doing
link |
01:34:20.760
better.
link |
01:34:21.760
So there's something to be said about sort of positive thinking, about optimism, about
link |
01:34:25.360
sort of just getting your brain and your mind into the right mindset that helps your body
link |
01:34:34.920
and helps your entire biology.
link |
01:34:37.360
From a science perspective, that's just fascinating.
link |
01:34:39.920
Obviously, most things about the brain is a total mystery for now, but that's a fascinating
link |
01:34:45.200
interplay that the brain can reduce.
link |
01:34:51.480
The brain can help cure cancer, I don't even know what to do with that.
link |
01:34:56.000
I mean, the way to think about that is the following.
link |
01:34:59.520
The converse of the equation is something that we are much more comfortable with.
link |
01:35:03.320
Like, oh, if you're stressed, then your heart might rise and all kinds of sort of toxins
link |
01:35:10.200
might be released and that can have a detrimental effect on your body, et cetera, et cetera.
link |
01:35:15.240
So maybe it's easier to understand your body healing from your mind by your mind is not
link |
01:35:21.800
killing your body, or at least it's killing it less.
link |
01:35:24.920
So I think that aspect of the stress equation is a little easier for most of us to conceptualize,
link |
01:35:31.840
but then the healing part is perhaps the same pathways, perhaps different pathways, but
link |
01:35:36.200
again, something that is totally untapped scientifically.
link |
01:35:39.440
I think we try to bring this question up a couple of times, but let's return to it again
link |
01:35:44.520
as what do you think is the difference between the way a computer represents information,
link |
01:35:49.480
the human genome represents and stores information, and maybe broadly, what is the difference between
link |
01:35:56.600
how you think about computers and how you think about biological systems?
link |
01:35:59.920
So I made a very provocative claim earlier that we are a digital computer, like that
link |
01:36:04.760
at the core lies a digital code, and that's true in many ways, but surrounding that digital
link |
01:36:08.880
core, there's a huge amount of analog.
link |
01:36:11.000
If you look at our brain, it's not really digital, if you look at our sort of RNA and
link |
01:36:15.840
all of that stuff inside ourselves, not really digital, it's really analog in many ways.
link |
01:36:21.160
But let's start with the code, and then we'll expand to the rest.
link |
01:36:24.920
So the code itself is digital.
link |
01:36:27.920
So there's genes, you can think of the genes as, I don't know, the procedures, the functions
link |
01:36:32.680
inside your language.
link |
01:36:34.240
And then somehow you have to turn these functions on, how do you call the gene?
link |
01:36:37.400
How do you call that function?
link |
01:36:39.440
The way that you would do it in old programming languages is go to, address whatever in your
link |
01:36:44.080
memory, and then you'd start running from there.
link |
01:36:46.640
And modern programming languages have encapsulated this into functions and objects and all of
link |
01:36:51.720
that, and it's nice and cute, but in the end, deep down, there's still an assembly code
link |
01:36:55.520
that says go to that instruction, and it runs that instruction.
link |
01:36:59.800
If you look at the human genome, and the genome of pretty much most species out there, there's
link |
01:37:07.040
no go to function.
link |
01:37:08.240
You just don't start in transcribing in position 1300, 13,527 in chromosome 12.
link |
01:37:18.920
You instead have content based indexing.
link |
01:37:22.000
So at every location in the genome, in front of the genes that need to be turned on, I
link |
01:37:28.760
don't know when you drink coffee, there's a little coffee marker in front of all of them.
link |
01:37:34.800
And whenever your cells that metabolize coffee need to metabolize coffee, they basically see
link |
01:37:40.400
coffee and they're like, ooh, let's go turn on all the coffee marked genes.
link |
01:37:44.840
So there's basically these small motifs, these small sequences that we call regulatory motifs.
link |
01:37:50.920
They're like patterns of DNA.
link |
01:37:52.120
They're only eight characters long or so, like GAT, GCA, et cetera.
link |
01:37:58.360
And these motifs work in combinations, and every one of them has some recruitment affinity
link |
01:38:06.320
for a different protein that will then come and bind it.
link |
01:38:09.800
And together, collections of these motifs create regions that we call regulatory regions
link |
01:38:15.440
that can be either promoters near the beginning of the gene, and that basically tells you
link |
01:38:20.080
where the function actually starts, where you call it, and then enhancers that are looping
link |
01:38:24.760
around of the DNA that basically bring the machinery that binds those enhancers and then
link |
01:38:30.000
bring it onto the promoter, which then recruits the right sort of the ribosome and the polymerase
link |
01:38:36.680
and all of that thing, which will first transcribe and then export and then eventually translate
link |
01:38:41.560
in the cytoplasm, you know, whatever RNA molecule.
link |
01:38:45.720
So the beauty of the way that the digital computer that's the genome works is that it's
link |
01:38:56.040
extremely fault tolerant.
link |
01:38:58.840
If I took your hard drive and I messed with 20% of the letters in it, of those zeros and
link |
01:39:04.560
ones and I flipped them, you'd be in trouble.
link |
01:39:08.240
If I take the genome and I flipped 20% of the letters, you probably won't even notice.
link |
01:39:14.760
And that resilience is a key design principle, and again, I'm thrombomorphizing here, but
link |
01:39:21.520
it's a key driving principle of how biological systems work.
link |
01:39:25.160
They're first resilient and then anything else.
link |
01:39:28.680
And when you look at these incredible beauty of life from the most, I don't know, beautiful,
link |
01:39:36.120
I don't know, human genome maybe of humanity and all of the ideals that come with it to
link |
01:39:41.560
the most terrifying genome like, I don't know, COVID 19, SARS COVID 2 and the current pandemic,
link |
01:39:48.360
you basically see this elegance as the epitome of clean design, but it's dirty.
link |
01:39:56.640
It's a mess.
link |
01:39:58.080
It's, you know, the way to get there is hugely messy.
link |
01:40:02.960
And that's something that we as computer scientists don't embrace.
link |
01:40:06.760
We like to have clean code, you know, as like in engineering, they teach you about compartmentalization,
link |
01:40:12.920
about sort of separating functions, about modularity, about hierarchical design, none
link |
01:40:18.080
of that applies in biology.
link |
01:40:20.080
Testing.
link |
01:40:21.080
Testing, sure.
link |
01:40:22.080
Yeah, biology does plenty of that, but I mean, through evolutionary exploration.
link |
01:40:27.120
But if you look at biological systems, first they are robust and then they specialize to
link |
01:40:35.000
become anything else.
link |
01:40:36.840
And if you look at viruses, the reason why they're so elegant when you look at the design
link |
01:40:42.200
of this, you know, genome, it seems so elegant.
link |
01:40:46.560
And the reason for that is that it's been stripped down from something much larger because
link |
01:40:52.080
of the pressure to keep it compact.
link |
01:40:54.080
So many compact genomes out there have ancestors that were much larger.
link |
01:40:58.760
You don't start small and become big.
link |
01:41:00.880
You go through a loop of add a bunch of stuff, increase complexity, and then, you know, slim
link |
01:41:06.280
it down.
link |
01:41:07.600
And one of my early papers was, in fact, on genome duplication.
link |
01:41:12.240
One of the things we found is that baker's yeast, which is the yeast that you use to
link |
01:41:16.960
make bread, but also the yeast that you use to make wine, which is basically the dominant
link |
01:41:20.440
species when you go in the fields of Tuscany and you say, you know, what's out there?
link |
01:41:24.080
It's basically Saccharomyces cerevisiae.
link |
01:41:26.480
Or the way my Italian friends say, Saccharomyces cerevisiae.
link |
01:41:31.480
Which means what?
link |
01:41:34.480
Oh, Saccharomyces.
link |
01:41:35.480
Okay.
link |
01:41:36.480
I'm sorry.
link |
01:41:37.480
I'm Greek.
link |
01:41:38.480
So, yeah.
link |
01:41:39.480
Zaharo.
link |
01:41:40.480
Zaharo is sugar.
link |
01:41:41.480
Micis is fungus.
link |
01:41:42.480
Yes.
link |
01:41:43.480
Cerevisiae.
link |
01:41:44.480
Cerveza.
link |
01:41:45.480
Beer.
link |
01:41:46.480
So it means the sugar fungus of beer.
link |
01:41:47.480
Yeah.
link |
01:41:48.480
You know, let's, let's, let's be sounding to the.
link |
01:41:51.160
Still poetic.
link |
01:41:52.160
Yeah.
link |
01:41:53.160
Anyway, Saccharomyces cerevisiae, basically the major baker's yeast out there is the descendant
link |
01:41:58.240
of a whole gene duplication.
link |
01:42:00.520
Why would a whole gene duplication even happen when it happened is coinciding with about
link |
01:42:06.800
a hundred million years ago and the emergence of fruit bearing plants?
link |
01:42:14.480
Why fruit bearing plants?
link |
01:42:15.760
Because animals would eat the fruit and would walk around and poop huge amounts of nutrients
link |
01:42:23.000
along with a seed for the plants to spread.
link |
01:42:26.760
Before that, plants were not spreading through animals.
link |
01:42:29.000
They were spreading through wind and all kinds of other ways.
link |
01:42:32.440
But basically the moment you have fruit bearing plants, the, the, the, the, these plants are
link |
01:42:36.800
basically creating this abundance of sugar in the environment.
link |
01:42:40.440
So there's an evolutionary niche that gets created.
link |
01:42:43.160
And in that evolutionary niche, you basically have enough sugar that a whole gene duplication
link |
01:42:48.760
which initially is a very messy event allows you to then, you know, relieve some of that
link |
01:42:55.080
complexity.
link |
01:42:56.080
So to pause, what does genome duplication mean?
link |
01:42:59.840
That basically means that instead of having eight chromosomes, you're gonna have 16 chromosomes.
link |
01:43:05.760
So, but the duplication at first, when you have six, when you go to 16, you're not using
link |
01:43:13.600
that.
link |
01:43:14.600
Oh yeah, you are.
link |
01:43:15.600
Yeah.
link |
01:43:16.600
So when you go to the next, you went from having eight chromosomes to having 16 chromosomes,
link |
01:43:20.800
probably a non disjunction event during a duplication, during a division.
link |
01:43:24.440
So you basically divide the cell instead of half the genome going this way and half the
link |
01:43:28.200
genome going the other way.
link |
01:43:29.440
After duplication of the genome, you basically have all of it going to one cell.
link |
01:43:33.280
And then there's a sufficient messiness there that you end up with slight differences that
link |
01:43:38.560
make most of these chromosomes be actually preserved.
link |
01:43:42.560
It's a long story short to basically, but it's a big upgrade, right?
link |
01:43:45.240
So that's not necessarily because what happens immediately thereafter is that you start massively
link |
01:43:49.720
losing tons of those duplicated genes.
link |
01:43:52.520
So 90% of those genes were actually lost very rapidly after holding duplication.
link |
01:43:58.400
And the reason for that is that biology is not intelligent.
link |
01:44:02.000
It's just ruthless selection, random mutation.
link |
01:44:06.680
So the ruthless selection basically means that as soon as one of the random mutations
link |
01:44:10.160
hit one gene, ruthless selection just kills off that gene.
link |
01:44:13.600
It's just, you know, if you have a pressure to maintain a small compact genome, you will
link |
01:44:19.840
very rapidly lose the second copy of most of your genes.
link |
01:44:23.120
And a small number, 10%, were kept into copies.
link |
01:44:26.040
And those had to do a lot with environment adaptation, with the speed of replication,
link |
01:44:31.360
with the speed of translation, and with sugar processing.
link |
01:44:34.520
So I'm making a long story short to basically say that evolution is messy.
link |
01:44:39.040
The only way, like so, you know, the example that I was giving of messing with 20% of your
link |
01:44:44.440
bits in your computer, totally bogus, duplicating all your functions and just throwing them
link |
01:44:49.480
out there in the same, you know, function, just totally bogus, like this would never
link |
01:44:53.640
work in an engineer system.
link |
01:44:55.480
But biological systems, because of this content based indexing and because of this modularity
link |
01:45:01.080
that comes from the fact that the gene is controlled by a series of tags.
link |
01:45:05.440
And now if you need this gene in another setting, you just add some more tags that will basically
link |
01:45:10.680
turn it on also in those settings.
link |
01:45:12.880
So this gene is now pressured to do two different functions.
link |
01:45:17.760
And it builds up complexity.
link |
01:45:19.960
I see a whole gene duplication and gene duplication in general as a way to relieve that complexity.
link |
01:45:24.880
So you have this gradual buildup of complexity as function gets sort of added onto the existing
link |
01:45:30.320
genes.
link |
01:45:31.320
And then boom, you duplicate your workforce, and you now have two copies of this gene.
link |
01:45:37.160
One will probably specialize to do one, and the other one will specialize to do the other,
link |
01:45:40.720
or one will maintain the ancestral function, the other one will sort of be free to evolve
link |
01:45:44.920
and specialize while losing the ancestral function and so on and so forth.
link |
01:45:48.920
So that's how genomes evolve.
link |
01:45:50.560
They're just messy things, but they're extremely fault tolerant, and they're extremely able
link |
01:45:56.720
to deal with mutations because that's the very way that you generate new functions.
link |
01:46:04.120
So new functionalization comes from the very thing that breaks it.
link |
01:46:08.000
So even in the current pandemic, many people are asking me which mutations matter the most.
link |
01:46:12.840
And what I tell them is, well, we can study the evolutionary dynamics of the current genome
link |
01:46:17.520
to then understand which mutations have previously happened or not, and which mutations happen
link |
01:46:26.400
in genes that evolve rapidly or not.
link |
01:46:29.280
And one of the things we found, for example, is that the genes that evolved rapidly in
link |
01:46:34.280
the past are still evolving rapidly now in the current pandemic.
link |
01:46:37.840
The genes that evolved slowly in the past are still evolving slowly.
link |
01:46:41.160
Which means that they're useful.
link |
01:46:43.000
Which means that they're under the same evolutionary pressures, but then the question is what happens
link |
01:46:48.560
in specific mutations.
link |
01:46:50.600
So if you look at the D614 gene mutation that's been all over the news, so in positions
link |
01:46:55.560
D614 in the amino acids, D614 of the S protein, there's a D2G mutation that sort of has crept
link |
01:47:04.520
over the population.
link |
01:47:07.880
That mutation we found out through my work disrupts a perfectly conserved nucleotide
link |
01:47:13.600
position that has never been changed in the history of millions of years of equivalent
link |
01:47:18.480
mammalian evolution of these viruses.
link |
01:47:23.240
That basically means that it's a completely new adaptation to human.
link |
01:47:27.640
And that mutation has now gone from 1% frequency to 90% frequency in almost all outbreaks.
link |
01:47:33.920
So there's a mutation, I like how you said in the mute, the 416, what was it?
link |
01:47:39.520
Yeah, 614, sorry.
link |
01:47:41.160
614.
link |
01:47:42.160
D614G.
link |
01:47:43.160
D614G.
link |
01:47:44.160
So literally, so what you're saying is this is like a chest move.
link |
01:47:48.560
So it just mutated one letter to another.
link |
01:47:50.640
Exactly.
link |
01:47:51.640
It didn't happen before.
link |
01:47:54.440
And this somehow, this mutation is really useful.
link |
01:47:58.760
It's really useful in the current environment of the genome, which is moving from human
link |
01:48:04.000
to human.
link |
01:48:05.080
When it was moving from bat to bat, it couldn't care less for that mutation.
link |
01:48:08.800
But it's environment specific, so now that it's moving from human to human, it's moving
link |
01:48:13.320
way better by orders of magnitude.
link |
01:48:17.320
So you're tracking this evolutionary dynamics, which is fascinating, but what do you do with
link |
01:48:23.680
that?
link |
01:48:24.680
So what does that mean?
link |
01:48:25.680
What do you make of this mutation in trying to anticipate, I guess, is one of the things
link |
01:48:33.520
you're trying to do is anticipate where, how this unrolls into the future, this evolutionary
link |
01:48:39.160
dynamics.
link |
01:48:40.160
Such a great question.
link |
01:48:41.160
So there's two things.
link |
01:48:43.200
Remember when I was saying earlier, mutation is the path to new things, but also the path
link |
01:48:47.720
to break old things.
link |
01:48:49.880
So what we know is that this position was extremely preserved through gazillions of mutations.
link |
01:48:56.920
That mutation was never tolerated when it was moving from bats to bats.
link |
01:49:00.320
So that basically means that that position is extremely important in the function of
link |
01:49:04.800
that protein.
link |
01:49:05.800
That's the first thing it tells.
link |
01:49:07.000
The second one is that that position was very well suited to bat transmission, but now is
link |
01:49:13.320
not well suited to human transmission, so it got rid of it, and it now has a new version
link |
01:49:17.400
of that amino acid that basically makes it much easier to transmit from human to human.
link |
01:49:22.880
So in terms of the evolutionary history teaching us about the future, it basically tells us
link |
01:49:30.640
here's the regions that are currently mutating.
link |
01:49:35.080
Here's the regions that are most likely to mutate going forward.
link |
01:49:38.040
As you're building a vaccine, here's what you should be focusing on in terms of the
link |
01:49:42.440
most stable regions that are the least likely to mutate, or here's the newly evolved functions
link |
01:49:48.200
that are the most likely to be important because they've overcome this local maximum
link |
01:49:54.480
that it had reached in the bat transmission.
link |
01:49:59.480
So anyway, it's a tangent to basically say that evolution works in messy ways, and the
link |
01:50:04.480
thing that you would break is the thing that actually allows you to first go through a
link |
01:50:11.560
lull and then reach a new local maximum.
link |
01:50:15.560
And I often like to say that if engineers had basically designed evolution, we would
link |
01:50:21.560
still be perfectly replicating bacteria because it's by making the bacterium worse that you
link |
01:50:29.720
allow evolution to reach a new optimum.
link |
01:50:32.440
That's just a pause on that, that's so profound for the entirety of this scientific and engineering
link |
01:50:42.920
disciplines.
link |
01:50:43.920
Exactly.
link |
01:50:44.920
We as engineers need to embrace breaking things.
link |
01:50:48.660
We as engineers need to embrace robustness as the first principle beyond perfection because
link |
01:50:54.360
nothing's going to ever be perfect.
link |
01:50:56.320
And when you're sending a satellite to Mars, when something goes wrong, it'll break down
link |
01:51:01.080
as opposed to building systems that tolerate failure and are resilient to that and in fact
link |
01:51:09.280
get better through that.
link |
01:51:11.240
So the SpaceX approach versus NASA for the...
link |
01:51:15.720
For example.
link |
01:51:17.720
Is there something we can learn about the incredible, take lessons from the incredible
link |
01:51:22.560
biological systems in their resilience, in their in the mushiness, the messiness to our
link |
01:51:29.120
computing systems, to our computers?
link |
01:51:31.920
It would basically be starting from scratch in many ways.
link |
01:51:35.360
It would basically be building new paradigms that don't try to get the right answer all
link |
01:51:41.480
the time, but try to get the right answer most of the time or a lot of the time.
link |
01:51:47.080
Do you see deep learning systems in the whole world of machine learning as kind of taking
link |
01:51:50.760
a step in that direction?
link |
01:51:52.360
Absolutely.
link |
01:51:53.360
Basically, by allowing this much more natural evolution of these parameters, you basically...
link |
01:52:01.360
And if you look at sort of deep learning systems, again, they're not inspired by the genome aspect
link |
01:52:06.640
of biology.
link |
01:52:07.640
They're inspired by the brain aspect of biology.
link |
01:52:10.280
And again, I want you to pause for a second and realize the complexity of the entire human
link |
01:52:17.320
brain with trillions of connections within our neurons, with millions of cells talking
link |
01:52:25.120
to each other, is still encoded within that same genome, that same genome encodes every
link |
01:52:34.400
single freaking cell type of the entire body.
link |
01:52:38.200
Every single cell is encoded by the same code, and yet specialization allows you to have
link |
01:52:45.240
the single viral like genome that self replicates, the single module, modular automaton, work
link |
01:52:54.600
with other copies of itself, it's mind boggling.
link |
01:52:59.080
Create complex organs through which blood flows.
link |
01:53:02.760
And what is that blood?
link |
01:53:03.760
The same freaking genome.
link |
01:53:07.320
Create organs that communicate with each other.
link |
01:53:10.960
And what are these organs?
link |
01:53:12.200
The exact same genome.
link |
01:53:14.640
Create a brain that is innervated by massive amounts of blood pumping energy to it, 20%
link |
01:53:22.920
of our energetic needs, to the brain from the same genome.
link |
01:53:28.280
And all of the neuronal connections, all of the auxiliary cells, all of the immune cells,
link |
01:53:33.960
the astrocytes, the ligature size, the neurons, the excitatory, the inhibitory neurons, all
link |
01:53:37.400
of the different classes of parasites, the blood brain barrier, all of that, same genome.
link |
01:53:43.000
One way to see that in a sad, this one is beautiful, the sad thing is thinking about
link |
01:53:49.120
the trillions of organisms that died to create that.
link |
01:53:55.320
You mean on the evolutionary path?
link |
01:53:56.760
Yeah, on the evolutionary path to humans.
link |
01:53:59.720
It's crazy, there's two dissenters of apes just talking on a podcast, okay, so mind boggling.
link |
01:54:08.640
Just to boggle our minds a little bit more.
link |
01:54:11.280
Just talking to each other.
link |
01:54:14.080
We are basically generating a series of vocal utterances through our pulsating of vocal
link |
01:54:21.280
cords received through this.
link |
01:54:23.520
The people who listen to this are taking a completely different path to that information
link |
01:54:30.040
transfer yet through language.
link |
01:54:33.200
But imagine if we could connect these brains directly to each other.
link |
01:54:39.160
The amount of information that I'm condensing into a small number of words is a huge funnel
link |
01:54:46.320
which then you receive and you expand into a huge number of thoughts from that small
link |
01:54:52.280
funnel.
link |
01:54:55.840
In many ways, engineers would love to have the whole information transfer, just take
link |
01:55:00.720
the whole set of neurons and throw them away, I mean throw them to the other person.
link |
01:55:05.560
This might actually not be better because in your misinterpretation of every word that
link |
01:55:11.840
I'm saying, you are creating new interpretation that might actually be way better than what
link |
01:55:16.280
I meant in the first place.
link |
01:55:18.040
The ambiguity of language perhaps might be the secret to creativity.
link |
01:55:25.280
Every single time you work on a project by yourself, you only bounce ideas with one person
link |
01:55:31.080
and your neurons are basically fully cognizant of what these ideas are.
link |
01:55:35.960
But the moment you interact with another person, the misinterpretations that happen might be
link |
01:55:41.360
the most creative part of the process.
link |
01:55:43.880
With my students, every time we have a research meeting, I very often pause and say, let me
link |
01:55:47.760
repeat what you just said in a different way.
link |
01:55:50.680
And I sort of go on and brainstorm with what they were saying, but by the third time, it's
link |
01:55:56.080
not what they were saying at all.
link |
01:55:58.120
And when they pick up what I'm saying, they're like, oh, well, da, da, da, now they've sort
link |
01:56:02.200
of learned something very different from what I was saying.
link |
01:56:05.440
And that is the same kind of messiness that I'm describing in the genome itself.
link |
01:56:11.040
It's sort of embracing the messiness.
link |
01:56:13.640
And that's a feature, not a book.
link |
01:56:15.480
Exactly.
link |
01:56:16.480
And in the same way, when you're thinking about these deep learning systems that will
link |
01:56:20.280
allow us to sort of be more creative perhaps or learn better approximations of these complex
link |
01:56:26.360
functions, again, tuned to the universe that we inhabit, you have to embrace the breaking.
link |
01:56:33.680
You have to embrace the, you know, how do we get out of these local optima?
link |
01:56:38.080
And a lot of the design paradigms that have made deep learning so successful are ways
link |
01:56:43.840
to get away from that, ways to get better training by sort of sending long range messages,
link |
01:56:50.560
these LSTM models and the, you know, sort of feed forward loops that, you know, sort
link |
01:56:57.840
of jump through layers of a convolutional neural network.
link |
01:57:01.000
All of these things are basically ways to push you out of these local maxima.
link |
01:57:07.480
And that's sort of what evolution does.
link |
01:57:08.880
That's what language does.
link |
01:57:09.880
That's what conversation and brainstorming does.
link |
01:57:12.400
That's what our brain does.
link |
01:57:13.680
So, you know, this design paradigm is something that's pervasive and yet not taught in schools
link |
01:57:20.480
not taught in engineering schools where everything's minutely modularized to make sure that we
link |
01:57:25.120
never deviate from, you know, whatever signal we're trying to emit, as opposed to let all
link |
01:57:30.440
hell breaks loose because that's the, that's the path to paradise.
link |
01:57:34.000
The path to paradise.
link |
01:57:35.000
Yeah.
link |
01:57:36.000
I mean, it's difficult to know how to teach that and what to do with it.
link |
01:57:39.280
I mean, it's, it's difficult to know how to build up a sign, the scientific method around
link |
01:57:45.720
messiness.
link |
01:57:46.720
I mean, it's not all messiness.
link |
01:57:49.960
We need, we need some cleanness.
link |
01:57:52.120
And going back to the example with Mars, that's probably the place where I want to sort of
link |
01:57:57.000
moderate error as much as possible and sort of control the environment as much as possible.
link |
01:58:01.120
But if you're trying to repopulate Mars, well, maybe messing is a good thing then.
link |
01:58:06.520
On that, you quickly mentioned this in terms of us using our vocal cords to speak on a
link |
01:58:13.480
podcast, so Elon Musk and Neuralink are working on trying to plug as per our discussion with
link |
01:58:22.720
computers and biological systems to connect it to, he's trying to connect our brain to
link |
01:58:29.360
a computer to create a brain computer interface where they can communicate back and forth.
link |
01:58:36.320
On this line of thinking, do you think this is possible to bridge the gap between our engineered
link |
01:58:43.560
computing systems and the messy biological systems?
link |
01:58:49.400
My answer would be absolutely.
link |
01:58:52.040
We, we, you know, there's no doubt that we can understand more and more about what goes
link |
01:58:55.600
on in the brain and we can sort of train the brain.
link |
01:59:00.280
I don't know if you remember the Palm pilot.
link |
01:59:02.840
Yeah.
link |
01:59:03.840
Palm pilot.
link |
01:59:04.840
Do you remember this whole sort of alphabet that they had created?
link |
01:59:07.640
Am I thinking of the same thing?
link |
01:59:11.040
It's basically you had, you had a little pen and for every character you had a little scribble
link |
01:59:16.800
that was unique that the machine could understand and that instead of trying the machine, trying
link |
01:59:22.880
to teach the machine to recognize human characters, you had basically, they figured out that it's
link |
01:59:27.400
better and easier to train humans to create human like characters that the machine is
link |
01:59:32.520
better at recognizing.
link |
01:59:34.920
So in the same way, I think what will happen is that humans will be trained to be able
link |
01:59:41.000
to create the mind pattern that the machine will respond to before the machine truly comprehends
link |
01:59:46.680
our thoughts.
link |
01:59:47.880
So the first human brain interfaces will be tricking humans to speak the machine language
link |
01:59:53.600
where with the right set of electrodes, I can sort of trick my brain into doing this.
link |
01:59:58.000
And this is the same way that many people teach, like learn to control artificial limbs.
link |
02:00:03.080
You basically try a bunch of stuff and eventually you figure out how your limbs work.
link |
02:00:07.040
That might not be very different from how humans learn to use their natural limbs when
link |
02:00:11.560
they first grow up.
link |
02:00:13.360
Basically you have these, you know, neoteny period of, you know, this puddle of soup inside
link |
02:00:20.400
your brain trying to figure out how to even make neural connections before you're born.
link |
02:00:25.720
And then learning sounds in utero of, you know, all kinds of echoes and, you know, eventually
link |
02:00:34.400
getting out in the real world.
link |
02:00:35.920
And I don't know if you've seen newborns, but they just stare around a lot.
link |
02:00:39.960
You know, one way to think about this as a machine learning person is, oh, they're just
link |
02:00:43.480
training their edge detectors.
link |
02:00:46.240
And eventually they figure out how to train their edge detectors.
link |
02:00:48.720
They work through the second layer of the visual cortex and the third layer and so on
link |
02:00:51.760
so forth.
link |
02:00:53.000
And you basically have this learning how to control your limbs that probably comes at
link |
02:01:00.240
the same time.
link |
02:01:01.240
You're sort of, you know, throwing random things there and you realize that, oh, wow,
link |
02:01:04.840
when I do this thing, my limb moves.
link |
02:01:08.360
Let's do the following experiment.
link |
02:01:09.360
Take a breath.
link |
02:01:12.040
What muscles did you flex?
link |
02:01:13.600
Now take another breath and think what muscles do I flex?
link |
02:01:16.800
The first thing that you're thinking when you're taking a breath is the impact that
link |
02:01:21.520
it has on your lungs.
link |
02:01:22.520
You're like, oh, I'm now going to increase my lungs or I'm not going to bring air in.
link |
02:01:25.560
But what you're actually doing is just changing your diaphragm.
link |
02:01:29.400
That's not conscious, of course.
link |
02:01:32.000
You never think of the diaphragm as a thing.
link |
02:01:35.240
And why is that?
link |
02:01:36.240
That's probably the same reason why I think of moving my finger when I actually move my
link |
02:01:39.480
finger.
link |
02:01:40.560
I think of the effect instead of actually thinking of whatever muscle is twitching that actually
link |
02:01:44.560
causes my finger to move.
link |
02:01:46.600
So we basically, in our first years of life, build up this massive lookup table between
link |
02:01:53.160
whatever neuronal firing we do and whatever action happens in our body that we control.
link |
02:02:01.040
If you have a kid grow up with a third limb, I'm sure they'll figure out how to control
link |
02:02:05.760
them probably at the same rate as their natural limbs.
link |
02:02:09.680
And a lot of the work would be done by the, if a third limb is a computer, you kind of
link |
02:02:15.920
have a, not a faith, but a thought that the brain might be able to figure out, like the
link |
02:02:24.440
plasticity would come from the brain, like the brain would be cleverer than the machine
link |
02:02:28.560
at first.
link |
02:02:29.560
When I talk about a third limb, that's exactly what I'm saying, an artificial limb that basically
link |
02:02:32.700
just controls your mouse while you're typing, you know, perfectly natural thing.
link |
02:02:36.640
I mean, again, you know, in a few hundred years, maybe sooner than that.
link |
02:02:42.000
But basically, there's, as long as the machine is consistent in the way that it will respond
link |
02:02:47.680
to your brain impulses, you'll figure out how to control that and you could play tennis
link |
02:02:52.480
with your third limb.
link |
02:02:54.320
And let me go back to consistency.
link |
02:02:57.840
People who have dramatic accidents that basically take out a whole chunk of their brain can
link |
02:03:04.080
be taught to coopt other parts of the brain to then control that part.
link |
02:03:08.640
You can basically build up that tissue again and eventually train your body how to walk
link |
02:03:13.160
again and how to read again and how to play again and how to think again, how to speak
link |
02:03:16.440
a language again, et cetera.
link |
02:03:18.240
So there's a massive amount of malleability that happens, you know, naturally in our way
link |
02:03:24.800
of controlling our body, our brain, our thoughts, our vocal cords, our limbs, et cetera.
link |
02:03:31.160
And human machine interfaces are all inevitable if we sort of figure out how to read these
link |
02:03:37.160
electric impulses.
link |
02:03:39.480
But the resolution at which we can understand human thought right now is nil, is ridiculous.
link |
02:03:46.680
So how are human thoughts encoded?
link |
02:03:48.840
It's basically combinations of neurons that cofire and these create these things called
link |
02:03:55.040
engrams that eventually form memories and so on and so forth.
link |
02:03:59.080
We know nothing of all that stuff.
link |
02:04:02.080
So before we can actually read into your brain that you want to build a program that does
link |
02:04:06.920
this and this and that, we need a lot of neuroscience.
link |
02:04:11.120
Well, so to push back on that, do you think it's possible that without understanding the
link |
02:04:17.320
functionally about the brain or from the neuroscience or the cognitive science or psychology, whichever
link |
02:04:22.360
level of the brain we look at, do you think we just connect them just like per your previous
link |
02:04:28.400
point?
link |
02:04:29.400
If we just have a high enough resolution between connection between Wikipedia and your brain,
link |
02:04:34.440
the brain will just figure it out without understanding.
link |
02:04:38.360
Because that's one of the innovations of Neuralink is they're increasing the number of connections
link |
02:04:43.440
to the brain to like several thousand, which before was, you know, in the dozens or whatever.
link |
02:04:48.360
You're still off by a few orders of magnitude on the order of seven.
link |
02:04:53.240
Right.
link |
02:04:54.240
But the thing is, the hope is if you increase that number more and more and more, maybe you
link |
02:04:59.120
don't need to understand anything about the actual, how human thought is represented in
link |
02:05:04.520
the brain.
link |
02:05:05.520
You can just let it, let it figure it out by itself.
link |
02:05:08.280
Yeah, Keanu Reeves waking up and saying, I know cook food.
link |
02:05:11.120
Yeah, exactly.
link |
02:05:12.120
Yeah.
link |
02:05:13.120
Exactly.
link |
02:05:14.120
So yeah, sure.
link |
02:05:15.120
You don't have faith in the, the plasticity of the brain to that degree.
link |
02:05:18.440
It's not about brain plasticity.
link |
02:05:19.960
It's about the input aspect.
link |
02:05:21.520
Basically, I think on the output aspect, being able to control a machine is something that
link |
02:05:25.480
you can probably train your neural impulses that you're sending out to sort of match whatever
link |
02:05:31.480
response you see in the environment.
link |
02:05:33.360
If this thing moved every single time I thought, a particular thought, then I could figure
link |
02:05:37.000
out, I could hack my way into moving this thing with just a series of thoughts.
link |
02:05:40.920
I could think guitar, piano, tennis ball, and then this thing would be moving.
link |
02:05:47.640
And then, you know, I would just have the series of thoughts that would sort of result
link |
02:05:51.840
in the impulses that will move this thing the way that I want.
link |
02:05:53.920
And then eventually it'll become natural because I won't even think about it.
link |
02:05:56.920
I mean, in the same way that we control our limbs in a very natural way, but babies don't
link |
02:06:00.920
do that.
link |
02:06:01.920
Babies have to figure it out.
link |
02:06:03.440
And you know, some of it is hard coded, but some of it is actually learned based on the
link |
02:06:07.360
whatever soup of neurons you ended up with, whatever connections you pruned them to, and
link |
02:06:13.560
eventually you were born with.
link |
02:06:15.240
You know, a lot of that is coded in the genome, but a huge chunk of that is stochastic instead
link |
02:06:19.880
of the way that you sort of create all these neurons that migrate, they form connections,
link |
02:06:23.440
they sort of, you know, spread out, they have particular branching patterns, but then the
link |
02:06:26.800
connectivity itself, unique in every single new person.
link |
02:06:31.400
All this to say that on the output side, absolutely, I'm very, very, you know, hopeful
link |
02:06:37.200
that we can have machines that read thousands of these neuronal connections on the output
link |
02:06:42.520
side, but on the input side, oh boy, I don't expect any time in the near future we'll be
link |
02:06:51.520
able to sort of send a series of impulses that will tell me, oh, earth to sun distance,
link |
02:06:56.520
7.5 million, et cetera, like nowhere.
link |
02:07:00.640
I mean, I think language will still be the input way rather than sort of any kind of
link |
02:07:06.400
more complex.
link |
02:07:07.400
It's a really interesting notion that the ambiguity of language is a feature.
link |
02:07:12.760
And we evolved for millions of years to take advantage of that ambiguity.
link |
02:07:20.600
And yet no one teaches us the subtle differences between words that are near cognates and yet
link |
02:07:26.560
evokes so much more than, you know, one from the other.
link |
02:07:30.880
And yet, you know, when you're choosing words from a list of 20 synonyms, you know exactly
link |
02:07:37.600
the connotation of every single one of them.
link |
02:07:40.120
And that's something that, you know, is there.
link |
02:07:42.920
So yes, there's ambiguity, but there's all kinds of connotations.
link |
02:07:46.840
And in the way that we select our words, we have so much baggage that we're sending along,
link |
02:07:51.360
the way that we're emoting, the way that we're moving our hands every single time we speak,
link |
02:07:56.080
the, you know, the pauses, the eye contact, et cetera, so much higher bar rate than just
link |
02:08:00.800
a vocal, you know, string of characters.
link |
02:08:03.920
Well, let me just take a small tangent on that.
link |
02:08:07.080
Oh, tangent.
link |
02:08:08.080
We haven't done that yet.
link |
02:08:09.080
That's a good idea.
link |
02:08:10.080
Let's just tangent.
link |
02:08:11.080
We'll return to the origin of life after.
link |
02:08:16.360
So I mean, you're Greek, but I'm going on this personal journey.
link |
02:08:22.320
I'm going to Paris for the explicit purpose of talking to one of the most famous, a couple
link |
02:08:29.920
who's a famous translators of Russian literature, Dostoevsky Tolstoy.
link |
02:08:35.280
And they go, that's their art is the translation and everything I've learned about the translation
link |
02:08:43.200
art, it makes me feel it's so profound in a way that's so much more profound than the
link |
02:08:53.720
natural language processing papers I read in the machine learning community, that there's
link |
02:08:57.840
such depth to language that I don't know what to do with.
link |
02:09:03.120
I don't know if you've experienced that in your own life with knowing multiple languages.
link |
02:09:08.720
I don't know what to, I don't know how to make sense of it.
link |
02:09:11.800
But there's so much loss in translation between Russian and English and getting a sense of
link |
02:09:16.880
that.
link |
02:09:17.880
Like for example, there's like just taking a single sentence from Dostoevsky and like
link |
02:09:24.120
there's a lot of them.
link |
02:09:25.880
You could talk for hours about how to translate that sentence properly.
link |
02:09:30.280
That captures the meaning, the period, the culture, the humor, the wit, the suffering
link |
02:09:37.240
that was in the context of the time, all of that could be a single sentence.
link |
02:09:43.080
You could talk forever about what it takes to translate that correctly.
link |
02:09:47.120
I don't know what to do with that.
link |
02:09:48.840
So being Greek, it's very hard for me to think of a sentence or even a word without going
link |
02:09:55.880
into the full etymology of that word, breaking up every single atom of that sentence and
link |
02:10:04.920
every single atom of these words and rebuilding it back up.
link |
02:10:09.760
I have three kids and the way that I teach them Greek is the same way that the documentary
link |
02:10:16.720
I was mentioning earlier about understanding the deep roots of all of these words.
link |
02:10:24.320
And it's very interesting that every single time I hear a new word that I've never heard
link |
02:10:32.080
before, I go and figure out the etymology of that word because I will never appreciate
link |
02:10:35.840
that word without understanding how it was initially formed.
link |
02:10:39.520
Interesting.
link |
02:10:40.520
But how does that help?
link |
02:10:42.320
Because that's not the full picture.
link |
02:10:43.800
No, no, of course, of course.
link |
02:10:45.000
But what I'm trying to say is that knowing the components teaches you about the context
link |
02:10:50.400
of the formation of that word and sort of the original usage of that word.
link |
02:10:55.280
And then of course, the word takes new meaning as you create it from its parts.
link |
02:11:01.200
And that meaning then gets augmented.
link |
02:11:04.280
And two synonyms that sort of have different roots will actually have implications that
link |
02:11:09.240
carry a lot of that baggage of the historical provenance of these words.
link |
02:11:14.280
So before working on genome evolution, my passion was evolution of language and sort
link |
02:11:20.240
of tracing cognates across different languages through their etymologies.
link |
02:11:27.440
And that's fascinating that there's parallels between, I mean, the idea that there's evolutionary
link |
02:11:33.520
dynamics through our language.
link |
02:11:36.440
Yeah.
link |
02:11:38.560
Every single word that you utter, parallels, parallels, what does parallels mean?
link |
02:11:44.200
Para means side by side, alleles from alleles, which means identical twins, parallels.
link |
02:11:50.760
I mean, name any word, and there's so much baggage, so much beauty in how that word came
link |
02:11:57.560
to be and how this word took a new meaning than the sum of its parts.
link |
02:12:02.320
Yeah, and they're just words, they don't have any physical grounding.
link |
02:12:08.920
And now you take these words and you weave them into a sentence.
link |
02:12:13.720
The emotional invocations of that weaving are fathomless.
link |
02:12:19.400
And all of those emotions all live in the brains of humans.
link |
02:12:25.600
In the eye of the beholder, no, seriously, you have to embrace this concept of the eye
link |
02:12:31.040
of the beholder.
link |
02:12:33.040
It's the conceptualization that nothing takes meaning with one person creating it.
link |
02:12:39.720
Everything takes meaning in the receiving end.
link |
02:12:42.720
The emergent properties of these communication networks where every single, if you look at
link |
02:12:49.880
the network of our cells and how they're communicating with each other, every cell has its own code.
link |
02:12:54.280
This code is modulated by the epigenome.
link |
02:12:56.280
This creates a bunch of different cell types.
link |
02:12:58.080
Each cell type now has its own identity, yet they all have the common root of the stem
link |
02:13:01.800
cells that sort of led to them.
link |
02:13:04.880
Each of these identities is now communicating with each other.
link |
02:13:08.240
They take meaning in their interaction.
link |
02:13:11.920
There's an emergent property that comes from a bunch of cells being together that is not
link |
02:13:16.160
in any one of the parts.
link |
02:13:17.480
If you look at neurons communicating, again, these engrams don't exist in any one neuron.
link |
02:13:23.400
They exist in the connection, in the combination of neurons.
link |
02:13:26.600
And the meaning of the words that I'm telling you is empty until it reaches you and it affects
link |
02:13:33.480
you in a very different way than it affects whoever's listening to this conversation now.
link |
02:13:38.920
Because of the emotional baggage that I've grown up with, that you've grown up with,
link |
02:13:42.360
and that they've grown up with.
link |
02:13:44.480
And that's, I think, the magic of translation.
link |
02:13:47.880
If you start thinking of translation as just simply capturing that emotional set of reactions
link |
02:13:56.560
that you evoke, you need a different set of words to evoke that same set of reactions
link |
02:14:02.560
to a French person than to a Russian person because of the baggage of the culture that
link |
02:14:07.400
we grew up in.
link |
02:14:08.400
Yeah.
link |
02:14:09.400
So basically, you shouldn't find the best word.
link |
02:14:13.800
Sometimes it's a completely different sentence structure that you will need matched to the
link |
02:14:19.440
cultural context of the target audience that you have.
link |
02:14:24.760
I usually don't think about this, but right now there's this feeling, as a reminder, there's
link |
02:14:30.600
just you and I talking, but there's several hundred thousand people who will listen to
link |
02:14:35.720
this.
link |
02:14:36.720
There's a guy in Russia right now running, like in Moscow, listening to us.
link |
02:14:44.160
There's somebody in India, I guarantee you, there's somebody in China and South America.
link |
02:14:48.480
There's somebody in Texas, and they all have different emotional baggage.
link |
02:14:54.040
They probably got angry earlier on about the whole discussion about coronavirus and about
link |
02:15:00.320
some aspect of it.
link |
02:15:02.000
Yeah, and there's that network effect.
link |
02:15:06.960
It's a beautiful thing, and these lateral transfer of information, that's what makes
link |
02:15:11.240
the collective, quote unquote, genome of humanity so unique from any other species.
link |
02:15:19.960
So you somehow miraculously wrapped it back to the very beginning of when we were talking
link |
02:15:25.040
about the beauty of the human genome.
link |
02:15:29.160
So I think this is the right time, unless we want to go for a six to eight hour conversation.
link |
02:15:34.880
We're going to have to talk again, but I think for now, to wrap it up, this is the right
link |
02:15:39.640
time to talk about the biggest, most ridiculous question of all, meaning of life.
link |
02:15:46.000
Off mic, you mentioned to me that you had your 42nd birthday, 42nd being a very special,
link |
02:15:56.120
absurdly special number, and you had to kind of get together with friends to discuss the
link |
02:16:03.560
meaning of life.
link |
02:16:04.560
So let me ask you, as a biologist, as a computer scientist, and as a human, what is the meaning
link |
02:16:13.840
of life?
link |
02:16:15.640
I've been asking this question for a long time, ever since my 42nd birthday, but well
link |
02:16:21.440
before that, in even planning the meaning of life symposium, and symposium, symp means
link |
02:16:29.080
together, posi actually means to drink together, so symposium is actually a drinking party.
link |
02:16:34.560
Can you actually elaborate about this meaning of life symposium that you put together?
link |
02:16:39.560
It's like the most genius idea I've ever heard.
link |
02:16:42.360
So 42 is obviously the answer to life, the universe, and everything from the Hitchhackers
link |
02:16:46.120
guy to the galaxy.
link |
02:16:48.120
And as I was turning 42, I've had the theme for every one of my birthdays.
link |
02:16:51.840
When I was turning 32, it's 10000 in binary, so I celebrated my 100,000th binary birthday,
link |
02:17:00.160
and I had a theme of going back 100,000 years, you know, let's dress something in the last
link |
02:17:06.440
100,000 years.
link |
02:17:07.440
Anyway, I've always had these...
link |
02:17:09.640
It's such an interesting human being.
link |
02:17:12.240
Okay, that's awesome.
link |
02:17:13.240
I've always had these sort of numerology related announcements for my birthday party.
link |
02:17:21.840
So what came out of that meaning of life symposium is that I basically asked 42 of my colleagues,
link |
02:17:29.520
42 of my friends, 42 of my collaborators to basically give seven minute species on the
link |
02:17:36.280
meaning of life, each from their perspective.
link |
02:17:38.840
And I really encourage you to go there because it's mind boggling that every single person
link |
02:17:44.120
said a different answer.
link |
02:17:46.400
Every single person started with, I don't know what the meaning of life is, but, and
link |
02:17:51.000
then give this beautifully eloquently answer, eloquently answer.
link |
02:17:55.840
And they were all different, but they all were consistent with each other and mutually
link |
02:18:01.960
synergistic and together forming a beautiful view of what it means to be human in many
link |
02:18:07.080
ways.
link |
02:18:08.760
Some people talked about the loss of their loved one, their life partner for many, many
link |
02:18:13.680
years, and how their life changed through that.
link |
02:18:16.680
Some people talked about the origin of life.
link |
02:18:19.400
Some people talked about the difference between purpose and meaning.
link |
02:18:24.160
I'll maybe quote one of the answers, which is this linguistics professor, a friend of
link |
02:18:31.120
mine at Harvard, who basically said that she was gonna...
link |
02:18:36.720
She's Greek as well, and she said, I will give a very Pythian answer.
link |
02:18:40.160
So Pythia was the oracle of Delphi, who would basically give these very cryptic answers,
link |
02:18:45.400
very short, but interpretable in many different ways.
link |
02:18:47.920
There was this whole set of priests who were tasked with interpreting what Pythia had said,
link |
02:18:53.840
and very often you would not get a clean interpretation, but she said, I will be like Pythian, give
link |
02:18:59.160
you a very short and multiply interpretable answer, but unlike her, I will actually also
link |
02:19:04.400
give you three interpretations.
link |
02:19:07.120
And she said, the answer to the meaning of life is become one.
link |
02:19:12.920
And the first interpretation is, like a child, become one year old with the excitement of
link |
02:19:18.720
discovering everything about the world.
link |
02:19:21.640
Second interpretation, in whatever you take on, become one, the first, the best, excel,
link |
02:19:29.240
drive yourself to perfection for every one of your tasks, and become one when people
link |
02:19:36.440
are separate, become one, come together, learn to understand each other.
link |
02:19:42.960
Damn, that's an answer.
link |
02:19:45.720
And one way to summarize this whole meaning of life symposium is that the very symposium
link |
02:19:51.280
was illustrating the quest for meaning, which might itself be the meaning of life.
link |
02:19:58.240
This constant quest for something sublime, something human, something intangible, some
link |
02:20:06.520
aspect of what defines us as a species and as an individual, both a quest of me as a
link |
02:20:13.720
person through my own life, but the meaning of life could also be the meaning of all of
link |
02:20:20.200
life, what is the whole point of life, why life, why life itself, because we've been
link |
02:20:24.800
talking about the history and evolution of life, but we haven't talked about why life
link |
02:20:30.080
in the first place, is life inevitable, is life part of physics, does life transcend
link |
02:20:36.680
physics, but by fighting against entropy, by compartmentalizing and increasing concentrations
link |
02:20:42.880
rather than diluting away, is life a distinct entity in the universe beyond the traditional
link |
02:20:53.160
very simple physical rules that govern gravity and electromagnetism and all of these forces,
link |
02:21:00.840
is life another force, is there a life force, is there a unique kind of set of principles
link |
02:21:05.400
that emerge, of course, built on top of the hardware of physics, but is it sort of a new
link |
02:21:10.000
layer of software or a new layer of a computer system?
link |
02:21:14.720
So that's at the level of big questions.
link |
02:21:18.560
There's another aspect of gratitude, of basically what I like to say is during this pandemic,
link |
02:21:27.920
I've basically worked from 6 a.m. until 7 p.m. every single day nonstop, including Saturday
link |
02:21:33.520
and Sunday.
link |
02:21:34.520
I've basically broken all boundaries of where life, personal life begins and work life ends.
link |
02:21:42.400
And that has been exhilarating for me, just the intellectual pleasure that I get from
link |
02:21:51.160
a day of exhaustion, where at the end of the day, my brain is hurting, I'm telling my wife,
link |
02:21:57.080
wow, I was useful today.
link |
02:22:00.800
And there's a certain pleasure that comes from feeling useful.
link |
02:22:08.480
And there's a certain pleasure that comes from feeling grateful.
link |
02:22:12.560
So I've written this little sort of prayer for my kids to say at bedtime every night,
link |
02:22:19.600
where they basically say, thank you, God, for all you have given me and give me the strength
link |
02:22:26.800
to give unto others with the same love that you have given unto me.
link |
02:22:33.320
Me as a species are so special, the only ones who worry about the meaning of life.
link |
02:22:40.960
And maybe that's what makes us human.
link |
02:22:44.920
And what I like to say to my wife and to my students during this pandemic, work extravaganza,
link |
02:22:53.400
is every now and then they ask me, but how do you do this?
link |
02:22:56.440
And I'm like, I'm a workaholic.
link |
02:22:58.840
I love this.
link |
02:23:00.960
This is me in the most unfiltered way.
link |
02:23:04.840
The ability to do something useful, to feel that my brain is being used, to interact with
link |
02:23:10.560
the smartest people on the planet day in, day out, and to help them discover aspects
link |
02:23:15.800
of the human genome, of the human brain, of human disease and the human condition that
link |
02:23:22.040
no one has seen before with data that we're capturing that has never been observed.
link |
02:23:30.000
And that's another aspect which is on the personal life.
link |
02:23:34.640
Many people say, oh, I'm not going to have kids.
link |
02:23:36.200
Why bother?
link |
02:23:37.520
I can tell you as a father, they're missing half the picture, if not the whole picture.
link |
02:23:46.200
Teaching my kids about my view of the world and watching through their eyes the naivete
link |
02:23:53.120
with which they start and the sophistication with which they end up.
link |
02:23:58.440
The understanding that they have of not just the natural world around them, but of me too.
link |
02:24:06.600
The unfiltered criticism that you get from your own children that knows no bounds of
link |
02:24:16.760
honesty.
link |
02:24:19.400
And I've grown components of my heart that I didn't know I had until you sense that fragility
link |
02:24:28.000
that vulnerability of the children, that immense love and passion, the unfiltered egoism that
link |
02:24:41.320
we as adults learn how to hide so much better.
link |
02:24:44.480
It's just this back of emotions that tell me about the raw materials that make a human
link |
02:24:51.680
being and how these raw materials can be arranged with more sophistication that we learn through
link |
02:24:57.160
life to become truly human adults.
link |
02:25:02.280
But there's something so beautiful about seeing that progression between them, the complexity
link |
02:25:08.040
of the language growing as more neural connections are formed to realize that the hardware is
link |
02:25:16.720
getting rearranged as their software is getting implemented on that hardware.
link |
02:25:22.760
But their frontal cortex continues to grow for another 10 years.
link |
02:25:28.080
There's neural connections that are continuing to form, new neurons that actually get replicated
link |
02:25:32.160
and formed.
link |
02:25:34.160
It's just incredible that we have these, not just you grow the hardware for 30 years and
link |
02:25:40.760
then you feed it all of the knowledge, no, no, the knowledge is fed throughout and is
link |
02:25:45.420
shaping these neural connections as they're forming.
link |
02:25:48.640
So seeing that transformation from either your own blood or from an adopted child is
link |
02:25:54.680
the most beautiful thing you can do as a human being.
link |
02:25:57.640
And it completes you, it completes that path, that journey, the create life, oh sure, that's
link |
02:26:03.320
that conception, that's easy, but create human life to add the human part that takes decades
link |
02:26:09.840
of compassion, of sharing, of love and of anger and of impatience and patience.
link |
02:26:19.000
And as a parent, I think I've become a very different kind of teacher.
link |
02:26:26.160
Because again, I'm a professor, my first role is to bring adult human beings into a more
link |
02:26:32.600
mature level of adulthood where they learn not just to do science, but they learn the
link |
02:26:38.600
process of discovery and the process of collaboration, the process of sharing, the process of conveying
link |
02:26:43.840
the knowledge of encapsulating something incredibly complex and sort of giving it up in sort
link |
02:26:49.960
of bite sized chunks that the rest of humanity can appreciate.
link |
02:26:54.400
I tell my students all the time, when a tree falls in the forest and no one's there to
link |
02:27:01.360
listen, has it really fallen?
link |
02:27:03.080
The same way, you do this awesome research, if you write an impenetrable paper that no
link |
02:27:06.880
one will understand, it's as if you never did the awesome research.
link |
02:27:11.160
So conveying of knowledge, conveying this lateral transfer that I was talking about
link |
02:27:15.840
at the very beginning of sort of humanity and sort of the sharing of information, all
link |
02:27:22.600
of that has gotten so much more rich by seeing human beings grow in my own home.
link |
02:27:32.560
Because that makes me a better parent and that makes me a better teacher and a better
link |
02:27:37.300
mentor to the nurturing of my adult children, which are my research group.
link |
02:27:43.400
First of all, beautifully put, connects beautifully to the vertical and the horizontal inheritance
link |
02:27:51.520
of ideas that we talked about at the very beginning.
link |
02:27:54.320
I don't think there's a better way to end it on this poetic and powerful note.
link |
02:28:00.840
Manolis, thank you so much for talking to us.
link |
02:28:03.040
A huge honor.
link |
02:28:04.040
We'll have to talk again about the origin of life, about epigenetics, epigenomics, and
link |
02:28:11.960
some of the incredible research you're doing.
link |
02:28:13.720
Truly an honor.
link |
02:28:14.720
Thanks so much for talking to us.
link |
02:28:15.720
Thank you.
link |
02:28:16.720
Such a pleasure.
link |
02:28:17.720
It's such a pleasure.
link |
02:28:18.720
I mean, your questions are outstanding.
link |
02:28:19.720
I've had such a blast here.
link |
02:28:20.720
I can't wait to be back.
link |
02:28:21.720
Awesome.
link |
02:28:22.720
Thanks for listening to this conversation with Manolis Kellis.
link |
02:28:26.080
And thank you to our sponsors, Blinkist, Aidsleep, and Masterclass.
link |
02:28:31.520
Please consider supporting this podcast by going to blinkist.com slash lex, aidsleep.com
link |
02:28:36.520
slash lex, and masterclass.com slash lex.
link |
02:28:41.280
Click the links by the stuff.
link |
02:28:43.400
Get the discount.
link |
02:28:44.400
It's the best way to support this podcast.
link |
02:28:46.440
If you enjoy this thing, subscribe on YouTube, review it with five stars on Apple Podcast.
link |
02:28:51.280
Support it on Patreon or connect with me on Twitter at Lex Freedman.
link |
02:28:55.560
And now let me leave you with some words from Charles Darwin that I think Manolis represents
link |
02:29:00.600
quite beautifully.
link |
02:29:01.600
If I had my life to live over again, I would have made a rule to read some poetry and listen
link |
02:29:07.920
to some music at least once every week.
link |
02:29:11.800
Thank you for listening and hope to see you next time.