Transcript
00:00:00 The following is a conversation with Manolis Kellis.
00:00:03 He’s a professor at MIT and head
00:00:05 of the MIT Computational Biology Group.
00:00:08 He’s interested in understanding the human genome
00:00:11 from a computational, evolutionary, biological,
00:00:14 and other cross disciplinary perspectives.
00:00:17 He has more big, impactful papers and awards
00:00:20 than I can list, but most importantly,
00:00:22 he’s a kind, curious, brilliant human being,
00:00:26 and just someone I really enjoy talking to.
00:00:28 His passion for science and life in general is contagious.
00:00:32 The hours honestly flew by,
00:00:34 and I’m sure we’ll talk again on this podcast soon.
00:00:37 Quick summary of the ads.
00:00:39 Three sponsors, Blinkist, Aidsleep, and Masterclass.
00:00:43 Please consider supporting this podcast
00:00:44 by going to blinkist.com slash lex,
00:00:47 aidsleep.com slash lex,
00:00:49 and signing up at masterclass.com slash lex.
00:00:53 Click the links, buy the stuff, get the discount.
00:00:56 It’s the best way to support this podcast.
00:00:58 If you enjoy this thing, subscribe on YouTube,
00:01:01 review it with five stars on Apple Podcast,
00:01:03 support it on Patreon,
00:01:04 or connect with me on Twitter at Lex Friedman.
00:01:08 As usual, I’ll do a few minutes of ads now,
00:01:10 and never any ads in the middle
00:01:11 that can break the flow of the conversation.
00:01:14 This episode is supported by Blinkist,
00:01:17 my favorite app for learning new things.
00:01:19 Get it at blinkist.com slash lex
00:01:22 for a seven day free trial and 25% off afterwards.
00:01:26 Blinkist takes the key ideas
00:01:28 from thousands of nonfiction books
00:01:30 and condenses them down into just 15 minutes
00:01:33 that you can read or listen to.
00:01:35 I’m a big believer in reading at least an hour every day.
00:01:38 As part of that, I use Blinkist every day
00:01:41 to try out a book I may otherwise
00:01:43 never have a chance to read.
00:01:45 And in general, it’s a great way to broaden your view
00:01:48 of the ideal landscape out there,
00:01:50 and find books that you may want to read more deeply.
00:01:53 With Blinkist, you get unlimited access
00:01:56 to read or listen to a massive library
00:01:58 of condensed nonfiction books.
00:02:00 Go to blinkist.com slash lex to try it free for seven days
00:02:05 and save 25% off your new subscription.
00:02:08 That’s blinkist.com slash lex.
00:02:10 Blinkist, spelled B L I N K I S T.
00:02:15 This show is also sponsored by Asleep
00:02:18 and it’s pod pro mattress.
00:02:19 You can check out at Asleep.com slash lex
00:02:22 to get $200 off.
00:02:24 It controls temperature with a nap
00:02:26 and can cool down to as low as 55 degrees
00:02:29 on each side of the bed separately.
00:02:31 Research shows that temperature has a big impact
00:02:33 on the quality of our sleep.
00:02:35 Anecdotally, as been true for me,
00:02:37 it’s truly been a game changer.
00:02:39 I love it.
00:02:40 The pod pro is packed with sensors
00:02:42 that track heart rate, heart rate variability
00:02:44 and respiratory rate, showing it all in their app.
00:02:48 The app’s health metrics are amazing
00:02:50 but the cooling alone is honestly worth the money.
00:02:52 Check it out at Asleep.com slash lex to get $200 off.
00:02:57 This show is also sponsored by Masterclass.
00:03:00 Sign up at Masterclass.com slash lex
00:03:02 to get a discount and to support this podcast.
00:03:05 When I first heard about Masterclass,
00:03:07 I thought it was too good to be true.
00:03:09 For 180 bucks a year, you get an all access pass
00:03:12 to watch courses from, to list some of my favorites,
00:03:15 Chris Hadfield on Space Exploration,
00:03:17 Neil deGrasse Tyson on Scientific Thinking
00:03:19 and Communication, Will Wright,
00:03:21 one of my favorite game designers,
00:03:23 Carlos Santana, one of my favorite guitar players,
00:03:26 Garry Kasparov, of course,
00:03:27 the greatest chess player of all time, I’m not biased,
00:03:30 Daniel Negrano on poker and many more.
00:03:33 Chris Hadfield explaining how rockets work
00:03:35 and the experience of being launched into space alone
00:03:38 is worth the money.
00:03:40 By the way, you can watch it on basically any device.
00:03:43 Once again, sign up at Masterclass.com slash lex
00:03:46 to get a discount and to support this podcast.
00:03:49 And now, here’s my conversation with Manolis Kellis.
00:03:54 What to you is the most beautiful aspect
00:03:56 of the human genome?
00:03:58 Don’t get me started.
00:04:00 So. We’ve got time.
00:04:04 The first answer is that the beauty of genomes
00:04:06 transcends humanity.
00:04:07 So it’s not just about the human genome.
00:04:09 Genomes in general are amazingly beautiful.
00:04:12 And again, I’m obviously biased.
00:04:14 So in my view, the way that I like to introduce
00:04:18 the human genome and the way that I like to introduce
00:04:20 genomics to my class is by telling them,
00:04:22 you know, we’re not the inventors
00:04:24 of the first digital computer.
00:04:26 We are the descendants of the first digital computer.
00:04:30 Basically, life is digital.
00:04:32 And that’s absolutely beautiful about life.
00:04:34 The fact that at every replication step,
00:04:37 you don’t lose any information
00:04:38 because that information is digital.
00:04:40 If it was analog, if it was just sprouting concentrations,
00:04:43 you’d lose it after a few generations.
00:04:44 It would just dissolve away.
00:04:46 And that’s what the ancients
00:04:48 didn’t understand about inheritance.
00:04:50 The first person to understand digital inheritance
00:04:52 was Mendel, of course.
00:04:54 And his theory, in fact, stayed in a bookshelf
00:04:57 for like 50 years while Darwin was getting famous
00:05:00 about natural selection.
00:05:02 But the missing component was this digital inheritance,
00:05:05 the mechanism of evolution that Mendel had discovered.
00:05:09 So that aspect in my view is the most beautiful aspect
00:05:13 but it transcends all of life.
00:05:14 And can you elaborate maybe the inheritance part?
00:05:18 What was the key thing that the ancients didn’t understand?
00:05:22 So the very theory of inheritance as discrete units,
00:05:28 throughout the life of Mendel and well after he’s writing,
00:05:32 people thought that his P experiments
00:05:35 were just a little fluke,
00:05:36 that they were just a little exception
00:05:38 that would normally not even apply to humans,
00:05:41 that basically what they saw
00:05:44 is this continuum of eye color,
00:05:48 this continuum of skin color,
00:05:49 this continuum of hair color,
00:05:51 this continuum of height.
00:05:52 And all of these continuums did not fit
00:05:55 with a discrete type of inheritance
00:05:56 that Mendel was describing.
00:05:58 But what’s unique about genomics
00:06:00 and what’s unique about the genome
00:06:01 is really that there are two copies
00:06:03 and that you get a combination of these.
00:06:06 But for every trait,
00:06:08 there are dozens of contributing variables.
00:06:10 And it was only Ronald Fisher in the 20th century
00:06:14 that basically recognized that even five Mendelian traits
00:06:20 would add up to a continuum like inheritance pattern.
00:06:24 And he wrote a series of papers
00:06:27 that still are very relevant today
00:06:30 about sort of this Mendelian inheritance
00:06:32 of continuum like traits.
00:06:35 And I think that that was the missing step in inheritance.
00:06:38 So well before the discovery of the structure of DNA,
00:06:41 which is again, another amazingly beautiful aspect,
00:06:44 the double helix,
00:06:45 what I like to call the most noble molecule of our time,
00:06:50 holds within it the secret of that discrete inheritance,
00:06:54 but the conceptualization of discrete elements
00:06:58 is something that precedes that.
00:06:59 So even though it’s discrete,
00:07:01 when it materializes itself into actual traits that we see,
00:07:06 it can be continuous.
00:07:08 Basically arbitrarily rich and complex.
00:07:10 So if you have five genes that contribute to human height,
00:07:15 and there aren’t five, there’s a thousand.
00:07:16 If there’s only five genes
00:07:18 and you inherit some combination of them,
00:07:20 and every one makes you two inches taller
00:07:23 or two inches shorter,
00:07:24 it’ll look like a continuous trait.
00:07:28 But instead of five, there are thousands.
00:07:30 And every one of them contributes to less than one millimeter.
00:07:33 We change in height more during the day
00:07:36 than each of these genetic variants contributes.
00:07:40 So by the evening, you’re shorter than you walk up with.
00:07:43 Isn’t that weird then
00:07:45 that we’re not more different than we are?
00:07:48 Why are we all so similar
00:07:49 if there’s so much possibility to be different?
00:07:52 Yeah, so there are selective advantages to being medium.
00:07:57 If you’re extremely tall or extremely short,
00:07:59 you run into selective disadvantages.
00:08:02 So you have trouble breathing, you have trouble running,
00:08:04 you have trouble sitting if you’re too tall.
00:08:06 If you’re too short, you might, I don’t know,
00:08:08 have other selective pressures are acting against that.
00:08:11 If you look at natural history of human population,
00:08:13 there’s actually selection for height in Northern Europe
00:08:17 and selection against height in Southern Europe.
00:08:19 So there might actually be advantages
00:08:21 to actually being not super tall.
00:08:25 And if you look across the entire human population,
00:08:28 for many, many traits,
00:08:29 there’s a lot of push towards the middle.
00:08:32 Balancing selection is the usual term
00:08:35 for selection that sort of seeks to not be extreme
00:08:39 and to sort of have a combination of alleles
00:08:43 that sort of keep recombining.
00:08:45 And if you look at mate selection,
00:08:48 super, super tall people
00:08:50 will not tend to sort of marry super, super tall people.
00:08:53 Very often you see these couples
00:08:55 that are kind of compensating for each other.
00:08:57 And the best predictor of the kid’s age
00:09:00 is very often just take the average of the two parents
00:09:03 and then adjust for sex and boom, you get it.
00:09:07 It’s extremely heritable.
00:09:08 Let me ask, you kind of took a step back to the genome
00:09:12 outside of just humans,
00:09:13 but is there something that you find beautiful
00:09:15 about the human genome specifically?
00:09:18 So I think the genome,
00:09:21 if more people understood the beauty of the human genome,
00:09:24 there would be so many fewer wars,
00:09:26 so much less anger in the world.
00:09:28 I mean, what’s really beautiful about the human genome
00:09:31 is really the variation
00:09:33 that teaches us both about individuality
00:09:36 and about similarity.
00:09:38 So any two people on the planet are 99.9% identical.
00:09:43 How can you fight with someone who’s 99.9% identical to you?
00:09:47 It’s just counterintuitive.
00:09:49 And yet any two siblings of the same parents
00:09:53 differ in millions of locations.
00:09:57 So every one of them is basically two to the million unique
00:10:01 from any pair of parents,
00:10:03 let alone any two random parents on the planet.
00:10:05 So that’s, I think, something that teaches us
00:10:08 about sort of the nature of humanity in many ways,
00:10:11 that every one of us is as unique as any star
00:10:14 and way more unique in actually many ways.
00:10:17 And yet we’re all brothers and sisters.
00:10:22 Yeah, just like stars, most of it is just fusion reactions.
00:10:26 Yeah, you only have a few parameters to describe stars.
00:10:29 Mass, size, initial size, and stage of life.
00:10:33 Whereas for humans, it’s thousands of parameters
00:10:36 scattered across our genome.
00:10:38 So the other thing that makes humans unique,
00:10:41 the other things that makes inheritance unique in humans
00:10:45 is that most species inherit things vertically.
00:10:50 Basically instinct is a huge part of their behavior.
00:10:54 The way that, I mean, with my kids,
00:10:57 we’ve been watching this nest of birds
00:11:00 with two little eggs outside our window
00:11:03 for the last few months,
00:11:05 for the last few weeks as they’ve been growing.
00:11:07 And there’s so much behavior that’s hard coded.
00:11:12 Birds don’t just learn as they grow.
00:11:16 There’s no culture.
00:11:17 Like a bird that’s born in Boston
00:11:19 will be the same as a bird that’s born in California.
00:11:22 So there’s not as much inheritance of ideas, of customs.
00:11:27 A lot of it is hard coding in their genome.
00:11:30 What’s really beautiful about the human genome
00:11:32 is that if you take a person from today
00:11:35 and you place them back in ancient Egypt,
00:11:37 or if you take a person from ancient Egypt
00:11:39 and you place them here today,
00:11:41 they will grow up to be completely normal.
00:11:44 That is not genetics.
00:11:47 This is the other type of inheritance in humans.
00:11:51 So on one hand, we have the genetic inheritance,
00:11:53 which is vertical from your parents down.
00:11:56 On the other hand, we have horizontal inheritance,
00:11:58 which is the ideas that are built up at every generation
00:12:02 are horizontally transmitted.
00:12:04 And the huge amount of time
00:12:06 that we spend in educating ourselves,
00:12:09 a concept known as neoteny,
00:12:11 neo for newborn and then teny for holding.
00:12:15 So if you look at humans,
00:12:17 I mean, the little birds that were eggs two weeks ago,
00:12:20 and now one of them has already flown off.
00:12:22 The other one’s ready to fly off.
00:12:24 In two weeks, they’re ready to just fend for themselves.
00:12:27 Humans, 16 years, 18 years, 24, getting out of college.
00:12:33 I’m still learning.
00:12:34 So that’s so fascinating,
00:12:36 this picture of a vertical and the horizontal.
00:12:38 When you talk about the horizontal,
00:12:40 is it in the realm of ideas?
00:12:41 Exactly.
00:12:42 Okay, so it’s the actual social interactions.
00:12:45 That’s exactly right.
00:12:46 That’s exactly right.
00:12:47 So basically the concept of neoteny
00:12:49 is that you spend acquiring characteristics
00:12:52 from your environment
00:12:54 in an extremely malleable state of your brain
00:12:56 and the wiring of your brain for a long period of your life.
00:13:00 Compared to primates, we are useless.
00:13:03 You take any primate at seven weeks
00:13:05 and any human at seven weeks, we lose the battle.
00:13:08 But at 18 years, you know, all bets are off.
00:13:11 Like we basically, our brain continues to develop
00:13:14 in an extremely malleable form till very late.
00:13:17 And this is what allows education.
00:13:20 This is what allows the person from Egypt
00:13:22 to do extremely well now.
00:13:24 And the reason for that is that the wiring of our brain
00:13:31 and the development of that wiring is actually delayed.
00:13:34 So, you know, the longer you delay that,
00:13:37 the more opportunity you have to pass on knowledge,
00:13:40 to pass on concepts, ideals, ideas
00:13:44 from the parents to the child.
00:13:46 And what’s really absolutely beautiful about humans today
00:13:49 is that that lateral transfer of ideas and culture
00:13:52 is not just from uncles and aunts and teachers at school,
00:13:55 but it’s from Wikipedia and review articles on the web
00:14:00 and thousands of journals
00:14:02 that are sort of putting out information for free
00:14:05 and podcasts and videocasts and all of that stuff
00:14:08 where you can basically learn about any topic,
00:14:12 pretty much everything that would be in any
00:14:16 super advanced textbook in a matter of days,
00:14:19 instead of having to go to the library of Alexandria
00:14:22 and sail there to read three books
00:14:24 and then sail for another few days to get to Athens
00:14:27 and et cetera, et cetera, et cetera.
00:14:28 So the democratization of knowledge
00:14:31 and the spread, the speed of spread of knowledge
00:14:34 is what defines, I think, the human inheritance pattern.
00:14:38 So you sound excited about it, are you also a little bit
00:14:43 afraid or are you more excited by the power
00:14:46 of this kind of distributed spread of information?
00:14:49 So you put it very kindly that most people
00:14:52 are kind of using the internet and looking Wikipedia,
00:14:55 reading articles, reading papers and so on,
00:14:58 but if we’re honest, most people online,
00:15:02 especially when they’re younger,
00:15:03 probably looking at five second clips on TikTok
00:15:05 or whatever the new social network is,
00:15:08 are you, given this power of horizontal inheritance,
00:15:12 are you optimistic or a little bit pessimistic
00:15:16 about this new effect of the internet
00:15:22 and democratization of knowledge on our,
00:15:26 what would you call this, this genome,
00:15:29 would you use the term genome, by the way, for this?
00:15:31 Yeah, I think we use the genome to talk about DNA,
00:15:36 but very often we say, I’m Greek,
00:15:38 so people ask me, hey, what’s in the Greek genome?
00:15:40 And I’m like, well, yeah, what’s in the Greek genome
00:15:42 is both our genes and also our ideas
00:15:44 and our ideals and our culture.
00:15:46 So the poetic meaning of the word.
00:15:48 Exactly, exactly, yeah.
00:15:50 So I think that there’s a beauty
00:15:55 to the democratization of knowledge,
00:15:57 the fact that you can reach as many people
00:16:00 as any other person on the planet
00:16:02 and it’s not who you are,
00:16:04 it’s really your ideas that matter,
00:16:06 is a beautiful aspect of the internet.
00:16:11 I think there’s, of course, a danger of my ignorance
00:16:15 is as important as your expertise.
00:16:18 The fact that with this democratization
00:16:21 comes the abolishment of respecting expertise.
00:16:25 Just because you’ve spent 10,000 hours of your life
00:16:28 studying, I don’t know, human brain circuitry,
00:16:33 why should I trust you?
00:16:34 I’m just gonna make up my own theories
00:16:35 and they’ll be just as good as yours,
00:16:37 is an attitude that sort of counteracts
00:16:39 the beauty of the democratization.
00:16:42 And I think that within our educational system
00:16:47 and within the upbringing of our children,
00:16:49 we have to not only teach them knowledge,
00:16:52 but we have to teach them the means to get to knowledge.
00:16:55 And that, it’s very similar to sort of you fish,
00:16:59 you catch a fish for a man for one day,
00:17:01 you fed them for one day, you teach them how to fish,
00:17:03 you fed them for the rest of their life.
00:17:05 So instead of just gathering the knowledge
00:17:07 they need for any one task,
00:17:09 we can just tell them, all right,
00:17:11 here’s how you Google it,
00:17:12 here’s how you figure out what’s real and what’s not,
00:17:14 here’s how you check the sources,
00:17:16 here’s how you form a basic opinion for yourself.
00:17:19 And I think that inquisitive nature
00:17:22 is paramount to being able to sort through
00:17:26 this huge wealth of knowledge.
00:17:29 So you need a basic educational foundation
00:17:32 based on which you can then add on
00:17:35 the sort of domain specific knowledge,
00:17:38 but that basic educational foundation
00:17:39 should just not just be knowledge,
00:17:42 but it should also be epistemology,
00:17:45 the way to acquire knowledge.
00:17:47 I’m not sure any of us know how to do that
00:17:49 in this modern day, we’re actually learning.
00:17:51 One of the big surprising thing to me
00:17:53 about the coronavirus, for example,
00:17:57 is that Twitter has been
00:17:59 one of the best sources of information.
00:18:02 Basically like building your own network of experts,
00:18:07 as opposed to the traditional centralized expertise
00:18:11 of the WHO and the CDC,
00:18:13 or maybe any one particular respectable person
00:18:19 at the top of a department in some kind of institution,
00:18:21 you instead look at 10, 20, hundreds of people,
00:18:26 some of whom are young kids that are incredibly good
00:18:32 at aggregating data and plotting and visualizing that data.
00:18:35 That’s been really surprising to me.
00:18:37 I don’t know what to make of it.
00:18:39 I don’t know how that matures into something stable.
00:18:45 I don’t know if you have ideas.
00:18:47 If you were to just try to explain to your kids
00:18:49 of where should you go to learn about coronavirus,
00:18:54 what would you say?
00:18:56 It’s such a beautiful example.
00:18:58 And I think the current pandemic
00:18:59 and the speed at which the scientific community has moved
00:19:03 in the current pandemic,
00:19:04 I think exemplifies this horizontal transfer
00:19:08 and the speed of horizontal transfer of information.
00:19:10 The fact that the genome was first sequenced
00:19:15 in early January,
00:19:16 the first sample was obtained December 29, 2019,
00:19:20 a week after the publication of the first genome sequence,
00:19:23 Moderna had already finalized its vaccine design
00:19:27 and was moving to production.
00:19:29 I mean, this is phenomenal.
00:19:31 The fact that we go from not knowing
00:19:34 what the heck is killing people in Wuhan
00:19:36 to wow, it’s SARS CoV2 and here’s the set of genes,
00:19:41 here’s the genome, here’s the sequence,
00:19:43 here are the polymorphisms, et cetera,
00:19:45 in the matter of weeks is phenomenal.
00:19:48 In that incredible pace of transfer of knowledge,
00:19:52 there have been many mistakes.
00:19:54 So, some of those mistakes
00:19:56 may have been politically motivated
00:19:57 or other mistakes may have just been innocuous errors.
00:20:00 Others may have been misleading the public
00:20:02 for the greater good, such as don’t wear masks
00:20:05 because we don’t want the mask to run out.
00:20:07 I mean, that was very silly in my view
00:20:09 and a very big mistake.
00:20:11 But the spread of knowledge
00:20:15 from the scientific community was phenomenal.
00:20:17 And some people will point out to bogus articles
00:20:20 that snuck in and made the front page.
00:20:22 Yeah, they did.
00:20:23 But within 24 hours, they were debunked
00:20:26 and went out of the front page.
00:20:27 And I think that’s the beauty of science today.
00:20:30 The fact that it’s not, oh, knowledge is fixed.
00:20:33 It’s the ability to embrace that nothing is permanent
00:20:36 when it comes to knowledge,
00:20:37 that everything is the current best hypothesis
00:20:40 and the current best model that best fits the current data
00:20:42 and the willingness to be wrong.
00:20:45 The expectation that we’re gonna be wrong
00:20:48 and the celebration of success based on
00:20:50 how long was I not proven wrong for,
00:20:52 rather than, wow, I was exactly right.
00:20:55 Because no one is gonna be exactly right
00:20:57 with partial knowledge.
00:20:58 But the arc towards perfection,
00:21:03 I think is so much more important
00:21:05 than how far you are in your first step.
00:21:08 And I think that’s what sort of
00:21:10 the current pandemic has taught us.
00:21:13 The fact that, yeah, no, of course,
00:21:14 we’re gonna make mistakes,
00:21:16 but at least we’re gonna learn from those mistakes
00:21:18 and become better and learn better
00:21:20 and spread information better.
00:21:21 So if I were to answer the question of,
00:21:23 where would you go to learn about coronavirus?
00:21:27 First textbook, it all starts with a textbook.
00:21:29 Just open up a chapter on virology
00:21:32 and how coronaviruses work.
00:21:34 Then some basic epidemiology
00:21:36 and sort of how pandemics have worked in the past.
00:21:39 What are the basic principles surrounding
00:21:41 these first wave, second wave?
00:21:43 Why do they even exist?
00:21:45 Then understanding about growth,
00:21:47 understanding about the R0 and RT
00:21:50 at various time points.
00:21:52 And then understanding the means of spread,
00:21:55 how it spreads from person to person.
00:21:57 Then how does it get into your cells?
00:22:00 From when it gets into the cells,
00:22:01 what are the paths that it takes?
00:22:03 What are the cell types that express
00:22:05 the particular ACE2 receptor?
00:22:07 How is your immune system interacting with the virus?
00:22:09 And once your immune system launches a defense,
00:22:12 how is that helping or actually hurting your health?
00:22:15 What about the cytokine storm?
00:22:16 What are most people dying from?
00:22:18 Why are the comorbidities
00:22:20 and these risk factors even applying?
00:22:23 What makes obese people respond more
00:22:25 or elderly people respond more to the virus
00:22:28 while kids are completely,
00:22:32 very often not even aware that they’re spreading it?
00:22:36 So I think there’s some basic questions
00:22:41 that you would start from.
00:22:42 And then I’m sorry to say,
00:22:44 but Wikipedia is pretty awesome.
00:22:45 Yeah, it is. Google is pretty awesome.
00:22:47 It used to be a time,
00:22:48 it used to be a time maybe five years ago.
00:22:50 I forget when,
00:22:52 but people kind of made fun of Wikipedia
00:22:54 for being an unreliable source.
00:22:57 I never quite understood it.
00:22:58 I thought from the early days, it was pretty reliable
00:23:01 or better than a lot of the alternatives.
00:23:03 But at this point,
00:23:04 it’s kind of like a solid accessible survey paper
00:23:08 on every subject ever.
00:23:10 There’s an ascertainment bias and a writing bias.
00:23:14 So I think this is related to sort of people saying,
00:23:17 oh, so many nature papers are wrong.
00:23:20 And they’re like, why would you publish in nature?
00:23:22 So many nature papers are wrong.
00:23:23 And my answer is no, no, no.
00:23:26 So many nature papers are scrutinized.
00:23:29 And just because more of them are being proven wrong
00:23:31 than in other articles is actually evidence
00:23:35 that they’re actually better papers overall
00:23:37 because they’re being scrutinized at a rate
00:23:39 much higher than any other journal.
00:23:41 So if you basically judge Wikipedia
00:23:45 by not the initial content,
00:23:49 but by the number of revisions,
00:23:52 then of course it’s gonna be the best source
00:23:53 of knowledge eventually.
00:23:55 It’s still very superficial.
00:23:57 You then have to go into the review papers,
00:23:58 et cetera, et cetera, et cetera.
00:24:00 But I mean, for most scientific topics,
00:24:03 it’s extremely superficial,
00:24:05 but it is quite authoritative
00:24:07 because it is the place that everybody likes to criticize
00:24:10 as being wrong.
00:24:11 You say that it’s superficial.
00:24:13 And a lot of topics that I’ve studied a lot of,
00:24:18 I find it, I don’t know if superficial is the right word.
00:24:24 Because superficial kind of implies that it’s not correct.
00:24:27 No, no, no.
00:24:29 I don’t mean any implication of it not being correct.
00:24:31 It’s just superficial.
00:24:32 It’s basically only scratching the surface.
00:24:35 For depth, you don’t go to Wikipedia.
00:24:37 You go to the review articles.
00:24:38 But it can be profound in the way that articles rarely,
00:24:41 one of the frustrating things to me
00:24:43 about certain computer science,
00:24:46 like in the machine learning world,
00:24:48 articles, they don’t as often take the bigger picture view.
00:24:54 There’s a kind of data set and you show that it works
00:24:57 and you kind of show that here’s an architecture thing
00:24:59 that creates an improvement and so on and so forth.
00:25:02 But you don’t say, well, what does this mean
00:25:05 for the nature of intelligence for future data sets
00:25:08 we haven’t even thought about?
00:25:10 Or if you were trying to implement this,
00:25:11 like if we took this data set of 100,000 examples
00:25:15 and scale it to 100 billion examples with this method,
00:25:19 like look at the bigger picture,
00:25:21 which is what a Wikipedia article would actually try to do,
00:25:25 which is like, what does this mean in the context
00:25:28 of the broad field of computer vision or something like that?
00:25:32 Yeah, no, I agree with you completely, but it depends
00:25:35 on the topic.
00:25:36 I mean, for some topics, there’s been a huge amount of work.
00:25:38 For other topics, it’s just a stub.
00:25:40 So, you know.
00:25:41 I got it.
00:25:42 Yeah.
00:25:43 Well, yeah, actually the, which we’ll talk on,
00:25:46 genomics was not great.
00:25:48 Yeah, it’s very shallow, yeah, yeah.
00:25:50 It’s not wrong, it’s just shallow.
00:25:51 It’s shallow.
00:25:52 Yeah, every time I criticize something,
00:25:54 I should feel partly responsible.
00:25:56 Basically, if more people from my community went there
00:25:58 and edited, it would not be shallow.
00:26:01 It’s just that there’s different modes of communication
00:26:04 in different fields.
00:26:05 And in some fields, the experts have embraced Wikipedia.
00:26:08 In other fields, it’s relegated.
00:26:11 And perhaps the reason is that if it was any better
00:26:15 to start with, people would invest more time.
00:26:18 But if it’s not great to start with,
00:26:19 then you need a few initial pioneers who will basically
00:26:22 go in and say, ah, enough, we’re just gonna fix that.
00:26:26 And then I think it’ll catch on much more.
00:26:29 So if it’s okay, before we go on to genomics,
00:26:32 can we linger a little bit longer on the beauty
00:26:35 of the human genome?
00:26:37 You’ve given me a few notes.
00:26:38 What else do you find beautiful about the human genome?
00:26:41 So the last aspect of what makes the human genome unique,
00:26:44 in addition to the, you know, similarity and the differences
00:26:49 and the individuality is that, so very early on,
00:26:56 people would basically say, oh, you don’t do that
00:26:58 experiment in human, you have to learn about that in fly,
00:27:01 or you have to learn about that in yeast first,
00:27:03 or in mouse first, or in a primate first.
00:27:05 And the human genome was in fact relegated to sort of,
00:27:09 oh, the last place that you’re gonna go
00:27:11 to learn something new.
00:27:12 That has dramatically changed.
00:27:14 And the reason that changed is human genetics.
00:27:18 We are the species in the planet
00:27:22 that’s the most studied right now.
00:27:24 It’s embarrassing to say that,
00:27:26 but this was not the case a few years ago.
00:27:28 It used to be, you know, first viruses, then bacteria,
00:27:33 then yeast, then the fruit fly and the worm,
00:27:37 then the mouse, and eventually human was very far last.
00:27:42 So it’s embarrassing that it took us this long
00:27:44 to focus on it, or the…
00:27:46 It’s embarrassing that the model organisms
00:27:49 have been taken over because of the power of human genetics.
00:27:52 That right now, it’s actually simpler to figure out
00:27:55 the phenotype of something by mining
00:27:58 this massive amount of human data
00:28:01 than by going back to any of the other species.
00:28:04 And the reason for that is that if you look
00:28:05 at the natural variation that happens
00:28:07 in a population of seven billion,
00:28:09 you basically have a mutation in almost every nucleotide.
00:28:13 So every nucleotide you wanna perturb,
00:28:15 you can go find a living, breathing human being
00:28:18 and go test the function of that nucleotide
00:28:20 by sort of searching the database and finding that person.
00:28:22 Wait, why is that embarrassing?
00:28:23 It’s a beautiful data set.
00:28:24 It’s a beautiful data set.
00:28:26 It’s embarrassing for the model organism.
00:28:29 For the flies.
00:28:30 Yeah, exactly.
00:28:30 I mean, do you feel on a small tangent,
00:28:34 is there something of value in the genome of a fly
00:28:40 and other of these model organisms that you miss
00:28:43 that we wish we would be looking at deeper?
00:28:47 So directed perturbation, of course.
00:28:49 So I think the place where humans are still lagging
00:28:54 is the fact that in an animal model,
00:28:55 you can go and say,
00:28:56 well, let me knock out this gene completely
00:28:58 and let me knock out these three genes completely.
00:29:00 And the moment you get into combinatorics,
00:29:02 it’s something you can’t do in the human
00:29:04 because there just simply aren’t enough humans
00:29:05 on the planet.
00:29:07 And again, let me be honest,
00:29:08 we haven’t sequenced all seven billion people.
00:29:11 It’s not like we have every mutation,
00:29:12 but we know that there’s a carrier out there.
00:29:15 So if you look at the trend and the speed
00:29:17 with which human genetics has progressed,
00:29:19 we can now find thousands of genes involved
00:29:23 in human cognition, in human psychology,
00:29:27 in the emotions and the feelings
00:29:29 that we used to think are uniquely learned.
00:29:31 It turns out there’s a genetic basis to a lot of that.
00:29:34 So the human genome has continued to elucidate
00:29:42 through these studies of genetic variation,
00:29:44 so many different processes that we previously thought
00:29:47 were something like free will.
00:29:52 Free will is this beautiful concept
00:29:54 that humans have had for a long time.
00:29:58 In the end, it’s just a bunch of chemical reactions
00:29:59 happening in your brain.
00:30:00 And the particular abundance of receptors
00:30:03 that you have this day based on what you ate yesterday
00:30:06 or that you have been wired with based on your parents
00:30:10 and your upbringing, et cetera,
00:30:12 determines a lot of that quote unquote free will component
00:30:15 to sort of narrow and narrow sort of slices.
00:30:20 So how much on that point, how much freedom
00:30:24 do you think we have to escape the constraints
00:30:29 of our genome?
00:30:30 You’re making it sound like more and more
00:30:31 we’re discovering that our genome is actually has the,
00:30:35 a lot of the story already encoded into it.
00:30:37 How much freedom do we have?
00:30:39 I, so let me describe what that freedom would look like.
00:30:45 That freedom would be my saying,
00:30:47 ooh, I’m gonna resist the urge to eat that apple
00:30:51 because I choose not to.
00:30:54 But there are chemical receptors that made me
00:30:57 not resist the urge to prove my individuality
00:31:01 and my free will by resisting the apple.
00:31:04 So then the next question is,
00:31:05 well, maybe now I’ll resist the urge to resist the apple
00:31:08 and I’ll go for the chocolate instead
00:31:09 to prove my individuality.
00:31:10 But then what about those other receptors that, you know?
00:31:14 That might be all encoded in there.
00:31:17 So it’s kicking the bucket down the road
00:31:19 and basically saying, well, your choice
00:31:22 will may have actually been driven by other things
00:31:24 that you actually are not choosing.
00:31:27 So that’s why it’s very hard to answer that question.
00:31:30 It’s hard to know what to do with that.
00:31:31 I mean, if the genome has,
00:31:35 if there’s not much freedom, it’s a…
00:31:38 It’s the butterfly effect.
00:31:40 It’s basically that in the short term,
00:31:42 you can predict something extremely well
00:31:45 by knowing the current state of the system.
00:31:48 But a few steps down, it’s very hard to predict
00:31:50 based on the current knowledge.
00:31:52 Is that because the system is truly free?
00:31:55 When I look at weather patterns,
00:31:56 I can predict the next 10 days.
00:31:57 Is it because the weather has a lot of freedom
00:32:00 and after 10 days it chooses to do something else?
00:32:03 Or is it because in fact the system is fully deterministic
00:32:07 and there’s just a slightly different magnetic field
00:32:10 of the earth, slightly more energy arriving from the sun,
00:32:12 a slightly different spin of the gravitational pull
00:32:15 of Jupiter that is now causing all kinds of tides
00:32:18 and slight deviation of the moon, et cetera.
00:32:20 Maybe all of that can be fully modeled.
00:32:22 Maybe the fact that China is emitting
00:32:25 a little more carbon today is actually gonna affect
00:32:28 the weather in Egypt in three weeks.
00:32:31 And all of that could be fully modeled.
00:32:33 In the same way, if you take a complete view
00:32:36 of a human being now, I model everything about you.
00:32:42 The question is, can I predict your next step?
00:32:44 Probably, but at how far?
00:32:47 And if it’s a little further, is that because of stochasticity
00:32:51 and sort of chaos properties of unpredictability
00:32:54 of beyond a certain level?
00:32:56 Or was that actually true free will?
00:32:58 Yeah, so the number of variables might be so,
00:33:01 you might need to build an entire universe to be able to model.
00:33:05 To simulate a human, and then maybe that human
00:33:07 will be fully simulatable.
00:33:09 But maybe aspects of free will will exist.
00:33:12 And where’s that free will coming from?
00:33:13 It’s still coming from the same neurons
00:33:14 or maybe from a spirit inhabiting these neurons.
00:33:17 But again, it’s very difficult empirically
00:33:19 to sort of evaluate where does free will begin
00:33:22 and sort of chemical reactions and electric signals.
00:33:26 So on that topic, let me ask the most absurd question
00:33:31 that most MIT faculty rolled their eyes on.
00:33:33 But what do you think about the simulation hypothesis
00:33:38 and the idea that we live in a simulation?
00:33:40 I think it’s complete BS.
00:33:41 Okay.
00:33:44 There’s no empirical evidence.
00:33:45 No, it’s not. Absolutely not.
00:33:47 Not in terms of empirical evidence or not,
00:33:49 but in terms of a thought experiment,
00:33:52 does it help you think about the universe?
00:33:54 I mean, so if you look at the genome,
00:33:57 it’s encoding a lot of the information
00:33:59 that is required to create some of the beautiful
00:34:01 human complexity that we see around us.
00:34:04 It’s an interesting thought experiment.
00:34:05 How much parameters do we need to have
00:34:11 in order to model this full human experience?
00:34:15 Like if we were to build a video game,
00:34:17 how hard it would be to build a video game
00:34:19 that’s like convincing enough and fun enough
00:34:22 and it has consistent laws of physics, all that stuff.
00:34:28 It’s not interesting to use a thought experiment.
00:34:31 I mean, it’s cute, but it’s Occam’s razor.
00:34:35 I mean, what’s more realistic,
00:34:36 the fact that you’re actually a machine
00:34:38 or that you’re a person?
00:34:39 What’s the fact that all of my experiences exist
00:34:43 inside the chemical molecules that I have
00:34:45 or that somebody is actually simulating all that?
00:34:49 Well, you did refer to humans
00:34:50 as a digital computer earlier.
00:34:52 Of course, of course.
00:34:53 But that does not.
00:34:54 It’s a kind of a machine, right?
00:34:55 I know, I know.
00:34:56 But I think the probability of all that is nil
00:35:01 and let the machines wake me up
00:35:03 and just terminate me now if it’s not.
00:35:07 I challenge you machines.
00:35:08 They’re gonna wait a little bit
00:35:10 to see what you’re gonna do next.
00:35:12 It’s fun.
00:35:13 It’s fun to watch, especially the clever humans.
00:35:17 What’s the difference to you
00:35:18 between the way a computer stores information
00:35:21 and the human genome stores information?
00:35:24 So you also have roots and your work.
00:35:27 Would you say when you introduce yourself at a bar.
00:35:31 It depends who I’m talking to.
00:35:34 Would you say it’s computational biology?
00:35:36 Do you reveal your expertise in computers?
00:35:43 It depends who I’m talking to, truly.
00:35:45 I mean, basically, if I meet someone who’s in computers,
00:35:47 I’ll say, oh, I’m a professor in computer science.
00:35:51 If I meet someone who’s in engineering,
00:35:52 I say computer science and electrical engineering.
00:35:54 If I meet someone in biology,
00:35:55 I’ll say, hey, I work in genomics.
00:35:57 If I meet someone in medicine,
00:35:58 I’m like, hey, I work on genetics.
00:36:00 So you’re a fun person to meet at a bar.
00:36:02 I got you, but so.
00:36:03 No, no, but what I’m trying to say is that I don’t,
00:36:07 I mean, there’s no single attribute
00:36:09 that I will define myself as.
00:36:11 There’s a few things I know.
00:36:12 There’s a few things I study.
00:36:13 There’s a few things I have degrees on
00:36:15 and there’s a few things that I grant degrees in.
00:36:17 And I publish papers across the whole gamut,
00:36:22 the whole spectrum of computation to biology, et cetera.
00:36:26 I mean, the complete answer is that I use computer science
00:36:31 to understand biology.
00:36:34 So I develop methods in AI and machine learning,
00:36:39 statistics and algorithms, et cetera.
00:36:41 But the ultimate goal of my career
00:36:44 is to really understand biology.
00:36:45 If these things don’t advance our understanding
00:36:47 of biology, I’m not as fascinated by them.
00:36:51 Although there are some beautiful computational problems
00:36:54 by themselves, I’ve sort of made it my mission
00:36:57 to apply the power of computer science
00:37:01 to truly understand the human genome, health, disease,
00:37:07 and the whole gamut of how our brain works,
00:37:10 how our body works and all of that,
00:37:11 which is so fascinating.
00:37:13 And so the dream, there’s not an equivalent
00:37:16 sort of complimentary dream of understanding
00:37:20 human biology in order to create an artificial life
00:37:23 or an artificial brain or artificial intelligence
00:37:26 that supersedes the intelligence
00:37:27 and the capabilities of us humans.
00:37:30 It’s an interesting question.
00:37:31 It’s a fascinating question.
00:37:33 So understanding the human brain is undoubtedly coupled
00:37:39 to how do we make better AI?
00:37:42 Because so much of AI has in fact been inspired
00:37:46 by the brain.
00:37:47 It may have taken 50 years
00:37:49 since the early days of neural networks
00:37:51 till we have all of these amazing progress
00:37:55 that we’ve seen with deep belief networks
00:38:00 and all of these advances in Go, in Chess,
00:38:06 in image synthesis, in deep fakes, in you name it.
00:38:10 But the underlying architecture is very much inspired
00:38:17 by the human brain,
00:38:18 which actually posits a very, very interesting question.
00:38:22 Why are neural networks performing so well?
00:38:27 And they perform amazingly well.
00:38:28 Is it because they can simulate any possible function?
00:38:32 And the answer is no, no.
00:38:34 They simulate a very small number of functions.
00:38:37 Is it because they can simulate every function
00:38:39 in the universe?
00:38:40 And that’s where it gets interesting.
00:38:41 The answer is actually, yeah, a little closer to that.
00:38:44 And here’s where it gets really fun.
00:38:47 If you look at human brain and human cognition,
00:38:51 it didn’t evolve in a vacuum.
00:38:53 It evolved in a world with physical constraints,
00:38:58 like the world that inhabits us.
00:39:00 It is the world that we inhabit.
00:39:03 And if you look at our senses, what do they perceive?
00:39:08 They perceive different parts of the electromagnetic spectrum.
00:39:13 The hearing is just different movements in air,
00:39:17 the touch, et cetera.
00:39:18 I mean, all of these things,
00:39:20 we’ve built intuitions for the physical world
00:39:22 that we inhabit.
00:39:23 And our brains and the brains of all animals evolved
00:39:27 for that world.
00:39:29 And the AI systems that we have built
00:39:32 happen to work well with images
00:39:34 of the type that we encounter
00:39:36 in the physical world that we inhabit.
00:39:38 Whereas if you just take noise and you add random signal
00:39:42 that doesn’t match anything in our world,
00:39:44 neural networks will not do as well.
00:39:46 And that actually basically has this whole loop around this,
00:39:52 which is this was designed by studying our own brain,
00:39:57 which was evolved for our own world.
00:39:59 And they happen to do well in our own world.
00:40:01 And they happen to make the same types of mistakes
00:40:04 that humans make many times.
00:40:07 And of course you can engineer images
00:40:08 by adding just the right amount of sort of pixel deviations
00:40:12 to make a zebra look like a bamboo and stuff like that,
00:40:15 or like a table.
00:40:18 But ultimately the undoctored images at least
00:40:23 are very often mistaken, I don’t know,
00:40:25 between muffins and dogs, for example,
00:40:28 in the same way that humans make those mistakes.
00:40:31 So there’s no doubt in my view
00:40:35 that the more we understand about the tricks
00:40:38 that our human brain has evolved
00:40:40 to understand the physical world around us,
00:40:42 the more we will be able to bring
00:40:44 new computational primitives in our AI systems
00:40:48 to again better understand not just the world around us,
00:40:52 but maybe even the world inside us,
00:40:54 and maybe even the computational problems that arise
00:40:57 from new types of data that we haven’t been exposed to,
00:41:00 but are yet inhabiting the same universe that we live in
00:41:03 with a very tiny little subset of functions
00:41:06 from all possible mathematical functions.
00:41:08 Yeah, and that small subset of functions,
00:41:10 all that matters to us humans really, that’s what makes.
00:41:12 It’s all that has mattered so far.
00:41:14 And even within our scientific realm,
00:41:17 it’s all that seems to continue to matter.
00:41:19 But I mean, I always like to think about our senses
00:41:24 and how much of the physical world around us we perceive.
00:41:29 And if you look at the LIGO experiment
00:41:35 over the last year and a half has been all over the news.
00:41:38 What did LIGO do?
00:41:39 It created a new sense for human beings,
00:41:42 a sense that has never been sensed
00:41:45 in the history of our planet.
00:41:48 Gravitational waves have been traversing the earth
00:41:53 since its creation a few billion years ago.
00:41:55 Life has evolved senses to sense things
00:41:59 that were never before sensed.
00:42:02 Light was not perceived by early life.
00:42:05 No one cared.
00:42:07 And eventually photoreceptors evolved
00:42:11 and the ability to sense colors
00:42:14 by sort of catching different parts
00:42:16 of that electromagnetic spectrum.
00:42:19 And hearing evolved and touch evolved, et cetera.
00:42:23 But no organism evolved a way to sense neutrinos
00:42:27 floating through earth or gravitational waves
00:42:29 flowing through earth, et cetera.
00:42:31 And I find it so beautiful in the history
00:42:33 of not just humanity, but life on the planet
00:42:37 that we are now able to capture additional signals
00:42:40 from the physical world than we ever knew before.
00:42:43 And axions, for example, have been all over the news
00:42:46 in the last few weeks.
00:42:47 And the concept that we can capture and perceive
00:42:53 more of that physical world is as exciting
00:42:57 as the fact that we were blind to it
00:43:01 is traumatizing before.
00:43:04 Because that also tells us, you know, we’re in 2020.
00:43:09 Picture yourself in 3020 or in 20, you know.
00:43:12 What new senses might we discover?
00:43:15 Is it, you know, could it be that we’re missing
00:43:19 nine tenths of physics?
00:43:21 That like, there’s a lot of physics out there
00:43:23 that we’re just blind to, completely oblivious to it.
00:43:27 And yet they’re permeating us all the time.
00:43:29 Yeah, so it might be right in front of us.
00:43:31 So when you’re thinking about premonitions,
00:43:35 yeah, a lot of that is ascertainment bias.
00:43:37 Like, yeah, you know, every now and then you’re like,
00:43:39 oh, I remember my friend.
00:43:41 And then my friend doesn’t appear
00:43:42 and I’ll forget that I remembered my friend.
00:43:44 But every now and then my friend will actually appear.
00:43:45 I’m like, oh my God, I thought about you a minute ago.
00:43:48 You just called me, that’s amazing.
00:43:50 So, you know, some of that is this,
00:43:51 but some of that might be that there are,
00:43:55 within our brain, sensors for waves
00:43:59 that we emit that we’re not even aware of.
00:44:03 And this whole concept of when I hug my children,
00:44:07 there’s such an emotional transfer there
00:44:10 that we don’t comprehend.
00:44:12 I mean, sure, yeah, of course we’re all like hard wire
00:44:15 for all kinds of touchy feely things
00:44:16 between parents and kids, it’s beautiful,
00:44:18 between partners, it’s beautiful, et cetera.
00:44:20 But then there are intangible aspects
00:44:24 of human communication
00:44:27 that I don’t think it’s unfathomable
00:44:30 that our brain has actually evolved waves and sensors
00:44:32 for it that we just don’t capture.
00:44:33 We don’t understand the function
00:44:35 of the vast majority of our neurons.
00:44:37 And maybe our brain is already sensing it,
00:44:40 but even worse, maybe our brain is not sensing it at all.
00:44:43 And we’re oblivious to this until we build a machine
00:44:46 that suddenly is able to sort of capture
00:44:48 so much more of what’s happening in the natural world.
00:44:50 So what you’re saying is physics
00:44:52 is going to discover a sensor for love.
00:44:57 And maybe dogs are off scale for that.
00:45:01 And we’ve been oblivious to it the whole time
00:45:04 because we didn’t have the right sensor.
00:45:05 And now you’re gonna have a little wrist that says,
00:45:07 oh my God, I feel all this love in the house.
00:45:09 I sense a disturbance in the forest.
00:45:11 It’s all around us.
00:45:13 And dogs and cats will have zero.
00:45:15 None. None.
00:45:17 It’s just.
00:45:17 Oh, no signal.
00:45:20 But let’s take a step back to our unfortunate place.
00:45:24 To one of the 400 topics that we had actually planned for.
00:45:29 But to our sad time in 2020
00:45:31 when we only have just a few sensors
00:45:33 and very primitive early computers.
00:45:37 So you have a foot in computer science
00:45:41 and a foot in biology.
00:45:43 In your sense, how do computers represent information
00:45:48 differently than like the genome or biological systems?
00:45:52 So first of all, let me correct
00:45:55 that no, we’re in an amazing time in 2020.
00:46:00 Computer science is totally awesome.
00:46:02 And physics is totally awesome.
00:46:03 And we have understood so much of the natural world
00:46:06 than ever before.
00:46:08 So I am extremely grateful and feeling extremely lucky
00:46:13 to be living in the time that we are.
00:46:16 Cause you know, first of all,
00:46:17 who knows when the asteroid will hit.
00:46:20 And second, you know, of all times in humanity,
00:46:26 this is probably the best time to be a human being.
00:46:29 And this might actually be the best place
00:46:31 to be a human being.
00:46:31 So anyway, you know, for anyone who loves science,
00:46:34 this is it.
00:46:35 This is awesome.
00:46:36 This is a great time.
00:46:36 At the same time, just a quick comment.
00:46:39 All I meant is that if we look several hundred years
00:46:43 from now and we end up somehow not destroying ourselves,
00:46:48 people will probably look back at this time
00:46:50 in computer science and at your work of Manos at MIT.
00:46:55 As infantile.
00:46:56 As infantile and silly and how ignorant it all was.
00:46:59 I like to joke very often with my students
00:47:02 that, you know, we’ve written so many papers.
00:47:04 We’ve published so much.
00:47:05 We’ve been citing so much.
00:47:06 And every single time I tell my students, you know,
00:47:08 the best is ahead of us.
00:47:09 What we’re working on now
00:47:11 is the most exciting thing I’ve ever worked on.
00:47:13 So in a way, I do have this sense of, yeah,
00:47:16 even the papers I wrote 10 years ago,
00:47:18 they were awesome at the time,
00:47:20 but I’m so much more excited about where we’re heading now.
00:47:22 And I don’t mean to minimize any of the stuff
00:47:24 we’ve done in the past,
00:47:25 but you know, there’s just this sense of excitement
00:47:29 about what you’re working on now
00:47:30 that as soon as a paper is submitted,
00:47:33 it’s like, ugh, it’s old.
00:47:35 You know, I can’t talk about that anymore.
00:47:37 I’m not gonna talk about it.
00:47:37 At the same time, you’re not,
00:47:38 you probably are not going to be able to predict
00:47:41 what are the most impactful papers and ideas
00:47:45 when people look back 200 years from now at your work,
00:47:47 what would be the most exciting papers.
00:47:50 And it may very well be not the thing that you expected.
00:47:54 Or the things you got awards for or, you know.
00:47:58 This might be true in some fields.
00:47:59 I don’t know.
00:48:00 I feel slightly differently about it in our field.
00:48:02 I feel that I kind of know what are the important ones.
00:48:05 And there’s a very big difference
00:48:07 between what the press picks up on
00:48:09 and what’s actually fundamentally important for the field.
00:48:11 And I think for the fundamentally important ones,
00:48:13 we kind of have a pretty good idea what they are.
00:48:15 And it’s hard to sometimes get the press excited
00:48:18 about the fundamental advances,
00:48:20 but you know, we take what we get
00:48:23 and celebrate what we get.
00:48:24 And sometimes, you know, one of our papers,
00:48:27 which was in a minor journal,
00:48:28 made the front page of Reddit
00:48:30 and suddenly had like hundreds of thousands of views.
00:48:33 Even though it was in a minor journal
00:48:34 because, you know, somebody pitched it the right way
00:48:37 that it suddenly caught everybody’s attention.
00:48:39 Whereas other papers that are sort of truly fundamental,
00:48:42 you know, we have a hard time
00:48:43 getting the editors even excited about them
00:48:46 when so many hundreds of people
00:48:47 are already using the results and building upon them.
00:48:50 So I do appreciate that there’s a discrepancy
00:48:54 between the perception and the perceived success
00:48:57 and the awards that you get for various papers.
00:48:59 But I think that fundamentally, I know that, you know,
00:49:02 some paper, I’m so, so when you write.
00:49:04 So is there a paper that you’re most proud of?
00:49:06 See, now you just, you trapped yourself.
00:49:09 No, no, no, no, I mean.
00:49:10 Is there a line of work that you have a sense
00:49:14 is really powerful that you’ve done to date?
00:49:17 You’ve done so much work in so many directions,
00:49:20 which is interesting.
00:49:21 Is there something where you think is quite special?
00:49:25 I mean, it’s like asking me to say
00:49:28 which of my three children I love best.
00:49:30 I mean.
00:49:34 Exactly.
00:49:34 So, I mean, and it’s such a gimme question
00:49:38 that is so, so difficult not to brag
00:49:42 about the awesome work that my team
00:49:44 and my students have done.
00:49:47 And I’ll just mention a few off the top of my head.
00:49:49 I mean, basically there’s a few landmark papers
00:49:53 that I think have shaped my scientific path.
00:49:56 And, you know, I like to somehow describe it
00:50:00 as a linear continuation of one thing led to another
00:50:03 and led to another led to another.
00:50:05 And, you know, it kind of all started with,
00:50:11 skip, skip, skip, skip, skip.
00:50:12 Let me try to start somewhere in the middle.
00:50:14 So my first PhD paper was the first comparative analysis
00:50:20 of multiple species.
00:50:21 So multiple complete genomes.
00:50:23 So for the first time we basically developed the concept
00:50:27 of genome wide evolutionary signatures.
00:50:29 The fact that you could look across the entire genome
00:50:32 and understand how things evolve.
00:50:35 And from these signatures of evolution
00:50:38 you could go back and study any one region
00:50:41 and say, that’s a protein coding gene.
00:50:44 That’s an RNA gene.
00:50:45 That’s a regulatory motif.
00:50:47 That’s a, you know, binding site and so on and so forth.
00:50:50 So.
00:50:50 I’m sorry, so comparing different.
00:50:52 Different species.
00:50:53 Species of the same.
00:50:55 So take human, mouse, rat and dog.
00:50:57 Yeah.
00:50:58 You know, they’re all animals, they’re all mammals.
00:50:59 They’re all performing similar functions with their heart,
00:51:02 with their brain, with their lungs, et cetera, et cetera.
00:51:05 So there’s many functional elements
00:51:08 that make us uniquely mammalian.
00:51:10 And those mammalian elements are actually conserved.
00:51:14 99% of our genome does not code for protein.
00:51:18 1% codes for protein.
00:51:20 The other 99%, we frankly didn’t know what it does
00:51:25 until we started doing this comparative genomic studies.
00:51:28 So basically these series of papers in my career
00:51:32 have basically first developed that concept
00:51:34 of evolutionary signatures and then apply them to yeast,
00:51:37 apply them to flies, apply them to four mammals,
00:51:40 apply them to 17 fungi,
00:51:41 apply them to 12 Drosophila species,
00:51:43 apply them to then 29 mammals and now 200 mammals.
00:51:46 So sorry, so can we.
00:51:48 So the evolutionary signatures seems like
00:51:51 it’s such a fascinating idea.
00:51:53 And we’re probably gonna linger on your early PhD work
00:51:57 for two hours.
00:51:58 But what is, how can you reveal something interesting
00:52:04 about the genome by looking at the multiple,
00:52:08 multiple species and looking at the evolutionary signatures?
00:52:11 Yeah, so you basically align
00:52:16 the matching regions.
00:52:20 So everything evolved from a common ancestor way, way back.
00:52:23 And mammals evolved from a common ancestor
00:52:26 about 60 million years back.
00:52:27 So after the meteor that killed off the dinosaurs landed
00:52:35 near Machu Picchu, we know the crater.
00:52:38 It didn’t allegedly land.
00:52:41 That was the aliens, okay.
00:52:42 No, just slightly north of Machu Picchu
00:52:44 in the Gulf of Mexico, there’s a giant hole
00:52:47 that that meteor impact.
00:52:49 Sorry, is that definitive to people?
00:52:51 Have people conclusively figured out
00:52:56 what killed the dinosaurs?
00:52:58 I think so.
00:52:59 So it was a meteor?
00:53:00 Well, volcanic activity, all kinds of other stuff
00:53:04 is coinciding, but the meteor is pretty unique
00:53:09 and we now have. That’s also terrifying.
00:53:11 I wouldn’t, we still have a lot of 2020 left,
00:53:14 so if anything.
00:53:15 No, no, but think about it this way.
00:53:17 So the dinosaurs ruled the earth for 175 million years.
00:53:24 We humans have been around for what?
00:53:28 Less than 1 million years.
00:53:29 If you’re super generous about what you call humans
00:53:32 and you include chimps basically.
00:53:35 So we are just getting warmed up
00:53:38 and we’ve ruled the planet much more ruthlessly
00:53:42 than Tyrannosaurus Rex.
00:53:46 T Rex had much less of an environmental impact
00:53:48 than we did.
00:53:49 And if you give us another 174 million years,
00:53:54 humans will look very different if we make it that far.
00:53:58 So I think dinosaurs basically are much more
00:54:02 of life history on earth than we are in all respects.
00:54:06 But look at the bright side, when they were killed off,
00:54:08 another life form emerged, mammals.
00:54:10 And that’s that whole evolutionary branching
00:54:14 that’s happened.
00:54:15 So you kind of have,
00:54:17 when you have these evolutionary signatures,
00:54:19 is there basically a map of how the genome changed?
00:54:22 Yeah, exactly, exactly.
00:54:23 So now you can go back to this early mammal
00:54:26 that was hiding in caves and you can basically ask
00:54:29 what happened after the dinosaurs were wiped out.
00:54:31 A ton of evolutionary niches opened up
00:54:34 and the mammals started populating all of these niches.
00:54:37 And in that diversification,
00:54:40 there was room for expansion of new types of functions.
00:54:44 So some of them populated the air with bats flying,
00:54:50 a new evolution of flight.
00:54:53 Some populated the oceans with dolphins and whales
00:54:57 going off to swim, et cetera.
00:54:58 But we all are fundamentally mammals.
00:55:01 So you can take the genomes of all these species
00:55:04 and align them on top of each other
00:55:06 and basically create nucleotide resolution correspondences.
00:55:11 What my PhD work showed is that when you do that,
00:55:14 when you line up species on top of each other,
00:55:17 you can see that within protein coding genes,
00:55:19 there’s a particular pattern of evolution
00:55:21 that is dictated by the level at which
00:55:25 evolutionary selection acts.
00:55:27 If I’m coding for a protein and I change
00:55:30 the third codon position of a triplet
00:55:34 that codes for that amino acid,
00:55:36 the same amino acid will be encoded.
00:55:38 So that basically means that any kind of mutation
00:55:42 that preserves that translation that is invariant
00:55:46 to that ultimate functional assessment
00:55:49 that evolution will give is tolerated.
00:55:52 So for any function that you’re trying to achieve,
00:55:55 there’s a set of sequences that encode it.
00:55:57 You can now look at the mapping,
00:56:00 the graph isomorphism, if you wish,
00:56:04 between all of the possible DNA encodings
00:56:07 of a particular function and that function.
00:56:09 And instead of having just that exact sequence
00:56:12 at the protein level, you can think of the set
00:56:15 of protein sequences that all fulfill the same function.
00:56:18 What’s evolution doing?
00:56:19 Evolution has two components.
00:56:20 One component is random, blind, and stupid mutation.
00:56:25 The other component is super smart, ruthless selection.
00:56:32 That’s my mom calling from Greece.
00:56:35 Yes, I might be a fully grown man, but I am a Greek.
00:56:40 Did you just cancel the call?
00:56:41 Wow, you’re in trouble.
00:56:42 I know, I’m in trouble.
00:56:43 No, she’s gonna be calling the cops.
00:56:44 Honey, are you okay?
00:56:45 I’m gonna edit this clip out and send it to her.
00:56:47 Sure.
00:56:51 So there’s a lot of encoding
00:56:53 for the same kind of function.
00:56:54 Yeah, so you now have this mapping
00:56:56 between all of the set of functions
00:56:58 that could all encode the same,
00:57:00 all of the set of sequences
00:57:02 that can all encode the same function.
00:57:04 What evolutionary signatures does
00:57:06 is that it basically looks at the shape
00:57:08 of that distribution of sequences
00:57:11 that all encode the same thing.
00:57:13 And based on that shape, you can basically say,
00:57:15 ooh, proteins have a very different shape
00:57:17 than RNA structures, than regulatory motifs, et cetera.
00:57:21 So just by scanning a sequence, ignoring the sequence
00:57:24 and just looking at the patterns of change,
00:57:26 I’m like, wow, this thing is evolving like a protein
00:57:29 and that thing is evolving like a motif
00:57:31 and that thing is evolving.
00:57:33 So that’s exactly what we just did for COVID.
00:57:35 So our paper that we posted in bioRxiv about coronavirus
00:57:39 basically took this concept of evolutionary signatures
00:57:42 and applied it on the SARS CoV2 genome
00:57:45 that is responsible for the COVID 19 pandemic.
00:57:48 And comparing it to?
00:57:50 To 44 serbicovirus species.
00:57:52 So this is the beta.
00:57:53 What word did you just use, serbicovirus?
00:57:56 Serbicovirus, so SARS related beta coronavirus.
00:58:00 It’s a portmanteau of a bunch.
00:58:01 So that whole family of viruses.
00:58:03 Yeah, so.
00:58:03 How big is that family by the way?
00:58:05 We have 44 species that, or I mean.
00:58:07 There’s 44 species in the family?
00:58:09 Yeah. Virus is a clever bunch.
00:58:11 No, no, but there’s just 44.
00:58:12 And again, we don’t call them species in viruses.
00:58:15 We call them strains.
00:58:16 But anyway, there’s 44 strains.
00:58:18 And that’s a tiny little subset of maybe another 50 strains
00:58:22 that are just far too distantly related.
00:58:24 Most of those only infect bats as the host
00:58:29 and a subset of only four or five have ever infected humans.
00:58:34 And we basically took all of those
00:58:35 and we aligned them in the same exact way
00:58:37 that we’ve aligned mammals.
00:58:39 And then we looked at what proteins are,
00:58:42 which of the currently hypothesized genes
00:58:44 for the coronavirus genome
00:58:47 are in fact evolving like proteins and which ones are not.
00:58:50 And what we found is that ORF10,
00:58:52 the last little open reading frame,
00:58:54 the last little gene in the genome is bogus.
00:58:56 That’s not a protein at all.
00:58:58 What is it?
00:58:59 It’s an RNA structure.
00:59:01 That doesn’t have a.
00:59:03 It doesn’t get translated into amino acids.
00:59:05 And that, so it’s important to narrow down
00:59:08 to basically discover what’s useful and what’s not.
00:59:10 Exactly.
00:59:11 Basically, what is even the set of genes?
00:59:13 The other thing that these evolutionary signatures showed
00:59:15 is that within ORF3A lies a tiny little additional gene
00:59:20 encoded within the other gene.
00:59:22 So you can translate a DNA sequence
00:59:24 in three different reading frames.
00:59:26 If you start in the first one, it’s ATG, et cetera.
00:59:30 If you start on the second one, it’s TGC, et cetera.
00:59:32 And there’s a gene within a gene.
00:59:36 So there’s a whole other protein
00:59:37 that we didn’t know about that might be super important.
00:59:41 So we don’t even know the building blocks of SARS COVID 2.
00:59:45 So if we want to understand coronavirus biology
00:59:48 and eventually find it successfully,
00:59:50 we need to even have the set of genes
00:59:51 and these evolutionary signatures
00:59:53 that I developed in my PhD work.
00:59:55 Are you really useful here?
00:59:56 We just recently used.
00:59:57 You know what, let’s run with that tangent
00:59:59 for a little bit, if it’s okay.
01:00:01 Can we talk about the COVID 19 a little bit more?
01:00:08 What’s your sense about the genome, the proteins,
01:00:13 the functions that we understand about COVID 19?
01:00:16 Where do we stand in your sense?
01:00:18 What are the big open problems?
01:00:21 And also, you kind of said it’s important to understand
01:00:25 what are the important proteins
01:00:29 and why is that important?
01:00:34 So what else does the comparison of these species tell us?
01:00:39 What it tells us is how fast are things evolving?
01:00:43 It tells us about at what level is the acceleration
01:00:46 or deceleration pedal set for every one of these proteins.
01:00:50 So the genome has 30 some genes.
01:00:54 Some genes evolve super, super fast.
01:00:56 Others evolve super, super slow.
01:00:59 If you look at the polymerase gene
01:01:00 that basically replicates the genome,
01:01:01 that’s a super slow evolving one.
01:01:04 If you look at the nucleocapsid protein,
01:01:06 that’s also super slow evolving.
01:01:09 If you look at the spike one protein,
01:01:11 this is the part of the spike protein
01:01:13 that actually touches the ACE2 receptor
01:01:15 and then enables the virus to attach to your cells.
01:01:21 That’s the thing that gives it that visual…
01:01:23 Yeah, the corona look basically.
01:01:24 The corona look, yeah.
01:01:26 So basically the spike protein sticks out of the virus
01:01:28 and there’s a first part of the protein S1
01:01:31 which basically attaches to the ACE2 receptor.
01:01:34 And then S2 is the latch that sort of pushes and channels
01:01:39 the fusion of the membranes
01:01:41 and then the incorporation of the viral RNA inside our cells
01:01:47 which then gets translated into all of these 30 proteins.
01:01:50 So the S1 protein is evolving ridiculously fast.
01:01:55 So if you look at the stop versus gas pedal,
01:01:59 the gas pedal is all the way down.
01:02:02 ORF8 is also evolving super fast
01:02:05 and ORF6 is evolving super fast.
01:02:06 We have no idea what they do.
01:02:08 We have some idea but nowhere near what S1 is.
01:02:11 So what the…
01:02:12 Isn’t that terrifying that S1 is evolving?
01:02:14 That means that’s a really useful function
01:02:16 and if it’s evolving fast,
01:02:18 doesn’t that mean new strains could be created
01:02:20 or it does something?
01:02:21 That means that it’s searching for how to match,
01:02:24 how to best match the host.
01:02:26 So basically anything in general in evolution,
01:02:29 if you look at genomes,
01:02:30 anything that’s contacting the environment
01:02:32 is evolving much faster than anything that’s internal.
01:02:34 And the reason is that the environment changes.
01:02:37 So if you look at the evolution of the cervical viruses,
01:02:42 the S1 protein has evolved very rapidly
01:02:44 because it’s attaching to different hosts each time.
01:02:47 We think of them as bats,
01:02:48 but there’s thousands of species of bats
01:02:50 and to go from one species of bat to another species of bat,
01:02:52 you have to adjust S1 to the new ACE2 receptor
01:02:55 that you’re gonna be facing in that new species.
01:02:58 Sorry, quick tangent.
01:02:59 Is it fascinating to you that viruses are doing this?
01:03:03 I mean, it feels like they’re this intelligent organism.
01:03:06 I mean, does it give you pause how incredible it is
01:03:12 that the evolutionary dynamics that you’re describing
01:03:16 is actually happening and they’re freaking out,
01:03:19 figuring out how to jump from bats to humans
01:03:22 all in this distributed fashion?
01:03:24 And then most of us don’t even say
01:03:25 they’re alive or intelligent or whatever.
01:03:27 So intelligence is in the eye of the beholder.
01:03:31 Stupid is as stupid does, as Forrest Gump would say,
01:03:34 and intelligent is as intelligent does.
01:03:36 So basically if the virus is finding solutions
01:03:39 that we think of as intelligent,
01:03:40 yeah, it’s probably intelligent,
01:03:42 but that’s again in the eye of the beholder.
01:03:43 Do you think viruses are intelligent?
01:03:45 Oh, of course not.
01:03:47 Really?
01:03:48 No.
01:03:49 It’s so incredible.
01:03:50 So remember when I was talking about the two components
01:03:52 of evolution, one is the stupid mutation,
01:03:55 which is completely blind,
01:03:57 and the other one is the super smart selection,
01:04:00 which is ruthless.
01:04:01 So it’s not viruses who are smart.
01:04:04 It’s this component of evolution that’s smart.
01:04:06 So it’s evolution that sort of appears smart.
01:04:10 And how is that happening?
01:04:12 By huge parallel search across thousands of parallel
01:04:17 of parallel infections throughout the world right now.
01:04:21 Yes, but so to push back on that,
01:04:23 so yes, so then the intelligence is in the mechanism,
01:04:27 but then by that argument,
01:04:31 viruses would be more intelligent
01:04:32 because there’s just more of them.
01:04:34 So the search, they’re basically the brute force search
01:04:38 that’s happening with viruses
01:04:40 because there’s so many more of them than humans,
01:04:43 then they’re taken as a whole are more intelligent.
01:04:47 I mean, so you don’t think it’s possible that,
01:04:51 I mean, who runs, would we even be here if viruses weren’t,
01:04:55 I mean, who runs this thing?
01:04:58 So humans or viruses?
01:04:59 So let me answer, yeah, let me answer your question.
01:05:03 So we would not be here if it wasn’t for viruses.
01:05:10 And part of the reason is that
01:05:11 if you look at mammalian evolution early on
01:05:14 in this mammalian radiation
01:05:16 that basically happened after the death of the dinosaurs
01:05:18 is that some of the viruses that we had in our genome
01:05:22 spread throughout our genome
01:05:24 and created binding sites
01:05:27 for new classes of regulatory proteins.
01:05:30 And these binding sites that landed all over our genome
01:05:33 are now control elements that basically control our genes
01:05:36 and sort of help the complexity of the circuitry
01:05:40 of mammalian genomes.
01:05:42 So, you know, everything’s coevolution.
01:05:45 That’s fascinating, we’re working together.
01:05:47 And yet you say they’re dumb.
01:05:48 We’ve coopted them.
01:05:49 No, I never said they’re dumb.
01:05:51 They just don’t care.
01:05:53 They don’t care.
01:05:54 Another thing, oh, is the virus trying to kill us?
01:05:56 No, it’s not.
01:05:58 The virus is not trying to kill you.
01:05:59 It’s actually actively trying to not kill you.
01:06:02 So when you get infected, if you die,
01:06:05 bomber, I killed him,
01:06:07 is what the reaction of the virus will be.
01:06:09 Why? Because that virus won’t spread.
01:06:12 Many people have a misconception of,
01:06:13 oh, viruses are smart or oh, viruses are mean.
01:06:16 They don’t care.
01:06:18 It’s like, you have to clean yourself
01:06:20 of any kind of anthropomorphism out there.
01:06:23 I don’t know.
01:06:24 Oh, yes.
01:06:24 So there’s a sense when taken as a whole that there’s…
01:06:31 It’s in the eye of the beholder.
01:06:32 Stupid is as stupid does.
01:06:34 Intelligent is as intelligent does.
01:06:35 So if you want to call them intelligent, that’s fine.
01:06:38 Because the end result is that
01:06:40 they’re finding amazing solutions.
01:06:42 I mean, I am in awe.
01:06:44 They’re so dumb about it.
01:06:45 They’re just doing dumb.
01:06:46 They don’t care.
01:06:47 They’re not dumb and they’re just don’t care.
01:06:48 They don’t care.
01:06:50 The care word is really interesting.
01:06:51 I mean, there could be an argument that they’re conscious.
01:06:54 They’re just dividing.
01:06:55 They’re not.
01:06:56 They’re just dividing.
01:06:57 They’re just a little entity
01:07:00 which happens to be dividing and spreading.
01:07:02 It just doesn’t want to kill us.
01:07:04 In fact, it prefers not to kill us.
01:07:06 It just wants to spread.
01:07:07 And when I say wants, again, I’m anthropomorphizing,
01:07:11 but it’s just that if you have two versions of a virus,
01:07:15 one acquires a mutation that spreads more,
01:07:17 that’s going to spread more.
01:07:18 One acquires a mutation that spreads less,
01:07:20 that’s going to be lost.
01:07:21 One acquires a mutation that enters faster,
01:07:24 that’s going to be kept.
01:07:25 One acquires a mutation that kills you right away,
01:07:27 it’s going to be lost.
01:07:28 So over evolutionary time,
01:07:30 the viruses that spread super well
01:07:32 but don’t kill the host
01:07:33 are the ones that are going to survive.
01:07:36 Yeah, but so you brilliantly described
01:07:39 the basic mechanisms of how it all happens,
01:07:41 but when you zoom out and you see the entirety of viruses,
01:07:46 maybe across different strains of viruses,
01:07:49 it seems like a living organism.
01:07:52 I am in awe of biology.
01:07:55 I find biology amazingly beautiful.
01:07:58 I find the design of the current coronavirus,
01:08:01 however lethal it is, amazingly beautiful.
01:08:04 The way that it is encoded,
01:08:06 the way that it tricks your cells
01:08:08 into making 30 proteins from a single RNA.
01:08:12 Human cells don’t do that.
01:08:14 Human cells make one protein from each RNA molecule.
01:08:18 They don’t make two, they make one.
01:08:20 We are hardwired to make only one protein
01:08:22 from every RNA molecule.
01:08:23 And yet this virus goes in,
01:08:25 throws in a single messenger RNA.
01:08:28 Just like any messenger RNA,
01:08:29 we have tens of thousands of messenger RNAs
01:08:32 in our cells in any one time.
01:08:34 In every one of our cells.
01:08:35 It throws in one RNA and that RNA is so,
01:08:40 I’m gonna use your word here, not my word, intelligent.
01:08:44 That it hijacks the entire machinery of your human cell.
01:08:49 It basically has at the beginning,
01:08:52 a giant open reading frame.
01:08:54 That’s a giant protein that gets translated.
01:08:57 Two thirds of that RNA make a single giant protein.
01:09:01 That single protein is basically
01:09:03 what a human cell would make.
01:09:04 It’s like, oh, here’s a start code.
01:09:06 I’m gonna start translating here.
01:09:07 Human cells are kind of dumb.
01:09:08 I’m sorry.
01:09:09 Again, this is not the words I would normally use.
01:09:12 But the human cell basically says,
01:09:13 oh, this is an RNA, must be mine.
01:09:15 Let me translate.
01:09:15 And it starts translating it.
01:09:17 And then you’re in trouble.
01:09:18 Why?
01:09:19 Because that one protein as it’s growing,
01:09:22 gets cleaved into about 20 different peptides.
01:09:26 The first peptide and the second peptide start interacting
01:09:30 and the third one and the fourth one.
01:09:32 And they shut off the ribosome of the whole cell
01:09:37 to not translate human RNAs anymore.
01:09:42 So the virus basically hijacks your cells
01:09:46 and it cuts, it cleaves every one of your human RNAs
01:09:50 to basically say to the ribosome,
01:09:52 don’t translate this one, junk.
01:09:53 Don’t look at this one, junk.
01:09:55 And it only spares its own RNAs
01:09:58 because they have a particular mark that it spares.
01:10:01 Then all of the ribosomes that normally make protein
01:10:04 in your human cells are now only able
01:10:06 to translate viral RNAs.
01:10:09 And then more and more and more and more of them.
01:10:11 That’s the first 20 proteins.
01:10:13 In fact, halfway down about protein 11,
01:10:16 between 11 and 12,
01:10:17 you basically have a translational slippage
01:10:21 where the ribosome skips reading frame.
01:10:23 And it translates from one reading frame
01:10:24 to another reading frame.
01:10:25 That means that about half of them
01:10:27 are gonna be translated from one to 11.
01:10:29 And the other half are gonna be translated
01:10:30 from 12 to 16.
01:10:32 It’s gorgeous.
01:10:34 And then you’re done.
01:10:37 Then that mRNA will never translate the last 10 proteins
01:10:40 but spike is the one right after that one.
01:10:42 So how does spike even get translated?
01:10:45 This positive strand RNA virus has a reverse transcriptase
01:10:50 which is an RNA based reverse transcriptase.
01:10:52 So from the RNA on the positive strand,
01:10:54 it makes an RNA on the negative strand.
01:10:56 And in between every single one of these genes,
01:10:59 these open reading frames,
01:11:01 there’s a little signal AACGCA or something like that,
01:11:05 that basically loops over to the beginning of the RNA.
01:11:09 And basically instead of sort of having
01:11:11 a single full negative strand RNA,
01:11:14 it basically has a partial negative strand RNA
01:11:16 that ends right before the beginning of that gene.
01:11:19 And another one that ends right before
01:11:20 the beginning of that gene.
01:11:21 These negative strand RNAs now make positive strand RNAs
01:11:25 that then look to the human whole cell
01:11:27 just like any other human mRNA.
01:11:29 It’s like, ooh, great, I’m gonna translate that one
01:11:31 because it doesn’t have the cleaving
01:11:32 that the virus has now put on all your human genes.
01:11:36 And then you’ve lost the battle.
01:11:38 That cell is now only making proteins for the virus
01:11:42 that will then create the spike protein,
01:11:45 the envelope protein, the membrane protein,
01:11:47 the nucleocapsid protein that will package up the RNA
01:11:50 and then sort of create new viral envelopes.
01:11:53 And these will then be secreted out of that cell
01:11:57 in new little packages
01:11:59 that will then infect the rest of the cells.
01:12:00 Repeat the whole process again.
01:12:01 It’s beautiful, right?
01:12:03 It’s mind boggling.
01:12:04 It’s hard not to anthropomorphize it.
01:12:05 I know, but it’s so gorgeous.
01:12:08 So there is a beauty to it.
01:12:09 Of course.
01:12:12 Is it terrifying to you?
01:12:13 So this is something that has happened throughout history.
01:12:16 Humans have been nearly wiped out
01:12:19 over and over and over again,
01:12:21 and yet never fully wiped out.
01:12:23 So yeah, I’m not concerned about the human race.
01:12:25 I’m not even concerned about the impact
01:12:29 on sort of our survival as a species.
01:12:33 This is absolutely something,
01:12:35 I mean, human life is so invaluable
01:12:38 and every one of us is so invaluable,
01:12:40 but if you think of it as sort of,
01:12:42 is this the end of our species?
01:12:44 By no means, basically.
01:12:46 So let me explain.
01:12:48 The Black Death killed what, 30% of Europe?
01:12:51 That has left a tremendous imprint,
01:12:55 a huge hole, a horrendous hole
01:12:59 in the genetic makeup of humans.
01:13:03 There’s been series of wiping out of huge fractions
01:13:08 of entire species or just entire species altogether.
01:13:12 And that has a consequence on the human immune repertoire.
01:13:17 If you look at how Europe was shaped
01:13:20 and how Africa was shaped by malaria, for example,
01:13:24 all the individuals that carry a mutation
01:13:26 that protects you from malaria
01:13:29 were able to survive much more.
01:13:31 And if you look at the frequency of sickle cell disease
01:13:33 and the frequency of malaria,
01:13:35 the maps are actually showing the same pattern,
01:13:38 the same imprint on Africa.
01:13:40 And that basically led people to hypothesize
01:13:42 that the reason why sickle cell disease
01:13:43 is so much more frequent is because
01:13:45 sickle cell disease is so much more frequent
01:13:47 in Americans of African descent
01:13:50 is because there was selection in Africa against malaria
01:13:55 leading to sickle cell, because when the cells sickle,
01:13:57 malaria is not able to replicate inside your cells as well.
01:14:01 And therefore you protect against that.
01:14:03 So if you look at human disease,
01:14:05 all of the genetic associations that we do
01:14:07 with human disease,
01:14:09 you basically see the imprint
01:14:13 of these waves of selection killing off
01:14:16 gazillions of humans.
01:14:18 And there’s so many immune processes that are coming up
01:14:23 as associated with so many different diseases.
01:14:25 The reason for that is similar
01:14:27 to what I was describing earlier,
01:14:28 where the outward facing proteins evolve much more rapidly
01:14:33 because the environment is always changing.
01:14:35 But what’s really interesting in the human genome
01:14:37 is that we have coopted many of these immune genes
01:14:40 to carry out nonimmune functions.
01:14:42 For example, in our brain,
01:14:43 we use immune cells to cleave off neuronal connections
01:14:48 that don’t get used.
01:14:50 This whole use it or lose it, we know the mechanism.
01:14:52 It’s microglia that cleave off neuronal synaptic connections
01:14:57 that are just not utilized.
01:14:59 When you utilize them, you mark them in a particular way
01:15:02 that basically when the microglia come,
01:15:04 tell it, don’t kill this one, it’s used now.
01:15:07 And the microglia will go off
01:15:08 and kill the ones you don’t use.
01:15:10 This is an immune function,
01:15:12 which is coopted to do nonimmune things.
01:15:14 If you look at our adipocytes,
01:15:16 M1 versus M2 macrophages inside our fat
01:15:19 will basically determine whether you’re obese or not.
01:15:22 And these are again, immune cells that are resident
01:15:24 and living within these tissues.
01:15:27 So many disease associations.
01:15:30 That’s it, that we coopt these kinds of things
01:15:33 for incredibly complicated functions.
01:15:36 Exactly, evolution works in so many different ways,
01:15:39 which are all beautiful and mysterious.
01:15:41 But not intelligent.
01:15:43 Not intelligent, it’s in the eye of the beholder.
01:15:45 But the point that I’m trying to make is that
01:15:51 if you look at the imprint that COVID will have,
01:15:54 hopefully it will not be big.
01:15:55 Hopefully the US will get attacked together
01:15:57 and stop the virus from spreading further.
01:16:00 But if it doesn’t, it’s having an imprint
01:16:03 on individuals who have particular genetic repertoires.
01:16:07 So if you look at now the genetic associations
01:16:10 of blood type and immune function cells, et cetera,
01:16:13 there’s actually association, genetic variation
01:16:15 that basically says how much more likely am I or you to die
01:16:18 if we contact the virus.
01:16:20 And it’s through these rounds of shaping the human genome
01:16:24 that humans have basically made it so far.
01:16:27 And selection is ruthless and it’s brutal
01:16:32 and it only comes with a lot of killing.
01:16:34 But this is the way that viruses and environments
01:16:38 have shaped the human genome.
01:16:39 Basically, when you go through periods of famine,
01:16:41 you select for particular genes.
01:16:43 And what’s left is not necessarily better,
01:16:46 it’s just whatever survived.
01:16:49 And it might have been the surviving one back then,
01:16:51 not because it was better,
01:16:53 maybe the ones that ran slower survived.
01:16:54 I mean, again, not necessarily better,
01:16:57 but the surviving ones are basically the ones
01:17:00 that then are shaped for any kind
01:17:02 of subsequent evolutionary condition
01:17:05 and environmental condition.
01:17:07 But if you look at, for example, obesity,
01:17:09 obesity was selected for basically the genes
01:17:12 that now predisposes to obesity
01:17:14 were at 2% frequency in Africa.
01:17:16 They rose to 44% frequency in Europe.
01:17:19 Wow, that’s fascinating.
01:17:20 Because you basically went through the ice ages
01:17:22 and there was a scarcity of food.
01:17:24 So there was a selection to being able to store
01:17:27 every single calorie you consume.
01:17:29 Eventually, environment changes.
01:17:31 So the better allele, which was the fat storing allele,
01:17:35 became the worst allele
01:17:36 because it’s the fat storing allele.
01:17:38 It still has the same function.
01:17:40 So if you look at my genome, speaking of mom calling,
01:17:44 mom gave me a bad copy of that gene, this FTO locus.
01:17:48 Basically, makes me.
01:17:49 The one that has to do with.
01:17:50 Obesity.
01:17:51 With obesity.
01:17:52 Yeah, I basically now have a bad copy from mom
01:17:54 that makes me more likely to be obese.
01:17:56 And I also have a bad copy from dad
01:17:59 that makes me more likely to be obese.
01:18:00 So homozygous.
01:18:01 And that’s the allele, it’s still the minor allele,
01:18:05 but it’s at 44% frequency in Southeast Asia,
01:18:09 42% frequency in Europe, even though it started at 2%.
01:18:12 It was an awesome allele to have 100 years ago.
01:18:16 Right now, it’s pretty terrible allele.
01:18:17 So the other concept is that diversity matters.
01:18:21 If we had 100 million nuclear physicists
01:18:25 living the earth right now, we’d be in trouble.
01:18:28 You need diversity, you need artists
01:18:31 and you need musicians and you need mathematicians
01:18:33 and you need politicians, yes, even those.
01:18:37 And you need like.
01:18:37 Well, let’s not get crazy.
01:18:39 But because then if a virus comes along or whatever.
01:18:43 Exactly, exactly.
01:18:44 So, no, there’s two reasons.
01:18:45 Number one, you want diversity in the immune repertoire
01:18:48 and we have built in diversity.
01:18:50 So basically, they are the most diverse.
01:18:53 Basically, if you look at our immune system,
01:18:54 there’s layers and layers of diversity.
01:18:57 Like the way that you create your cells generates diversity
01:19:01 because of the selection for the VDJ recombination
01:19:04 that basically eventually leads
01:19:06 to a huge number of repertoires.
01:19:08 But that’s only one small component of diversity.
01:19:10 The blood type is another one.
01:19:11 The major histocompatibility complex, the HLA alleles
01:19:15 are another source of diversity.
01:19:18 So the immune system of humans is by nature,
01:19:21 incredibly diverse and that basically leads to resilience.
01:19:25 So basically what I’m saying that I don’t worry
01:19:27 for the human species because we are so diverse immunologically,
01:19:32 we are likely to be very resilient
01:19:34 against so many different attacks like this current virus.
01:19:39 So you’re saying natural pandemics may not be something
01:19:42 that you’re really afraid of because of the diversity
01:19:45 in our genetic makeup.
01:19:48 What about engineered pandemics?
01:19:50 Do you have fears of us messing with the makeup of viruses
01:19:55 or well, yeah, let’s say with the makeup of viruses
01:19:58 to create something that we can’t control
01:20:00 and would be much more destructive
01:20:02 than it would come about naturally?
01:20:05 Remember how we were talking about how smart evolution is?
01:20:08 Humans are much dumber.
01:20:09 So.
01:20:10 You mean like human scientists, engineers?
01:20:11 Yeah, humans, humans just like.
01:20:13 Humans overall?
01:20:14 Yeah, humans overall.
01:20:14 Okay.
01:20:15 But I mean, even the sort of synthetic biologists
01:20:19 you know, basically if you were to create,
01:20:25 you know, virus like SARS that will kill a lot of people,
01:20:29 you would probably start with SARS.
01:20:32 So whoever, you know, would like to design such a thing
01:20:37 would basically start with a SARS tree
01:20:39 or at least some relative of SARS.
01:20:42 The source genome for the current virus
01:20:45 was something completely different.
01:20:47 It was something that has never infected anyone
01:20:49 and never infected humans.
01:20:50 No one in their right mind would have started there.
01:20:52 But when you say sources like the nearest.
01:20:55 The nearest relative.
01:20:56 Relative.
01:20:57 Is in a whole other branch.
01:20:58 Interesting.
01:20:59 No species of which has ever infected humans
01:21:00 in that branch.
01:21:02 So, you know, let’s put this to rest.
01:21:05 This was not designed by someone to kill off the human race.
01:21:08 So you don’t believe it was engineered?
01:21:12 The. Or likely.
01:21:13 Yeah, the path to engineering a deadly virus
01:21:16 did not come from this strain that was used.
01:21:21 Moreover, there’s been various claims of,
01:21:26 ha ha, this was mixed and matched in lab
01:21:29 because the S1 protein has three different components,
01:21:32 each of which has a different evolutionary tree.
01:21:34 So, you know, a lot of popular press basically said,
01:21:37 aha, this came from pangolin
01:21:39 and this came from, you know, all kinds of other species.
01:21:42 This is what has been happening
01:21:44 throughout the coronavirus tree.
01:21:46 So basically the S1 protein has been recombining
01:21:49 across species all the time.
01:21:50 Remember when I was talking about the positive strand,
01:21:52 the negative strand, sub genomic RNAs,
01:21:54 these can actually recombine.
01:21:55 And if you have two different viruses
01:21:57 infecting the same cell,
01:21:58 they can actually mix and match
01:21:59 between the positive strand and the negative strand
01:22:01 and basically create a new hybrid virus with recombination
01:22:04 that now has the S1 from one
01:22:06 and the rest of the genome from another.
01:22:08 And this is something that happens a lot in S1,
01:22:10 in Orfet, et cetera.
01:22:12 And that’s something that’s true of the whole tree.
01:22:13 For the whole family of viruses.
01:22:15 So it’s not like someone has been messing with this
01:22:18 for millions of years and, you know, changing.
01:22:20 This happens naturally.
01:22:21 That’s, again, beautiful that that somehow happens,
01:22:24 that they recombine.
01:22:25 So two different strands can infect the body
01:22:27 and then recombine.
01:22:30 So all of this actually magic happens inside hosts.
01:22:35 Like all, like.
01:22:36 Yeah, that’s why classification wise,
01:22:39 virus is not thought to be alive
01:22:40 because it doesn’t self replicate.
01:22:41 It’s not autonomous.
01:22:43 It’s something that enters a living cell
01:22:45 and then co ops it to basically make it its own.
01:22:48 But by itself, people ask me,
01:22:50 how do we kill this bastard?
01:22:51 I’m like, you stop it from replicating.
01:22:54 It’s not like a bacterium that will just live
01:22:57 in a, you know, puddle or something.
01:23:01 It’s a virus.
01:23:02 Viruses don’t live without their host.
01:23:04 And they only live with their host for very little time.
01:23:07 So if you stop it from replicating,
01:23:09 it’ll stop from spreading.
01:23:11 I mean, it’s not like HIV, which can stay dormant
01:23:13 for a long time.
01:23:14 Basically, coronaviruses just don’t do that.
01:23:15 They’re not integrating genomes.
01:23:16 They’re RNA genomes.
01:23:18 So if it’s not expressed, it degrades.
01:23:20 RNA degrades.
01:23:21 It doesn’t just stick around.
01:23:23 Well, let me ask also about the immune system you mentioned.
01:23:27 A lot of people kind of ask, you know,
01:23:31 how can we strengthen the immune system
01:23:34 to respond to this particular virus,
01:23:36 but the viruses in general.
01:23:37 Do you have from a biological perspective,
01:23:40 thoughts on what we can do as humans
01:23:43 to strengthen our immune system?
01:23:43 If you look at the death rates across different countries,
01:23:46 people with less vaccination have been dying more.
01:23:49 If you look at North Italy,
01:23:51 the vaccination rates are abysmal there.
01:23:53 And a lot of people have been dying.
01:23:55 If you look at Greece, very good vaccination rates.
01:23:58 Almost no one has been dying.
01:24:00 So yes, there’s a policy component.
01:24:03 So Italy reacted very slowly.
01:24:05 Greece reacted very fast.
01:24:07 So yeah, many fewer people died in Greece,
01:24:09 but there might actually be a component
01:24:11 of genetic immune repertoire.
01:24:14 Basically, how did people die off, you know,
01:24:16 in the history of the Greek population
01:24:19 versus the Italian population.
01:24:20 Wow. There’s a…
01:24:22 That’s interesting to think about.
01:24:24 And then there’s a component
01:24:25 of what vaccinations did you have as a kid
01:24:28 and what are the off target effects of those vaccinations?
01:24:32 So basically a vaccination can have two components.
01:24:34 One is training your immune system
01:24:37 against that specific insult.
01:24:39 The second one is boosting up your immune system
01:24:42 for all kinds of other things.
01:24:44 If you look at allergies,
01:24:47 Northern Europe, super clean environments,
01:24:50 tons of allergies.
01:24:51 Southern Europe, my kids grew up eating dirt.
01:24:54 No allergies.
01:24:57 So growing up, I never had even heard of what allergies are.
01:25:00 Like, was it really allergies?
01:25:01 And the reason is that I was playing in the garden.
01:25:03 I was putting all kinds of stuff in my mouth from,
01:25:05 you know, all kinds of dirt and stuff,
01:25:07 tons of viruses there, tons of bacteria there.
01:25:09 You know, my immune system was built up.
01:25:11 So the more you protect your immune system from exposure,
01:25:16 the less opportunity it has to learn
01:25:18 about non self repertoire in a way that prepares it
01:25:23 for the next insult.
01:25:24 So that’s the horizontal thing too,
01:25:25 like the, so it’s throughout your lifetime
01:25:28 and the lifetime of the people that, your ancestors,
01:25:33 that kind of thing.
01:25:34 What about the…
01:25:35 So again, it returns against free will.
01:25:37 On the free will side of things,
01:25:39 is there something we could do
01:25:40 to strengthen our immune system in 2020?
01:25:44 Is there like, you know, exercise, diet,
01:25:49 all that kind of stuff?
01:25:50 So it’s kind of funny.
01:25:52 There’s a cartoon that basically shows two windows
01:25:55 with a teller in each window.
01:25:58 One has a humongous line and the other one has no one.
01:26:02 The one that has no one above says health.
01:26:04 No, it says exercise and diet.
01:26:07 And the other one says pill.
01:26:10 And there’s a huge line for pill.
01:26:12 So we’re looking basically for magic bullets
01:26:13 for sort of ways that we can, you know,
01:26:16 beat cancer and beat coronavirus and beat this
01:26:18 and beat that.
01:26:19 And it turns out that the window with like,
01:26:21 just diet and exercise is the best way
01:26:23 to boost every aspect of your health.
01:26:26 If you look at Alzheimer’s, exercise and nutrition.
01:26:31 I mean, you’re like, really?
01:26:32 For my brain, neurodegeneration?
01:26:34 Absolutely.
01:26:36 If you look at cancer, exercise and nutrition.
01:26:40 If you look at coronavirus, exercise and nutrition,
01:26:43 every single aspect of human health gets improved.
01:26:47 And one of the studies we’re doing now
01:26:48 is basically looking at what are the effects
01:26:51 of diet and exercise?
01:26:52 How similar are they to each other?
01:26:55 We basically take in diet intervention
01:26:58 and exercise intervention in human and in mice.
01:27:01 And we’re basically doing single cell profiling
01:27:03 of a bunch of different tissues
01:27:04 to basically understand how are the cells,
01:27:08 both the stromal cells and the immune cells
01:27:10 of each of these tissues responding
01:27:13 to the effect of exercise.
01:27:15 What are the communication networks
01:27:16 between different cells?
01:27:18 Where the muscle that exercises sends signals
01:27:23 through the bloodstream, through the lymphatic system,
01:27:25 through all kinds of other systems
01:27:27 that give signals to other cells that I have exercised
01:27:31 and you should change in this particular way,
01:27:33 which basically reconfigure those receptor cells
01:27:37 with the effect of exercise.
01:27:39 How well understood is those reconfigurations?
01:27:43 Very little.
01:27:44 We’re just starting now, basically.
01:27:46 Is the hope there to understand the effect on,
01:27:52 so like the effect on the immune system?
01:27:54 On the immune system, the effect on brain,
01:27:56 the effect on your liver, on your digestive system,
01:27:59 on your adipocytes?
01:28:00 Adipose, the most misunderstood organ.
01:28:03 Everybody thinks, oh, fat, terrible.
01:28:05 No, fat is awesome.
01:28:07 Your fat cells is what’s keeping you alive
01:28:09 because if you didn’t have your fat cells,
01:28:11 all those lipids and all those calories
01:28:13 would be floating around in your blood
01:28:15 and you’d be dead by now.
01:28:16 Your adipocytes are your best friend.
01:28:18 They’re basically storing all these excess calories
01:28:21 so that they don’t hurt all of the rest of the body.
01:28:24 And they’re also fat burning in many ways.
01:28:28 So, again, when you don’t have
01:28:31 the homozygous version that I have,
01:28:33 your cells are able to burn calories much more easily
01:28:36 by sort of flipping a master metabolic switch
01:28:39 that involves this FTO locus that I mentioned earlier
01:28:42 and its target genes, RX3 and RX5,
01:28:45 that basically switch your adipocytes
01:28:47 during their three first days of differentiation
01:28:50 as they’re becoming mature adipocytes
01:28:52 to basically become either fat burning
01:28:54 or fat storing fat cells.
01:28:57 And the fat burning fat cells are your best friend.
01:28:58 They’re much closer to muscle
01:29:00 than they are to white adipocytes.
01:29:02 Is there a lot of difference between people
01:29:05 that you could give, science could eventually give advice
01:29:09 that is very generalizable
01:29:12 or is our differences in our genetic makeup,
01:29:16 like you mentioned, is that going to be basically
01:29:18 something we have to be very specialized individuals,
01:29:22 any advice we give in terms of diet,
01:29:24 like what we were just talking about?
01:29:25 Believe it or not, the most personalized advice
01:29:28 that you give for nutrition
01:29:29 don’t have to do with your genome.
01:29:31 They have to do with your gut microbiome,
01:29:34 with the bacteria that live inside you.
01:29:35 So most of your digestion is actually happening
01:29:37 by species that are not human inside you.
01:29:40 You have more nonhuman cells than you have human cells.
01:29:43 You’re basically a giant bag of bacteria
01:29:46 with a few human cells along.
01:29:48 And those do not necessarily have to do
01:29:53 with your genetic makeup.
01:29:54 They interact with your genetic makeup.
01:29:56 They interact with your epigenome.
01:29:58 They interact with your nutrition.
01:29:59 They interact with your environment.
01:30:01 They’re basically an additional source of variation.
01:30:07 So when you’re thinking about sort of
01:30:08 personalized nutritional advice,
01:30:10 part of that is actually how do you match your microbiome?
01:30:13 And part of that is how do we match your genetics?
01:30:17 But again, this is a very diverse set of contributors.
01:30:22 And the effect sizes are not enormous.
01:30:24 So I think the science for that is not fully developed yet.
01:30:27 Speaking of diets,
01:30:28 because I’ve wrestled in combat sports,
01:30:30 but sports my whole life were weight matters.
01:30:32 So you have to cut and all that stuff.
01:30:35 One thing I’ve learned a lot about my body,
01:30:38 and it seems to be, I think,
01:30:39 true about other people’s bodies,
01:30:41 is that you can adjust to a lot of things.
01:30:45 That’s the miraculous thing about this biological system,
01:30:48 is like I fast often.
01:30:52 I used to eat like five, six times a day
01:30:54 and thought that was absolutely necessary.
01:30:57 How could you not eat that often?
01:30:58 And then when I started fasting,
01:31:01 your body adjusted to that.
01:31:02 And you learn how to not eat.
01:31:04 And it was, if you just give it a chance
01:31:07 for a few weeks, actually,
01:31:09 over a period of a few weeks,
01:31:10 your body can adjust to anything.
01:31:11 And that’s a miraculous, that’s such a beautiful thing.
01:31:14 So I’m a computer scientist,
01:31:15 and I’ve basically gone through periods of 24 hours
01:31:18 without eating or stopping.
01:31:19 And then I’m like, oh, must eat.
01:31:22 And I eat a ton.
01:31:23 I used to order two pizzas just with my brother.
01:31:27 So I’ve gone through these extremes as well,
01:31:29 and I’ve gone the whole intermittent fasting thing.
01:31:32 So I can sympathize with you both on the seven meals a day
01:31:35 to the zero meals a day.
01:31:37 So I think when I say everything with moderation,
01:31:40 I actually think your body responds interestingly
01:31:44 to these different changes in diet.
01:31:47 I think part of the reason why we lose weight
01:31:49 with pretty much every kind of change in behavior
01:31:52 is because our epigenome and the set of proteins
01:31:55 and enzymes that are expressed and our microbiome
01:31:58 are not well suited to that nutritional source.
01:32:02 And therefore, they will not be able
01:32:03 to sort of catch everything that you give them.
01:32:06 And then a lot of that will go undigested.
01:32:09 And that basically means that your body can then
01:32:11 lose weight in the short term,
01:32:13 but very quickly will adjust to that new normal.
01:32:16 And then we’ll be able to sort of perhaps gain
01:32:18 a lot of weight from the diet.
01:32:20 So anyway, I mean, there’s also studies in factories
01:32:24 where basically people dim the lights
01:32:27 and then suddenly everybody started working better.
01:32:28 It was like, wow, that’s amazing.
01:32:30 Three weeks later, they made the lights a little brighter.
01:32:32 Everybody started working better.
01:32:34 So any kind of intervention has a placebo effect of,
01:32:39 wow, now I’m healthier and I’m gonna be running
01:32:41 more often, et cetera.
01:32:42 So it’s very hard to uncouple the placebo effect
01:32:44 of, wow, I’m doing something to intervene on my diet
01:32:47 from the, wow, this is actually the right thing for me.
01:32:50 So, you know.
01:32:51 Yeah, from the perspective from a nutrition science,
01:32:53 psychology, both things I’m interested in,
01:32:55 especially psychology, it seems that it’s extremely difficult
01:32:59 to do good science because there’s so many variables
01:33:03 involved, it’s so difficult to control the variables,
01:33:06 so difficult to do sufficiently large scale experiments,
01:33:10 both sort of in terms of the number of subjects
01:33:12 and temporal, like how long you do the study for,
01:33:17 that it just seems like it’s not even a real science
01:33:20 for now, like nutrition science.
01:33:22 I wanna jump into the whole placebo effect
01:33:24 for a little bit here.
01:33:25 And basically talk about the implications of that.
01:33:30 If I give you a sugar pill and I tell you it’s a sugar pill,
01:33:33 you won’t get any better.
01:33:35 But if I tell you a sugar pill and I tell you,
01:33:38 wow, this is an amazing drug,
01:33:40 it actually will stop your cancer,
01:33:42 your cancer will actually stop with much higher probability.
01:33:46 What does that mean?
01:33:47 That’s so amazing.
01:33:47 That means that if I can trick your brain
01:33:49 into thinking that I’m healing you,
01:33:51 your brain will basically figure out a way to heal itself,
01:33:54 to heal the body.
01:33:55 And that tells us that there’s so much
01:33:58 that we don’t understand in the interplay
01:34:01 between our cognition and our biology,
01:34:05 that if we were able to better harvest
01:34:08 the power of our brain to sort of impact the body
01:34:12 through the placebo effect,
01:34:14 we would be so much better in so many different things.
01:34:17 Just by tricking yourself into thinking
01:34:19 that you’re doing better, you’re actually doing better.
01:34:21 So there’s something to be said
01:34:22 about sort of positive thinking, about optimism,
01:34:25 about sort of just getting your brain
01:34:30 and your mind into the right mindset
01:34:33 that helps your body and helps your entire biology.
01:34:36 Yeah, from a science perspective, that’s just fascinating.
01:34:39 Obviously most things about the brain
01:34:41 is a total mystery for now,
01:34:43 but that’s a fascinating interplay
01:34:46 that the brain can help cure cancer.
01:34:54 I don’t even know what to do with that.
01:34:55 I mean, the way to think about that is the following.
01:34:59 The converse of the equation is something
01:35:01 that we are much more comfortable with.
01:35:03 Like, oh, if you’re stressed,
01:35:05 then your heart rate might rise
01:35:08 and all kinds of sort of toxins might be released
01:35:10 and that can have a detrimental effect in your body,
01:35:13 et cetera, et cetera, et cetera.
01:35:14 So maybe it’s easier to understand your body
01:35:18 healing from your mind
01:35:20 by your mind is not killing your body,
01:35:23 or at least it’s killing it less.
01:35:24 So I think that aspect of the stress equation
01:35:28 is a little easier for most of us to conceptualize,
01:35:31 but then the healing part is perhaps the same pathways,
01:35:35 perhaps different pathways,
01:35:36 but again, something that is totally untapped scientifically.
01:35:39 I think we try to bring this question up a couple of times,
01:35:42 but let’s return to it again,
01:35:44 is what do you think is the difference
01:35:46 between the way a computer represents information,
01:35:49 the human genome represents and stores information?
01:35:53 And maybe broadly, what is the difference
01:35:55 between how you think about computers
01:35:57 and how you think about biological systems?
01:36:00 So I made a very provocative claim earlier
01:36:02 that we are a digital computer.
01:36:04 Like I said, at the core lies a digital code
01:36:06 and that’s true in many ways,
01:36:07 but surrounding that digital core,
01:36:09 there’s a huge amount of analog.
01:36:11 If you look at our brain, it’s not really digital.
01:36:13 If you look at our sort of RNA
01:36:15 and all of that stuff inside our cell,
01:36:17 it’s not really digital.
01:36:18 It’s really analog in many ways,
01:36:21 but let’s start with the code
01:36:22 and then we’ll expand to the rest.
01:36:24 So the code itself is digital.
01:36:27 So there’s genes.
01:36:28 You can think of the genes as, I don’t know,
01:36:30 the procedures, the functions inside your language.
01:36:33 And then somehow you have to turn these functions on.
01:36:36 How do you call a gene?
01:36:37 How do you call that function?
01:36:39 The way that you would do it in old programming languages
01:36:41 is go to address whatever in your memory
01:36:44 and then you’d start running from there.
01:36:46 And modern programming languages
01:36:48 have encapsulated this into functions
01:36:50 and objects and all of that.
01:36:52 And it’s nice and cute, but in the end, deep down,
01:36:54 there’s still an assembly code
01:36:55 that says go to that instruction
01:36:57 and it runs that instruction.
01:36:59 If you look at the human genome
01:37:01 and the genome of pretty much most species out there,
01:37:06 there’s no go to function.
01:37:08 You just don’t start transcribing in position 13,000,
01:37:15 13,527 in chromosome 12.
01:37:18 You instead have content based indexing.
01:37:21 So at every location in the genome,
01:37:25 in front of the genes that need to be turned on,
01:37:28 I don’t know, when you drink coffee,
01:37:30 there’s a little coffee marker in front of all of them.
01:37:34 And whenever your cells that metabolize coffee
01:37:38 need to metabolize coffee,
01:37:39 they basically see coffee and they’re like,
01:37:41 ooh, let’s go turn on all the coffee marked genes.
01:37:44 So there’s basically these small motifs,
01:37:48 these small sequences that we call regulatory motifs.
01:37:50 They’re like patterns of DNA.
01:37:52 They’re only eight characters long or so,
01:37:54 like GAT, GCA, et cetera.
01:37:57 And these motifs work in combinations
01:38:01 and every one of them has some recruitment affinity
01:38:06 for a different protein that will then come and bind it
01:38:09 and together collections of these motifs
01:38:11 create regions that we call regulatory regions
01:38:15 that can be either promoters near the beginning of the gene
01:38:19 and that basically tells you
01:38:20 where the function actually starts, where you call it,
01:38:22 and then enhancers that are looping around of the DNA
01:38:26 that basically bring the machinery
01:38:28 that binds those enhancers
01:38:29 and then bring it onto the promoter,
01:38:32 which then recruits the right sort of the ribosome
01:38:36 and the polymerase and all of that thing,
01:38:37 which will first transcribe and then export
01:38:40 and then eventually translate in the cytoplasm,
01:38:42 you know, whatever RNA molecule.
01:38:45 So the beauty of the way
01:38:50 that the digital computer that’s the genome works
01:38:54 is that it’s extremely fault tolerant.
01:38:57 If I took your hard drive
01:38:59 and I messed with 20% of the letters in it,
01:39:03 of the zeros and ones and I flipped them,
01:39:05 you’d be in trouble.
01:39:07 If I take the genome and I flipped 20% of the letters,
01:39:11 you probably won’t even notice.
01:39:13 And that resilience.
01:39:15 That’s fascinating, yeah.
01:39:16 Is a key design principle.
01:39:18 And again, I’m anthropomorphizing here,
01:39:20 but it’s a key driving principle
01:39:22 of how biological systems work.
01:39:24 They’re first resilient and then anything else.
01:39:27 And when you look at this incredible beauty of life
01:39:32 from the most, I don’t know, beautiful,
01:39:35 I don’t know, human genome maybe of humanity
01:39:38 and all of the ideals that should come with it
01:39:40 to the most terrifying genome,
01:39:42 like, I don’t know, COVID 19, SARS COVID 2
01:39:45 and the current pandemic,
01:39:47 you basically see this elegance
01:39:50 as the epitome of clean design,
01:39:54 but it’s dirty.
01:39:55 It’s a mess.
01:39:57 It’s, you know, the way to get there is hugely messy.
01:40:02 And that’s something that we as computer scientists
01:40:04 don’t embrace.
01:40:06 We like to have clean code.
01:40:08 You know, like in engineering,
01:40:10 they teach you about compartmentalization,
01:40:12 about sort of separating functions,
01:40:13 about modularity, about hierarchical design.
01:40:17 None of that applies in biology.
01:40:19 Testing.
01:40:19 Testing, sure.
01:40:22 Yeah, biology does plenty of that.
01:40:24 But I mean, through evolutionary exploration.
01:40:26 But if you look at biological systems,
01:40:31 first they are robust
01:40:33 and then they specialize to become anything else.
01:40:36 And if you look at viruses,
01:40:38 the reason why they’re so elegant
01:40:41 when you look at the design of this, you know, genome,
01:40:44 it seems so elegant.
01:40:46 And the reason for that is that it’s been stripped down
01:40:49 from something much larger
01:40:51 because of the pressure to keep it compact.
01:40:53 So many compact genomes out there
01:40:56 have ancestors that were much larger.
01:40:58 You don’t start small and become big.
01:41:00 You go through a loop of add a bunch of stuff,
01:41:03 increase complexity, and then, you know, slim it down.
01:41:07 And one of my early papers was in fact on genome duplication.
01:41:12 One of the things we found is that baker’s yeast,
01:41:14 which is the, you know, yeast that you use to make bread,
01:41:17 but also the yeast that you use to make wine,
01:41:19 which is basically the dominant species
01:41:20 when you go in the fields of Tuscany
01:41:22 and you say, you know, what’s out there,
01:41:24 it’s basically saccharomyces cerevisiae,
01:41:26 or the way my Italian friends say,
01:41:27 saccharomyces cerevisiae.
01:41:30 So, so.
01:41:33 Oh, which means what?
01:41:34 Oh, saccharomyces, okay, I’m sorry, I’m Greek.
01:41:36 So yeah, zacharo, mikis, zacharo is sugar,
01:41:39 mikis is fungus.
01:41:41 Yes, cerevisiae, cerveza, beer.
01:41:44 So it means the sugar fungus of beer.
01:41:47 Yeah.
01:41:48 You know, less, less sounding to the ear.
01:41:51 Still poetic, yeah.
01:41:52 So anyway, saccharomyces cerevisiae,
01:41:54 basically the major baker’s yeast out there
01:41:57 is the descendant of a whole genome duplication.
01:42:00 Why would a whole gene duplication even happen?
01:42:02 When it happened is coinciding
01:42:06 with about a hundred million years ago
01:42:08 and the emergence of fruit bearing plants.
01:42:14 Why fruit bearing plants?
01:42:15 Because animals would eat the fruit
01:42:19 and would walk around and poop huge amounts of nutrients
01:42:23 along with a seed for the plants to spread.
01:42:26 Before that, plants were not spreading through animals,
01:42:29 they were spreading through wind
01:42:30 and all kinds of other ways.
01:42:32 But basically the moment you have fruit bearing plants,
01:42:34 these plants are basically creating this abundance
01:42:38 of sugar in the environment.
01:42:40 So there’s an evolutionary niche that gets created.
01:42:42 And in that evolutionary niche,
01:42:44 you basically have enough sugar
01:42:46 that a whole genome duplication,
01:42:48 which initially is a very messy event,
01:42:51 allows you to then, you know,
01:42:53 relieve some of that complexity.
01:42:56 So I had to pause, what does genome duplication mean?
01:42:59 That basically means that instead of having eight chromosomes,
01:43:03 you can now have 16 chromosomes.
01:43:06 So, but the duplication at first,
01:43:09 when you go to 16, you’re not using that.
01:43:13 Oh yeah, you are.
01:43:15 Yeah, so basically from one day to the next,
01:43:17 you went from having eight chromosomes
01:43:18 to having 16 chromosomes.
01:43:20 Probably a non disjunction event during a duplication,
01:43:22 during a division.
01:43:24 So you basically divide the cell
01:43:25 instead of half the genome going this way
01:43:27 and half the genome going the other way
01:43:29 after duplication of the genome,
01:43:30 you basically have all of it going to one cell
01:43:33 and then there’s sufficient messiness there
01:43:35 that you end up with slight differences
01:43:38 that make most of these chromosomes
01:43:39 be actually preserved.
01:43:42 It’s a long story short to me.
01:43:43 But that’s a big upgrade, right?
01:43:45 So that’s…
01:43:45 Not necessarily,
01:43:46 because what happens immediately thereafter
01:43:48 is that you start massively losing
01:43:50 tons of those duplicated genes.
01:43:52 So 90% of those genes were actually lost
01:43:55 very rapidly after whole gene duplication.
01:43:58 And the reason for that is that biology is not intelligent,
01:44:01 it’s just ruthless selection, random mutation.
01:44:06 So the ruthless selection basically means
01:44:08 that as soon as one of the random mutations hit one gene,
01:44:11 ruthless selection just kills off that gene.
01:44:13 It’s just,
01:44:16 if you have a pressure to maintain a small compact genome,
01:44:19 you will very rapidly lose the second copy
01:44:21 of most of your genes and a small number 10%
01:44:24 were kept in two copies.
01:44:25 And those had to do a lot with environment adaptation,
01:44:28 with the speed of replication,
01:44:31 with the speed of translation and with sugar processing.
01:44:34 So I’m making a long story short
01:44:36 to basically say that evolution is messy.
01:44:38 The only way…
01:44:39 Like, so the example that I was giving
01:44:42 of messing with 20% of your bits in your computer,
01:44:45 totally bogus.
01:44:47 Duplicating all your functions
01:44:48 and just throwing them out there in the same function,
01:44:51 just totally bogus.
01:44:52 Like this would never work in an engineer system.
01:44:55 But biological systems,
01:44:56 because of this content based indexing
01:44:59 and because of this modularity that comes
01:45:01 from the fact that the gene is controlled
01:45:04 by a series of tags.
01:45:05 And now if you need this gene in another setting,
01:45:08 you just add some more tags
01:45:09 that will basically turn it on also in those settings.
01:45:12 So this gene is now pressured to do two different functions
01:45:17 and it builds up complexity.
01:45:19 I see a whole gene duplication
01:45:21 and gene duplication in general
01:45:22 as a way to relieve that complexity.
01:45:24 So you have this gradual buildup of complexity
01:45:26 as functions get sort of added onto the existing genes.
01:45:30 And then boom, you duplicate your workforce.
01:45:34 And you now have two copies of this gene.
01:45:36 One will probably specialize to do one
01:45:38 and the other one will specialize to do the other
01:45:40 or one will maintain the ancestral function.
01:45:42 The other one will sort of be free to evolve
01:45:44 and specialize while losing the ancestral function
01:45:47 and so on and so forth.
01:45:48 So that’s how genomes evolve.
01:45:49 They’re just messy things,
01:45:52 but they’re extremely fault tolerant
01:45:54 and they’re extremely able to deal with mutations
01:45:58 because that’s the very way that you generate new functions.
01:46:03 So new functionalization comes
01:46:05 from the very thing that breaks it.
01:46:07 So even in the current pandemic,
01:46:09 many people are asking me which mutations matter the most.
01:46:12 And what I tell them is,
01:46:13 well, we can study the evolutionary dynamics
01:46:16 of the current genome to then understand
01:46:19 which mutations have previously happened or not.
01:46:23 And which mutations happen in genes
01:46:26 that evolve rapidly or not.
01:46:28 And one of the things we found, for example,
01:46:29 is that the genes that evolved rapidly in the past
01:46:33 are still evolving rapidly now in the current pandemic.
01:46:36 The genes that evolved slowly in the past
01:46:38 are still evolving slowly.
01:46:40 Which means that they’re useful?
01:46:41 Which means that they’re under
01:46:43 the same evolutionary pressures.
01:46:45 But then the question is what happens in specific mutations?
01:46:49 So if you look at the D614 gene mutations,
01:46:52 that’s been all over the news.
01:46:53 So in position 614, in the amino acids 614 of the S protein,
01:46:59 there’s a D2 gene mutation
01:47:02 that sort of has creeped over the population.
01:47:07 That mutation, we found out through my work,
01:47:10 disrupts a perfectly conserved nucleotide position
01:47:13 that has never been changed in the history
01:47:15 of millions of years of equivalent
01:47:17 per million evolution of these viruses.
01:47:23 That basically means that it’s a completely new adaptation
01:47:25 to human.
01:47:27 And that mutation has now gone from 1% frequency
01:47:30 to 90% frequency in almost all outbreaks.
01:47:33 So this mutation, I like how you say the 416,
01:47:38 what was it, okay.
01:47:39 Yeah, 614, sorry.
01:47:40 614.
01:47:41 D614G.
01:47:43 D614, so literally, so what you’re saying
01:47:46 is this is like a chess move.
01:47:48 So it just mutated one letter to another.
01:47:50 Exactly.
01:47:51 And that hasn’t happened before.
01:47:53 Yeah, never.
01:47:54 And this somehow, this mutation is really useful.
01:47:58 It’s really useful in the current environment of the genome,
01:48:02 which is moving from human to human.
01:48:04 When it was moving from bat to bat,
01:48:06 it couldn’t care less for that mutation,
01:48:08 but it’s environment specific.
01:48:09 So now that it’s moving from human to human,
01:48:12 it’s moving way better, like by orders of magnitude.
01:48:15 What do you, okay, so you’re like tracking
01:48:18 this evolutionary dynamics, which is fascinating,
01:48:22 but what do you do with that?
01:48:24 So what does that mean?
01:48:25 What does this mean, what do you make,
01:48:27 what do you make of this mutation
01:48:29 in trying to anticipate, I guess,
01:48:31 is one of the things you’re trying to do
01:48:34 is anticipate where, how this unrolls into the future,
01:48:37 this evolutionary dynamics.
01:48:39 Such a great question.
01:48:40 So there’s two things.
01:48:42 Remember when I was saying earlier,
01:48:44 mutation is the path to new things,
01:48:47 but also the path to break old things.
01:48:49 So what we know is that this position
01:48:52 was extremely preserved through gazillions of mutations.
01:48:56 That mutation was never tolerated
01:48:58 when it was moving from bats to bats.
01:49:00 So that basically means that that position
01:49:02 is extremely important in the function of that protein.
01:49:05 That’s the first thing it tells.
01:49:06 The second one is that that position
01:49:09 was very well suited to bat transmission,
01:49:12 but now is not well suited to human transmission,
01:49:14 so it got rid of it.
01:49:15 And it now has a new version of that amino acid
01:49:18 that basically makes it much easier
01:49:20 to transmit from human to human.
01:49:22 So in terms of the evolutionary history
01:49:27 teaching us about the future,
01:49:29 it basically tells us here’s the regions
01:49:31 that are currently mutating.
01:49:34 Here’s the regions that are most likely
01:49:36 to mutate going forward.
01:49:37 As you’re building a vaccine,
01:49:39 here’s what you should be focusing on
01:49:41 in terms of the most stable regions
01:49:43 that are the least likely to mutate.
01:49:45 Or here’s the newly evolved functions
01:49:48 that are the most likely to be important
01:49:50 because they’ve overcome this local maximum
01:49:54 that it had reached in the bat transmission.
01:49:59 So anyway, it’s a tangent to basically say
01:50:01 that evolution works in messy ways.
01:50:04 And the thing that you would break
01:50:07 is the thing that actually allows you
01:50:10 to first go through a lull
01:50:12 and then reaching new local maximum.
01:50:15 And I often like to say that if engineers
01:50:18 had basically designed evolution,
01:50:21 we would still be perfectly replicating bacteria
01:50:26 because it’s my making the bacterium worse
01:50:29 that you allow evolution to reach a new optimum.
01:50:32 That’s, just to pause on that,
01:50:34 that’s so profound.
01:50:35 That’s so profound for the entirety
01:50:39 of this scientific and engineering disciplines.
01:50:44 Exactly.
01:50:45 We as engineers need to embrace breaking things.
01:50:48 We as engineers need to embrace robustness
01:50:50 as the first principle beyond perfection
01:50:54 because nothing’s gonna ever be perfect.
01:50:56 And when you’re sending a satellite to Mars,
01:50:58 when something goes wrong, it’ll break down.
01:51:01 As opposed to building systems that tolerate failure
01:51:04 and are resilient to that.
01:51:08 And in fact, get better through that.
01:51:11 So the SpaceX approach versus NASA for the…
01:51:14 For example.
01:51:16 Is there something we can learn about the incredible,
01:51:21 take lessons from the incredible biological systems
01:51:23 in their resilience, in the mushiness, the messiness
01:51:27 to our computing systems, to our computers?
01:51:31 It would basically be starting from scratch in many ways.
01:51:35 It would basically be building new paradigms
01:51:38 that don’t try to get the right answer all the time,
01:51:42 but try to get the right answer most of the time
01:51:45 or a lot of the time.
01:51:47 Do you see deep learning systems in the whole world
01:51:49 of machine learning as kind of taking a step
01:51:51 in that direction?
01:51:52 Absolutely, absolutely.
01:51:53 Basically by allowing this much more natural evolution
01:51:57 of these parameters, you basically…
01:52:01 And if you look at sort of deep learning systems again,
01:52:04 they’re not inspired by the genome aspect of biology,
01:52:07 they’re inspired by the brain aspect of biology.
01:52:10 And again, I want you to pause for a second
01:52:12 and realize the complexity of the entire human brain
01:52:18 with trillions of connections within our neurons,
01:52:22 with millions of cells talking to each other,
01:52:26 is still encoded within that same genome.
01:52:29 That same genome encodes every single freaking cell type
01:52:36 of the entire body.
01:52:37 Every single cell is encoded by the same code.
01:52:41 And yet specialization allows you to have
01:52:45 the single viral like genome that self replicates,
01:52:50 the single module, modular automaton,
01:52:54 work with other copies of itself, it’s mind boggling.
01:52:57 Create complex organs through which blood flows.
01:53:01 And what is that blood?
01:53:02 The same freaking genome.
01:53:05 Create organs that communicate with each other.
01:53:09 And what are these organs?
01:53:11 The exact same genome.
01:53:13 Create a brain that is innervated by massive amounts
01:53:17 of blood pumping energy to it,
01:53:21 20% of our energetic needs to the brain from the same genome.
01:53:28 And all of the neuronal connections,
01:53:30 all of the auxiliary cells, all of the immune cells,
01:53:33 the astrocytes, the ligodendrocytes, the neurons,
01:53:35 the excitatory, the inhibitory neurons,
01:53:37 all of the different classes of parasites,
01:53:39 the blood brain barrier, all of that, same genome.
01:53:42 One way to see that in a sad, this one is beautiful.
01:53:47 The sad thing is thinking about the trillions
01:53:50 of organisms that died to create that.
01:53:55 You mean on the evolutionary path to humans?
01:53:57 On the evolutionary path to humans.
01:53:59 It’s crazy, there’s two descendant of apes
01:54:02 just talking on a podcast.
01:54:04 Okay, it’s just so mind boggling.
01:54:08 Just to boggle our minds a little bit more.
01:54:11 Us talking to each other,
01:54:13 we are basically generating a series of vocal utterances
01:54:18 through our pulsating of vocal cords received through this.
01:54:23 The people who listen to this
01:54:26 are taking a completely different path
01:54:29 to that information transfer, yet through language.
01:54:32 But imagine if we could connect these brains
01:54:36 directly to each other.
01:54:38 The amount of information that I’m condensing
01:54:41 into a small number of words is a huge funnel,
01:54:46 which then you receive and you expand
01:54:49 into a huge number of thoughts from that small funnel.
01:54:55 In many ways, engineers would love
01:54:58 to have the whole information transfer,
01:54:59 just take the whole set of neurons and throw them away.
01:55:02 I mean, throw them to the other person.
01:55:05 This might actually not be better
01:55:07 because in your misinterpretation
01:55:10 of every word that I’m saying,
01:55:13 you are creating new interpretation
01:55:14 that might actually be way better
01:55:16 than what I meant in the first place.
01:55:17 The ambiguity of language perhaps
01:55:21 might be the secret to creativity.
01:55:25 Every single time you work on a project by yourself,
01:55:28 you only bounce ideas with one person
01:55:31 and your neurons are basically fully cognizant
01:55:33 of what these ideas are.
01:55:35 But the moment you interact with another person,
01:55:37 the misinterpretations that happen
01:55:41 might be the most creative part of the process.
01:55:43 With my students, every time we have a research meeting,
01:55:45 I very often pause and say,
01:55:47 let me repeat what you just said in a different way.
01:55:50 And I sort of go on and brainstorm
01:55:52 with what they were saying,
01:55:53 but by the third time,
01:55:55 it’s not what they were saying at all.
01:55:58 And when they pick up what I’m saying,
01:55:59 they’re like, oh, well, dah, dah, dah.
01:56:01 Now they’ve sort of learned something very different
01:56:04 from what I was saying.
01:56:05 And that is the same kind of messiness
01:56:08 that I’m describing in the genome itself.
01:56:10 It’s sort of embracing the messiness.
01:56:13 And that’s a feature, not a book.
01:56:15 Exactly.
01:56:16 And in the same way, when you’re thinking
01:56:17 about sort of these deep learning systems
01:56:19 that will allow us to sort of be more creative perhaps
01:56:23 or learn better approximations of these complex functions,
01:56:27 again, tuned to the universe that we inhabit,
01:56:30 you have to embrace the breaking.
01:56:33 You have to embrace the,
01:56:35 how do we get out of these local optima?
01:56:38 And a lot of the design paradigms
01:56:40 that have made deep learning so successful
01:56:43 are ways to get away from that,
01:56:45 ways to get better training
01:56:47 by sort of sending long range messages,
01:56:50 these LSTM models and the sort of feed forward loops
01:56:55 that sort of jump through layers
01:56:59 of a convolutional neural network.
01:57:00 All of these things are basically ways to push you out
01:57:04 of these local maxima.
01:57:07 And that’s sort of what evolution does.
01:57:08 That’s what language does.
01:57:09 That’s what conversation and brainstorming does.
01:57:12 That’s what our brain does.
01:57:14 So this design paradigm is something that’s pervasive
01:57:18 and yet not taught in schools,
01:57:20 not taught in engineering schools
01:57:22 where everything’s minutely modularized
01:57:24 to make sure that we never deviate
01:57:26 from whatever signal we’re trying to emit
01:57:28 as opposed to let all hell breaks loose
01:57:31 because that’s the path to paradise.
01:57:34 The path to paradise.
01:57:35 Yeah, I mean, it’s difficult to know how to teach that
01:57:38 and what to do with it.
01:57:39 I mean, it’s difficult to know how to build up
01:57:43 the scientific method around messiness.
01:57:46 I mean, it’s not all messiness.
01:57:49 We need some cleanness.
01:57:51 And going back to the example with Mars,
01:57:54 that’s probably the place where I want
01:57:55 to sort of moderate error as much as possible
01:57:58 and sort of control the environment as much as possible.
01:58:01 But if you’re trying to repopulate Mars,
01:58:03 well, maybe messiness is a good thing then.
01:58:05 On that, you quickly mentioned this
01:58:09 in terms of us using our vocal cords
01:58:12 to speak on a podcast.
01:58:15 So Elon Musk and Neuralink are working
01:58:17 on trying to plug, as per our discussion
01:58:22 with computers and biological systems,
01:58:24 to connect the two.
01:58:25 He’s trying to connect our brain to a computer
01:58:30 to create a brain computer interface
01:58:32 where they can communicate back and forth.
01:58:36 On this line of thinking, do you think this is possible
01:58:40 to bridge the gap between our engineered computing systems
01:58:45 and the messy biological systems?
01:58:49 My answer would be absolutely.
01:58:51 You know, there’s no doubt that we can understand
01:58:54 more and more about what goes on in the brain
01:58:57 and we can sort of train the brain.
01:59:00 I don’t know if you remember the Palm Pilot.
01:59:03 Yeah, Palm Pilot, yeah.
01:59:04 Remember this whole sort of alphabet that they had created?
01:59:08 Am I thinking of the same thing?
01:59:10 It’s basically, you had a little pen
01:59:13 and for every character, you had a little scribble
01:59:17 that was unique that the machine could understand.
01:59:19 And that instead of trying the machine
01:59:22 and trying to teach the machine
01:59:23 to recognize human characters,
01:59:25 you had basically, they figured out
01:59:27 that it’s better and easier to train humans
01:59:29 to create human like characters
01:59:31 that the machine is better at recognizing.
01:59:34 So in the same way, I think what will happen
01:59:38 is that humans will be trained
01:59:40 to be able to create the mind pattern
01:59:43 that the machine will respond to
01:59:45 before the machine truly comprehends our thoughts.
01:59:47 So the first human brain interfaces
01:59:50 will be tricking humans to speak the machine language
01:59:53 where with the right set of electrodes,
01:59:55 I can sort of trick my brain into doing this.
01:59:57 And this is the same way that many people teach,
02:00:00 like learn to control artificial limbs.
02:00:02 You basically try a bunch of stuff
02:00:04 and eventually you figure out how your limbs work.
02:00:06 That might not be very different
02:00:08 from how humans learn to use their natural limbs
02:00:11 when they first grow up.
02:00:13 Basically, you have these, you know,
02:00:14 neoteny period of, you know,
02:00:17 this puddle of soup inside your brain,
02:00:21 trying to figure out how to even make neural connections
02:00:23 before you’re born and then learning sounds
02:00:27 in utero of, you know, all kinds of echoes
02:00:31 and, you know, eventually getting out in the real world.
02:00:35 And I don’t know if you’ve seen newborns,
02:00:37 but they just stare around a lot.
02:00:39 You know, one way to think about this
02:00:41 as a machine learning person is,
02:00:43 oh, they’re just training their edge detectors.
02:00:46 And eventually they figure out
02:00:47 how to train their edge detectors.
02:00:48 They work through the second layer of the visual cortex
02:00:50 and the third layer and so on and so forth.
02:00:52 And you basically have this learning
02:00:58 how to control your limbs
02:00:59 that probably comes at the same time.
02:01:01 You’re sort of, you know, throwing random things there
02:01:03 and you realize that, oh, wow,
02:01:04 when I do this thing, my limb moves.
02:01:08 Let’s do the following experiment.
02:01:09 Take a breath.
02:01:11 What muscles did you flex?
02:01:13 Now take another breath and think what muscles do I flex?
02:01:16 The first thing that you’re thinking
02:01:17 when you’re taking a breath
02:01:19 is the impact that it has on your lungs.
02:01:22 You’re like, oh, I’m now gonna increase my lungs
02:01:24 or I’m not gonna bring air in.
02:01:25 But what you’re actually doing
02:01:26 is just changing your diaphragm.
02:01:29 That’s not conscious, of course.
02:01:31 You never think of the diaphragm as a thing.
02:01:34 And why is that?
02:01:36 That’s probably the same reason
02:01:37 why I think of moving my finger
02:01:38 when I actually move my finger.
02:01:40 I think of the effect instead of actually thinking
02:01:42 of whatever muscle is twitching
02:01:44 that actually causes my finger to move.
02:01:46 So we basically in our first years of life
02:01:49 build up this massive lookup table
02:01:52 between whatever neuronal firing we do
02:01:55 and whatever action happens in our body that we control.
02:02:00 If you have a kid grow up with a third limb,
02:02:04 I’m sure they’ll figure out how to control them
02:02:06 probably at the same rate as their natural limbs.
02:02:09 And a lot of the work would be done by the…
02:02:13 If a third limb is a computer,
02:02:15 you kind of have a, not a faith, but a thought
02:02:20 that the brain might be able to figure out…
02:02:24 The plasticity would come from the brain.
02:02:26 The brain would be cleverer than the machine at first.
02:02:28 When I talk about a third limb,
02:02:29 that’s exactly what I’m saying, an artificial limb
02:02:32 that basically just controls your mouse while you’re typing.
02:02:35 Perfectly natural thing.
02:02:36 I mean, again, in a few hundred years.
02:02:40 Maybe sooner than that.
02:02:41 But basically, as long as the machine is consistent
02:02:46 in the way that it will respond to your brain impulses,
02:02:49 you’ll figure out how to control that
02:02:51 and you could play tennis with your third limb.
02:02:53 And let me go back to consistency.
02:02:57 People who have dramatic accidents
02:03:01 that basically take out a whole chunk of their brain
02:03:03 can be taught to coopt other parts of the brain
02:03:07 to then control that part.
02:03:08 You can basically build up that tissue again
02:03:10 and eventually train your body how to walk again
02:03:13 and how to read again and how to play again
02:03:15 and how to think again, how to speak a language again,
02:03:17 et cetera.
02:03:18 So there’s a massive amount of malleability
02:03:21 that happens naturally in our way of controlling our body,
02:03:26 our brain, our thoughts, our vocal cords, our limbs,
02:03:29 et cetera.
02:03:30 And human machine interfaces are inevitable
02:03:35 if we sort of figure out how to read these electric impulses,
02:03:39 but the resolution at which we can understand human thought
02:03:43 right now is nil, is ridiculous.
02:03:46 So how are human thoughts encoded?
02:03:49 It’s basically combinations of neurons that cofire
02:03:53 and these create these things called engrams
02:03:55 that eventually form memories and so on and so forth.
02:03:58 We know nothing of all that stuff.
02:04:01 So before we can actually read into your brain
02:04:05 that you wanna build a program
02:04:06 that does this and this and this and that,
02:04:08 we need a lot of neuroscience.
02:04:10 Well, so to push back on that,
02:04:13 do you think it’s possible that without understanding
02:04:16 the functionally about the brain or from the neuroscience
02:04:20 or the cognitive science or psychology,
02:04:22 whichever level of the brain we’ll look at,
02:04:24 do you think if we just connect them,
02:04:26 just like per your previous point,
02:04:29 if we just have a high enough resolution
02:04:30 between connection between a Wikipedia and your brain,
02:04:34 the brain will just figure it out with us understanding
02:04:38 because that’s one of the innovations of Neuralink
02:04:40 is they’re increasing the number of connections
02:04:43 to the brain to like several thousand,
02:04:45 which before was in the dozens or whatever.
02:04:48 You’re still off by a few orders of magnitude
02:04:51 on the order of seven.
02:04:52 Right, but the thing is, the hope is if you increase
02:04:57 that number more and more and more,
02:04:58 maybe you don’t need to understand anything
02:05:00 about the actual how human thought
02:05:03 is represented in the brain.
02:05:04 You can just let it figure it out by itself.
02:05:08 Keanu Reeves waking up and saying, I know cook food.
02:05:10 Yeah, exactly.
02:05:13 So yeah, sure.
02:05:14 You don’t have faith in the plasticity of the brain
02:05:16 to that degree.
02:05:18 It’s not about brain plasticity.
02:05:19 It’s about the input aspect.
02:05:21 Basically, I think on the output aspect,
02:05:23 being able to control a machine is something
02:05:25 that you can probably train your neural impulses
02:05:28 that you’re sending out to sort of match
02:05:30 whatever response you see in the environment.
02:05:33 If this thing moved every single time I thought
02:05:35 a particular thought, then I could figure out,
02:05:37 I could hack my way into moving this thing
02:05:39 with just a series of thoughts.
02:05:40 I could think guitar, piano, tennis ball,
02:05:45 and then this thing would be moving.
02:05:47 And then I would just have the series of thoughts
02:05:50 that would sort of result in the impulses
02:05:52 that will move this thing the way that I want it.
02:05:54 And then eventually it’ll become natural
02:05:55 because I won’t even think about it.
02:05:57 I mean, in the same way that we control our limbs
02:05:59 in a very natural way, but babies don’t do that.
02:06:01 Babies have to figure it out.
02:06:03 And some of that is hard coded,
02:06:04 but some of that is actually learned
02:06:06 based on whatever soup of neurons you ended up with,
02:06:10 whatever connections you pruned them to,
02:06:13 and eventually you were born with.
02:06:15 A lot of that is coded in the genome,
02:06:17 but a huge chunk of that is stochastic.
02:06:19 And sort of the way that you sort of create
02:06:21 all these neurons, they migrate, they form connections,
02:06:23 they sort of spread out,
02:06:25 they have particular branching patterns,
02:06:26 but then the connectivity itself,
02:06:28 unique in every single new person.
02:06:30 All this to say that on the output side,
02:06:34 absolutely, I’m very, very, you know,
02:06:36 hopeful that we can have machines
02:06:38 that read thousands of these neuronal connections
02:06:41 on the output side, but on the input side, oh boy.
02:06:47 I don’t expect any time in the near future
02:06:51 we’ll be able to sort of send a series of impulses
02:06:53 that will tell me, oh, earth to sun distance,
02:06:56 7.5 million, et cetera, et cetera.
02:06:58 Like nowhere.
02:07:00 I mean, I think language will still be the input way
02:07:04 rather than sort of any kind of more complex.
02:07:07 It’s a really interesting notion
02:07:08 that the ambiguity of language is a feature.
02:07:12 And we evolved for millions of years
02:07:16 to take advantage of that ambiguity.
02:07:19 Exactly.
02:07:20 And yet no one teaches us the subtle differences
02:07:23 between words that are near cognates,
02:07:26 and yet evoke so much more than, you know,
02:07:29 one from the other.
02:07:30 And yet, you know, when you’re choosing words
02:07:34 from a list of 20 synonyms,
02:07:36 you know exactly the connotation
02:07:38 of every single one of them.
02:07:40 And that’s something that, you know, is there.
02:07:42 So yes, there’s ambiguity,
02:07:45 but there’s all kinds of connotations.
02:07:46 And in the way that we select our words,
02:07:48 we have so much baggage that we’re sending along,
02:07:51 the way that we’re emoting,
02:07:52 the way that we’re moving our hands
02:07:54 every single time we speak,
02:07:56 the, you know, the pauses, the eye contact, et cetera.
02:07:58 So much higher baud rate than just a vocal,
02:08:01 you know, string of characters.
02:08:04 Well, let me just take a small tangent on that.
02:08:07 Oh, tangent?
02:08:07 We haven’t done that yet.
02:08:08 It’s a good idea.
02:08:09 Let’s do a tangent.
02:08:10 We’ll return to the origin of life after.
02:08:16 So, I mean, you’re Greek,
02:08:17 but I’m going on this personal journey.
02:08:20 I’m going to Paris for the explicit purpose
02:08:25 of talking to one of the most famous,
02:08:29 a couple who’s a famous translators of Russian literature,
02:08:33 Dostoevsky, Tolstoy, and they go,
02:08:36 that’s their art is the translation.
02:08:38 And everything I’ve learned about the translation art,
02:08:44 it makes me feel,
02:08:46 it’s so profound in a way that’s so much more profound
02:08:53 than the natural language processing papers
02:08:55 I read in the machine learning community,
02:08:57 that there’s such depth to language
02:09:00 that I don’t know what to do with.
02:09:03 I don’t know if you’ve experienced that in your own life
02:09:05 with knowing multiple languages.
02:09:08 I don’t know what to,
02:09:09 I don’t know how to make sense of it,
02:09:11 but there’s so much loss in translation
02:09:13 between Russian and English,
02:09:15 and getting a sense of that.
02:09:17 Like, for example,
02:09:19 there’s like just taking a single sentence
02:09:22 from Dostoevsky, and like, there’s a lot of them.
02:09:25 You could talk for hours
02:09:27 about how to translate that sentence properly.
02:09:30 That captures the meaning, the period,
02:09:34 the culture, the humor, the wit,
02:09:36 the suffering that was in the context of the time,
02:09:39 all of that could be a single sentence.
02:09:42 You could talk forever about what it takes
02:09:46 to translate that correctly.
02:09:47 I don’t know what to do with that.
02:09:48 So being Greek, it’s very hard for me
02:09:51 to think of a sentence or even a word
02:09:54 without going into the full etymology of that word,
02:09:59 breaking up every single atom of that sentence
02:10:04 and every single atom of these words
02:10:07 and rebuilding it back up.
02:10:09 I have three kids.
02:10:11 And the way that I teach them Greek
02:10:13 is the same way that, you know,
02:10:16 the documentary I was mentioning earlier
02:10:17 about sort of understanding the deep roots
02:10:19 of all of these, you know, words.
02:10:23 And it’s very interesting
02:10:29 that every single time I hear a new word
02:10:31 that I’ve never heard before,
02:10:33 I go and figure out the etymology of that word
02:10:34 because I will never appreciate that word
02:10:36 without understanding how it was initially formed.
02:10:40 Interesting, but how does that help?
02:10:42 Because that’s not the full picture.
02:10:44 No, no, of course, of course.
02:10:44 But what I’m trying to say is that knowing the components
02:10:48 teaches you about the context of the formation of that word
02:10:52 and sort of the original usage of that word.
02:10:54 And then of course the word takes new meaning
02:10:57 as you create it, you know, from its parts.
02:11:00 And that meaning then gets augmented.
02:11:04 And two synonyms that sort of have different roots
02:11:08 will actually have implications
02:11:09 that carry a lot of that baggage
02:11:11 of the historical provenance of these words.
02:11:14 So before working on genome evolution,
02:11:16 my passion was evolution of language
02:11:19 and sort of tracing cognates across different languages
02:11:24 through their etymologies.
02:11:27 That’s fascinating that there’s parallels between,
02:11:30 I mean, the idea that there’s evolutionary dynamics
02:11:34 to our language.
02:11:35 Yeah, every single word that you utter, parallels, parallels.
02:11:41 What does parallels mean?
02:11:42 Para means side by side.
02:11:44 Alleles from alleles, which means identical twins.
02:11:48 Parallels.
02:11:49 I mean, name any word and there’s so much baggage,
02:11:53 so much beauty in how that word came to be
02:11:56 and how this word took a new meaning
02:11:58 than the sum of its parts.
02:12:02 Yeah, and there’s just, there’s so many different words
02:12:05 that are just words.
02:12:06 They don’t have any physical grounding.
02:12:08 And now you take these words
02:12:10 and you weave them into a sentence.
02:12:13 The emotional invocations of that weaving are fathomless.
02:12:19 And all of those emotions all live in the brains of humans.
02:12:25 In the eye of the beholder.
02:12:28 No, seriously, you have to embrace this concept
02:12:30 of the eye of the beholder.
02:12:32 It’s the conceptualization that nothing takes meaning
02:12:37 with one person creating it.
02:12:39 Everything takes meaning in the receiving end
02:12:42 and the emergent properties of these communication networks
02:12:47 where every single, you know,
02:12:49 if you look at the network of our cells
02:12:50 and how they’re communicating with each other,
02:12:52 every cell has its own code.
02:12:54 This code is modulated by the epigenome.
02:12:56 This creates a bunch of different cell types.
02:12:57 Each cell type now has its own identity.
02:12:59 Yet they all have the common root of the stem cells
02:13:02 that sort of led to them.
02:13:04 Each of these identities is now communicating
02:13:06 with each other.
02:13:08 They take meaning in their interaction.
02:13:11 There’s an emergent property that comes
02:13:13 from a bunch of cells being together
02:13:15 that is not in any one of the parts.
02:13:17 If you look at neurons communicating,
02:13:19 again, these engrams don’t exist in any one neuron.
02:13:23 They exist in the connection and the combination of neurons.
02:13:26 And the meaning of the words that I’m telling you
02:13:29 is empty until it reaches you
02:13:31 and it affects you in a very different way
02:13:34 than it affects whoever’s listening
02:13:35 to this conversation now.
02:13:37 Because of the emotional baggage that I’ve grown up with,
02:13:40 that you’ve grown up with, and that they’ve grown up with.
02:13:43 And that’s, I think, the magic of translation.
02:13:46 If you start thinking of translation
02:13:48 as just simply capturing that emotional set of reactions
02:13:53 that you evoke, you need a different set of words
02:13:57 to evoke that same set of reactions to a French person
02:14:01 than to a Russian person,
02:14:02 because of the baggage of the culture that we grew up in.
02:14:05 Yeah, I mean, there’s…
02:14:07 So basically, you shouldn’t find the best word.
02:14:10 Sometimes it’s a completely different sentence structure
02:14:13 that you will need,
02:14:15 matched to the cultural context
02:14:18 of the target audience that you have.
02:14:20 Yeah, there’s a lot of different words
02:14:22 in the target audience that you have.
02:14:23 Yeah, it’s, I mean, you’re just…
02:14:26 I usually don’t think about this,
02:14:27 but right now, there’s this feeling,
02:14:30 as a reminder, that it’s just you and I talking,
02:14:32 but there’s several hundred thousand people
02:14:35 will listen to this.
02:14:36 There’s some guy in Russia right now running,
02:14:40 like in Moscow, listening to us.
02:14:44 There’s somebody in India, I guarantee you.
02:14:46 There’s somebody in China and South America.
02:14:48 There’s somebody in Texas,
02:14:51 they all have different…
02:14:53 Emotional baggage.
02:14:54 They probably got angry earlier on
02:14:56 about the whole discussion about coronavirus
02:14:58 and about some aspect of it.
02:15:02 Yeah, and there’s that network effect that’s…
02:15:06 It’s a beautiful thing.
02:15:08 And this lateral transfer of information,
02:15:10 that’s what makes the collective, quote unquote,
02:15:12 genome of humanity so unique from any other species.
02:15:17 Yeah.
02:15:19 So you somehow miraculously wrapped it back
02:15:22 to the very beginning of when we were talking
02:15:25 about the beauty of the human genome.
02:15:29 So I think this is the right time,
02:15:31 unless we wanna go for a six to eight hour conversation.
02:15:34 We’re gonna have to talk again,
02:15:36 but I think for now, to wrap it up,
02:15:39 this is the right time to talk about
02:15:41 the biggest, most ridiculous question of all,
02:15:44 meaning of life.
02:15:45 Off mic, you mentioned to me
02:15:47 that you had your 42nd birthday.
02:15:52 42nd being a very special, absurdly special number.
02:15:58 And you had a kind of get together with friends
02:16:03 to discuss the meaning of life.
02:16:04 So let me ask you,
02:16:05 in your, as a biologist, as a computer scientist,
02:16:09 and as a human, what is the meaning of life?
02:16:14 I’ve been asking this question for a long time,
02:16:18 ever since my 42nd birthday,
02:16:21 but well before that,
02:16:22 in even planning the meaning of life symposium.
02:16:25 And symposium, sim means together,
02:16:29 posy actually means to drink together.
02:16:31 So symposium is actually a drinking party.
02:16:33 So the meaning.
02:16:36 Can you actually elaborate about this meaning of life
02:16:37 symposium that you put together?
02:16:39 It’s like the most genius idea I’ve ever heard.
02:16:42 So 42 is obviously the answer to life,
02:16:44 the universe and everything,
02:16:45 from the Hitchhiker’s Guide to the Galaxy.
02:16:47 And as I was turning 42,
02:16:49 I’ve had the theme for every one of my birthdays.
02:16:51 When I was turning 32, it’s one, zero, zero, zero, zero, zero
02:16:55 in binary.
02:16:56 So I celebrated my 100,000th binary birthday,
02:17:00 and I had a theme of going back 100,000 years,
02:17:03 let’s dress something in the last 100,000 years.
02:17:07 Anyway, it was, I’ve always had these.
02:17:09 It’s such an interesting human being.
02:17:12 Okay, that’s awesome.
02:17:13 I’ve always had these sort of numerology
02:17:17 related announcements for my birthday parties.
02:17:21 So what came out of that meaning of life symposium
02:17:27 is that I basically asked 42 of my colleagues,
02:17:29 42 of my friends, 42 of my collaborators,
02:17:33 to basically give seven minutes species
02:17:35 on the meaning of life, each from their perspective.
02:17:38 And I really encourage you to go there
02:17:40 because it’s mind boggling
02:17:42 that every single person said a different answer.
02:17:46 Every single person started with,
02:17:48 I don’t know what the meaning of life is, but,
02:17:50 and then give this beautifully eloquently answer,
02:17:54 eloquent answer.
02:17:55 And they were all different,
02:17:57 but they all were consistent with each other
02:18:01 and mutually synergistic and together forming
02:18:04 a beautiful view of what it means to be human in many ways.
02:18:08 Some people talked about the loss of their loved one,
02:18:12 their life partner for many, many years
02:18:14 and how their life changed through that.
02:18:16 Some people talked about the origin of life.
02:18:19 Some people talked about the difference
02:18:21 between purpose and meaning.
02:18:24 I’ll maybe quote one of the answers,
02:18:28 which is this linguistics professor,
02:18:30 friend of mine at Harvard, who basically said,
02:18:35 that she was gonna, she’s Greek as well.
02:18:37 And she said, I will give a very Pythian answer.
02:18:40 So Pythia was the Oracle of Delphi,
02:18:42 who would basically give these very cryptic answers,
02:18:45 very short, but interpretable in many different ways.
02:18:48 There was this whole set of priests
02:18:50 who were tasked with interpreting what Pythia had said.
02:18:53 And very often you would not get a clean interpretation,
02:18:56 but she said, I will be like Pythia
02:18:59 and give you a very short and multiply interpretable answer.
02:19:02 But unlike her, I will actually also give you
02:19:04 three interpretations.
02:19:07 And she said, the answer to the meaning of life
02:19:09 is become one.
02:19:12 And the first interpretation is like a child,
02:19:16 become one year old with the excitement
02:19:18 of discovering everything about the world.
02:19:21 Second interpretation, in whatever you take on,
02:19:25 become one, the first, the best, excel,
02:19:28 drive yourself to perfection for every one of your tasks
02:19:32 and become one when people are separate,
02:19:38 become one, come together, learn to understand each other.
02:19:43 Damn, that’s an answer.
02:19:45 And one way to summarize
02:19:46 this whole meaning of life symposium
02:19:48 is that the very symposium was illustrating
02:19:52 the quest for meaning,
02:19:54 which might itself be the meaning of life.
02:19:58 This constant quest for something sublime,
02:20:01 something human, something intangible,
02:20:04 some aspect of what defines us as a species
02:20:09 and as an individual.
02:20:11 Both the quest of me as a person through my own life,
02:20:16 but the meaning of life could also be
02:20:19 the meaning of all of life.
02:20:20 What is the whole point of life?
02:20:22 Why life?
02:20:22 Why life itself?
02:20:24 Because we’ve been talking about the history
02:20:26 and evolution of life,
02:20:28 but we haven’t talked about why life in the first place?
02:20:31 Is life inevitable?
02:20:32 Is life part of physics?
02:20:35 Does life transcend physics
02:20:37 by fighting against entropy,
02:20:40 by compartmentalizing and increasing concentrations
02:20:42 rather than diluting away?
02:20:45 Is life a distinct entity in the universe
02:20:51 beyond the traditional very simple physical rules
02:20:55 that govern gravity and electromagnetism
02:20:58 and all of these forces?
02:21:00 Is life another force?
02:21:02 Is there a life force?
02:21:03 Is there a unique kind of set of principles that emerge,
02:21:05 of course, built on top of the hardware of physics,
02:21:09 but is it sort of a new layer of software
02:21:11 or a new layer of a computer system?
02:21:14 And so that’s at the level of big questions.
02:21:18 There’s another aspect of gratitude
02:21:21 of basically what I like to say is,
02:21:27 during this pandemic,
02:21:27 I’ve basically worked from 6 a.m. until 7 p.m.
02:21:30 every single day, nonstop, including Saturday and Sunday.
02:21:34 I’ve basically broken all boundaries
02:21:36 of where life, personal life begins
02:21:39 and work life ends.
02:21:42 And that has been exhilarating for me,
02:21:46 just the intellectual pleasure that I get
02:21:50 from a day of exhaustion,
02:21:53 where at the end of the day, my brain is hurting.
02:21:55 I’m telling my wife, wow, I was useful today.
02:22:00 And there’s a certain pleasure
02:22:04 that comes from feeling useful.
02:22:08 And there’s a certain pleasure
02:22:09 that comes from feeling grateful.
02:22:12 So I’ve written this little sort of prayer for my kids
02:22:16 to say at bedtime every night,
02:22:19 where they basically say,
02:22:21 thank you, God, for all you have given me
02:22:24 and give me the strength to give onto others
02:22:28 with the same love that you have given onto me.
02:22:33 We as a species are so special,
02:22:36 the only ones who worry about the meaning of life.
02:22:40 And maybe that’s what makes us human.
02:22:44 And what I like to say to my wife and to my students
02:22:47 during this pandemic work extravaganza
02:22:53 is every now and then they ask me, but how do you do this?
02:22:56 And I’m like, I’m a workaholic.
02:22:58 I love this.
02:23:00 This is me in the most unfiltered way.
02:23:04 The ability to do something useful,
02:23:07 to feel that my brain is being used,
02:23:09 to interact with the smartest people on the planet
02:23:12 day in, day out, and to help them discover aspects
02:23:15 of the human genome, of the human brain,
02:23:18 of human disease and the human condition
02:23:21 that no one has seen before
02:23:24 with data that we’re capturing that has never been observed.
02:23:29 And there’s another aspect, which is on the personal life.
02:23:34 Many people say, oh, I’m not gonna have kids, why bother?
02:23:37 I can tell you as a father,
02:23:41 they’re missing half the picture, if not the whole picture.
02:23:44 Teaching my kids about my view of the world
02:23:49 and watching through their eyes
02:23:51 the naivete with which they start
02:23:53 and the sophistication with which they end up,
02:23:56 the understanding that they have
02:24:00 of not just the natural world around them, but of me too.
02:24:05 The unfiltered criticism that you get from your own children
02:24:10 that knows no bounds of honesty.
02:24:15 And I’ve grown components of my heart
02:24:18 that I didn’t know I had
02:24:20 until you sense that fragility,
02:24:25 that vulnerability of the children,
02:24:30 that immense love and passion,
02:24:34 the unfiltered egoism,
02:24:36 that we as adults learn how to hide so much better.
02:24:40 It’s just this back of emotions
02:24:43 that tell me about the raw materials that make a human being
02:24:48 and how these raw materials can be arranged
02:24:50 with more sophistication that we learn through life
02:24:53 to become truly human adults.
02:24:57 But there’s something so beautiful
02:24:59 about seeing that progression between them
02:25:02 and seeing that progress and that progress
02:25:05 and that progression between them,
02:25:07 the complexity of the language growing
02:25:10 as more neural connections are formed
02:25:13 to realize that the hardware is getting rearranged
02:25:18 as their software is getting implemented on that hardware,
02:25:22 that their frontal cortex continues to grow
02:25:24 for another 10 years.
02:25:27 There’s neuronal connections that are continuing to form,
02:25:29 new neurons that actually get replicated and formed.
02:25:33 And it’s just incredible that we have these,
02:25:38 not just you grow the hardware for 30 years
02:25:40 and then you feed it all of the knowledge.
02:25:42 No, no, the knowledge is fed throughout
02:25:45 and is shaping these neural connections as they’re forming.
02:25:48 So seeing that transformation from either your own blood
02:25:52 or from an adopted child
02:25:54 is the most beautiful thing you can do as a human being.
02:25:57 And it completes you, it completes that path, that journey.
02:26:00 The create life, oh sure, that’s at conception, that’s easy.
02:26:04 But create human life to add the human part,
02:26:08 that takes decades of compassion, of sharing,
02:26:13 of love and of anger and of impatience and patience.
02:26:18 And as a parent,
02:26:21 I think I’ve become a very different kind of teacher
02:26:25 because again, I’m a professor.
02:26:27 My first role is to bring adult human beings
02:26:31 into a more mature level of adulthood
02:26:34 where they learn not just to do science,
02:26:37 but they learn the process of discovery
02:26:39 and the process of collaboration, the process of sharing,
02:26:42 the process of conveying the knowledge
02:26:44 of encapsulating something incredibly complex
02:26:48 and sort of giving it up in sort of bite sized chunks
02:26:51 that the rest of humanity can appreciate.
02:26:54 I tell my students all the time, if you, you know,
02:26:57 like when an apple fall,
02:26:58 when a tree falls in the forest
02:27:00 and no one’s there to listen, has it really fallen?
02:27:03 The same way you do this awesome research,
02:27:05 if you write an impenetrable paper that no one will understand,
02:27:08 it’s as if you never did the awesome research.
02:27:11 So conveying of knowledge, conveying this lateral transfer
02:27:15 that I was talking about at the very beginning
02:27:17 of sort of humanity and sort of the sharing of information,
02:27:22 all of that has gotten so much more rich
02:27:27 by seeing human beings grow in my own home
02:27:32 because that makes me a better parent
02:27:35 and that makes me a better teacher and a better mentor
02:27:38 to the nurturing of my adult children,
02:27:42 which are my research group.
02:27:43 First of all, beautifully put, connects beautifully
02:27:48 to the vertical and the horizontal inheritance of ideas
02:27:52 that we talked about at the very beginning.
02:27:54 I don’t think there’s a better way to end it
02:27:57 on this poetic and powerful note.
02:28:01 Manolis, thank you so much for talking to me.
02:28:02 It was a huge honor.
02:28:03 We’ll have to talk again about the origin of life,
02:28:07 about epigenetics, epigenomics,
02:28:10 and some of the incredible research you’re doing.
02:28:13 Truly an honor. Thanks so much for talking to me.
02:28:15 Thank you. Such a pleasure. It’s such a pleasure.
02:28:17 I mean, your questions are outstanding.
02:28:19 I’ve had such a blast here and I can’t wait to be back.
02:28:21 Awesome.
02:28:23 Thanks for listening to this conversation
02:28:24 with Manolis Kellis, and thank you to our sponsors,
02:28:28 Blinkist, 8sleep, and Masterclass.
02:28:31 Please consider supporting this podcast
02:28:33 by going to blinkist.com slash lex,
02:28:35 8sleep.com slash lex, and masterclass.com slash lex.
02:28:41 Click the links, buy the stuff, get the discount.
02:28:44 It’s the best way to support this podcast.
02:28:47 If you enjoy this thing, subscribe on YouTube,
02:28:48 review it with five stars on Apple Podcast,
02:28:50 support it on Patreon,
02:28:52 or connect with me on Twitter at lexfreedman.
02:28:55 And now let me leave you with some words
02:28:57 from Charles Darwin that I think Manolis
02:29:00 represents quite beautifully.
02:29:02 If I had my life to live over again,
02:29:04 I would have made a rule to read some poetry
02:29:07 and listen to some music at least once every week.
02:29:11 Thank you for listening, and hope to see you next time.