00:00hey I'm like solana thanks for checking
00:03out our live stream at Founders Fund on
00:05synthetic biology and longevity which
00:08we're gonna be talking about today let's
00:10quickly introduce my panel of genius
00:13friends and colleagues here
00:16hi my name's Dan some of skya working in
00:19engineering momentum cells that ecology
00:25I'm Laura diamond I run a venture
00:28capital fund focused on investing in
00:29companies that aim to extend human life
00:32and I'm Erin vandevander two scientists
00:35here at the Pechanga all right before
00:38really sorry I want to kind of just
00:40quickly frame the conversation I think a
00:42lot of people don't believe that
00:44extending the human lifespan longevity
00:47is a serious question talk about either
00:50kind understand what you're saying
00:51that's very much what we're gonna be
00:53talking about over the next 45 minutes
00:55it's just that this is a problem that
00:57can be solved and we're going to kind of
01:00come at it from a bunch of directions I
01:02want to start with just maybe talking
01:04about what the vehicles in synthetic
01:06biology are kind of give everybody a
01:08kind of that question and someone what
01:11is the holy grail here what are we
01:13actually working towards why is this
01:15such an exciting field
01:16well in biology it's really this is the
01:19biggest reverse engineering project in
01:21the history of mankind were trying to
01:23reverse engineer this four billion year
01:25old gaming technology and we've come a
01:27long way last century and the synthetic
01:31biology the real goal is to be able to
01:33make predictive changes to actually give
01:35an engineer cells that build and make up
01:38a body and to do so in a way that allows
01:41us to treat disease and perhaps one day
01:44soon even natural capabilities
01:47but along the way what we're figuring
01:50out how to do is how it actually
01:52specifically and the state play engineer
01:55cells to achieve some medical goal right
02:01and to use them as the new form of
02:06my view this is or this is pretty much
02:09how you think about it as well yeah I
02:10mean I think synthetic biology like one
02:12of the coolest things about it is I'm
02:14like it back at MIT original lab for a
02:15bit and and we literally were like you
02:17know three kids for summer like what if
02:19we just trained cells to like respond to
02:21pressure like create both of them like
02:23could we make like an armored thing
02:24that's what we could wear and like you
02:25have to get thrown in a car and I'd like
02:26to get a boat for the suit and like
02:28obviously we can even fight that far
02:29along but it's just amazing how quickly
02:31as like a kid you can just take stuff
02:33and in a summer like you make something
02:36that's actually you've been seemingly
02:37cool like in biology and I think that's
02:38really resetting and not quite in the
02:41public sphere as much as it should be
02:42like right now it's usually people
02:43throughout time I thought of biology is
02:45this fixed thing that maybe you could
02:47prod and poke and just slightly but
02:49we're now talking about Rican sieving
02:52what it means to be yeah we really
02:55really have like parts like we have a
02:56parts library and it's like put the
02:58parts together to like make actuators
03:00and stuff and it's like really cool
03:01there's existing technology what do you
03:04see like what is it what is a game what
03:05is the ultimate always perfect
03:07biological control look like Lego blocks
03:09like you know what you want and it's
03:12it's very straightforward and I mean
03:13like you know then it's very complicated
03:14of course because these systems are very
03:16complex but I think the thing that I
03:18love also the quality is just the
03:19engineer ability of it like that's the
03:25synthetic biology is great in that it is
03:28a a better understanding of what it
03:32really means to be a living organism and
03:34what it means to have a disease or old
03:37and I you know we've had medicine for a
03:39long time and the medicines that we give
03:42are just simple molecules and those are
03:46static there and if your if your
03:48understanding of a disease this is just
03:50a static condition like you have a
03:52bacteria right that's a
03:55sick state and I give you some
03:57antibiotic and then I can then I can
04:00change that but if your understanding of
04:03disease is more sophisticated than just
04:06like presidents or absence of a pathogen
04:08and then it's really about how all of
04:11the systems in your body is working
04:13together then being able to treat that
04:16with just a molecule that's a static
04:18fixed thing is really too limited so
04:21synthetic biology allows us to create
04:24living dynamical systems ie a cell which
04:29is a self-contained living dynamical
04:31system that we can program in soon and
04:33just in the same way that a synthetic
04:36chemists can make molecules a synthetic
04:38biologists can make cells or biological
04:42systems that can treat the disease which
04:45is a part of your living condition in
04:48motion it's no longer just a static
04:50model for what it means to be well or
04:53sick it's now a dynamic model of your
04:56whole metabolism functioning together
04:57and using these dynamic tools like cells
05:01we can synthesize ways of addressing
05:03those in motion what would you say are
05:05some of the huge problems that we're
05:06looking to to tackle with this kind of
05:09this kind of stuff probably the best
05:12most straightforward one would be like
05:13the FIB teas right which is a metabolic
05:18kind of issue where you have your blood
05:21sugar and your insulin are always
05:23chasing after each other and so we've
05:26been trying to treat that for years you
05:29know since for a hundred years by
05:32the static application of insulin but it
05:37never really works out it's very
05:39difficult and it's a very burdensome
05:40disease to manage where you're testing
05:43yourself multiple times a day injecting
05:45yourself multiple times a day because
05:47fundamentally you're trying to alter a
05:49dynamic disease State you're changing
05:52insulin level with a static test and
05:58and some of the sounds somewhat similar
06:00what you're working on that you guys are
06:01targeting diseases and reprogramming the
06:04human immune system to go after them can
06:06we talk a little bit about that and just
06:08kind of the interim steps it's kind of
06:10perfect biological control yeah well
06:13right now people are really interested
06:14in the immune system mainly because it's
06:16a very easy system where you can take
06:18the cells out of the body
06:20keep them passive to them make genetic
06:22modifications to them check them out on
06:23put them back in the in system was
06:27service a distributed system that's a
06:30very good place to initiative began
06:31trying to do real genetic there is
06:36I mean diabetes is where the interesting
06:38is very simple disease in some sense
06:40that it's a comes type one diabetes
06:42comes to nature let me to make sure I
06:44understand this correctly so you have a
06:46disease let's say I mean give me an
06:48example like right now we're focusing
06:50cancers so you have a certain kind of
06:52cancer then someone from your team or or
06:55someone would come in and actually
06:57change the way that your body reacts to
06:59its required bodies and kind of go after
07:02the cancer what you're trying to do is
07:03really sort of just teach the and system
07:05that cancer really is or an entity that
07:07needs to be destroyed so what people are
07:10doing right now they're trying to teach
07:11t-cells they're trying to engineer the
07:13t-cells to recognize the foreign
07:16presidents the prophets you know
07:19rebellions presidents inside you to
07:21detect the cancer cells in to just
07:24annihilate them I mean this happens
07:26rarely naturally but it's typically not
07:31and do you now have a system to clean
07:34your body with cancer but what people
07:35are trying hours to sort of accelerating
07:37the process of just teaching Arian
07:39systems recognize these bone cells into
07:41white porcelain they've purchased this I
07:43mean how do you get your immune system
07:44to do what you want to do well what you
07:46do you sort of incredibly complicated in
07:49beautiful national system for
07:50discovering foreign antigens that
07:53foreign shapes on themselves they're not
07:55around with bacterial cells or viruses
07:57but you really been evolved to get very
08:00good at detecting viruses and bacteria
08:03it's not so good at noticing when your
08:05inner-self on relative exactly the sort
08:07of avoids attacking them because that's
08:08the basis broad immunity so what people
08:12are doing right now they're basically
08:13short-circuiting the whole natural
08:15process of discovering what's for they
08:17literally just the designer that
08:18receptor to the design receptively a
08:21protein that is designed to bind to
08:23literally stick it to the cancer cell so
08:26what you're sort of just you're sort of
08:27very quickly is teaching the t-cell and
08:29this is what you want to go after if you
08:31want to go after this guy can you see it
08:33you kill it and so we're sort of
08:35short-circuiting a natural
08:37the natural process of discovery and
08:41directly by hand I'm wondering Laura I
08:45mean you're a biologist and you're also
08:47now a partner at and the founder deputy
08:50fund so you're investing in this kind of
08:51stuff how are companies dealing with
08:55these problems thinking about these
08:56problems and think about investing in
08:58companies like that so something like
08:59one really cool thing about this fuel
09:01right a synthetic biology does not like
09:02a recent phenomenon we think it is you
09:04know we're talking about it like it
09:05happened the past five years but like
09:06the evil at the 1980s right like the the
09:08founding of biotechnology as a whole
09:10like that was recombinant DNA like that
09:12was copy pasting some DNA into some cell
09:14of having that cell Korea the proof that
09:15you wanted to put into a person and
09:17we're talking like how it's like not a
09:19great fear for debuts and it's not you
09:20know but it's like it's incredible to me
09:21that like you know for 20-something
09:23years we've been taking human cells and
09:26getting them to make this protein just
09:27extremely complex purifying approach now
09:29that sticking into your human and that's
09:31really kind of synthetic biology right
09:33and so announced I was talking about you
09:35know these these cells that are very
09:36good at binding to kind of these on
09:38foreign entities that's antibiotic
09:40technology and that's the founding of
09:41biotechnology you know the 1980s again
09:43like that Genentech saying you know what
09:45if we just took some piece of a cancer
09:47and just put it into a mouse and had
09:49that now it's like figure out what the
09:50heck was going on try to attack that
09:52cancer and kill it and then get in and
09:54sort of like develop these really good
09:56binders for the cancer kind of molecule
09:59and so you know we even talked about to
10:00the co biology yes it's like the field
10:03complex and that ecology take multiple
10:05parts of putting them all together
10:05that's very recent but kind of like the
10:08field of nipple aiding cells in this way
10:10is actually kind of I think it's been
10:12the birthright of biotechnology since it
10:14first started about there years ago I
10:15would say there has been another it's
10:17more than just the complexity there's a
10:19shift in that the making recombinant
10:22insulin or making antibodies from cho
10:24cells is definitely synthetic biology
10:27but it's sort of like ancillary
10:29synthetic biology because it only
10:31happens in a dish and then we maybe take
10:34the results of that and we put it into
10:36your body but the way that when it
10:38actually like the rubber hits the road
10:40when the needle hits your vein that is
10:42still static molecules the region we're
10:45getting into now is where we're taking
10:47those reprogrammed cells and applying
10:50them into your body directly so it's
10:52like primary synthetic biology medicine
10:54rather than secondary what would you say
10:57we've had around here cell phone Valley
10:59we've had this tremendous wave of
11:02incredible companies emerging
11:04Computer Sciences software companies and
11:08hardware companies how does biology fit
11:11into that what are the differences maybe
11:14between starting up companies I mean
11:15both are are we poised to see another
11:17wave of maybe now biology companies it
11:20adds it becomes a lot cheaper to do this
11:21stuff and what is it what is that kind
11:23of stuff of like I would say we are we
11:26are we are policed for that and there
11:28are a lot of reasons for that but the
11:30one I want to highlight right now is the
11:33the predictability of biology has come a
11:37long way in the last few years and that
11:40one of the reasons why Silicon Valley
11:45create a lot of these technology
11:47companies very quickly was that you
11:49could do things cheap and you could do
11:51things reliably because when you run a
11:52program on a CPU on a computer processor
11:57it runs the same way every time and the
12:00tools that we had for doing biology were
12:04very noisy and so that prevented where
12:07as you know when you're saying studying
12:10physics as the science and using that as
12:12a foundation for engineering for
12:14computer architecture as the foundation
12:16for software engineering
12:19biology wasn't at the point where it was
12:22reliable enough keep enough to become a
12:24really good foundation for biological
12:27engineering and that has really started
12:29to change and that we've seen a lot of
12:33flat scale automation kinds of things
12:35happening and also just our basic
12:38fundamental understanding of things like
12:41genetics from cheap sequencing and the
12:44tools that we use to understand the
12:46biology have sort of gotten us to a
12:48place where it now makes sense to really
12:50think about biology as a mature enough
12:53scientist to use as ahead of this virgin
12:55ear I kind of one thing I think we have
12:59to talk about in this conversation is
13:01just CRISPR cats 9 and ketamine in
13:06can you just give us a quick and since
13:08you might meet me as there's like a
13:09quick for our viewers that what is
13:11CRISPR why is it important has fitted to
13:13the sort of other gene editing
13:16it's one of the biggest challenges say
13:20you know there's a mutation in a
13:22person's cell that is causing a disease
13:24cystic fibrosis is cocktail mutation
13:26you'd like to go in there just fix it
13:27you'd like to go in there to change it
13:29or say you want to put some new stuff in
13:31but you don't want to put it in the
13:33middle of the gene and destroy that gene
13:34you want to put it in a very specific
13:36place that that's been credibly hard
13:38thing to do just the ability of
13:40targeting them edit so being able to go
13:42into one small place in the 3 billion
13:45letters that makes up your code and make
13:48a change right there well people have
13:51done for most of the last week decades
13:54has been throwing something kind of
13:56randomly there's a few exceptions in
13:59mice or whatever reason actually there's
14:01a trick of let's see sort of it edit
14:03them somewhat decently and that's why
14:05minds become such a huge research model
14:07because this is somewhat easier in there
14:09but in general is incredibly hard to do
14:11this in in higher animals in the million
14:14room systems and so it CRISPR lets you
14:17do is to basically very specifically
14:19target a little protein that just cuts
14:22the DNA in a particular place and that
14:23Cootes all of the repair machinery at
14:25your sales muse that will then integrate
14:27in a little piece of DNA and also throw
14:30just you know a basic repair in within
14:33your edit and create that alteration and
14:36so being able to target you know being
14:38on being able to actually say I want to
14:39change that right there it's a huge deal
14:42and it really is Nathan a lot of your
14:43research so if right now today we are
14:45able to just target a gene and cut it
14:50what problems are loved to solve right
14:52that seems pretty great right like you
14:53can just bill just control but you know
14:56theoretically what is the what is what
14:58is preventing us from doing that
15:00there's two things that even as good as
15:03Christopher is it still needs to improve
15:04a little bit it still occasionally
15:06places you don't want it to you know
15:08there still it's so new there's still a
15:10lot of work happening but that progress
15:12we made very rapidly well we'll get it
15:13to be much more specific we'll use some
15:15other tricks and we'll make it really
15:17clean so that you can feel quite
15:18confident your only target is the change
15:20in Y to but but aside from the math are
15:22different challenges just knowing what
15:24to do I mean some problems very easy
15:26cystic fibrosis is very simple you know
15:28what the problem is you know it's a
15:30single appointment you know that you
15:31just want to revert that the natural
15:34it's challenge there is that you're
15:36typically dealing with that mutation in
15:39a bunch of opportunity of cells in your
15:41mucosal lining and you want to find a
15:43way to get this CRISPR and the altered
15:46gene into those cells so there's still a
15:47huge problem with delivery so
15:49traditionally easy viruses to do that
15:51and they work pretty well but they're
15:53not quite good enough yet and its really
15:56sort of packaging CRISPR along with all
15:58those viral viral delivery technologies
16:00to make these edits and that's just
16:03talking about making that just just
16:06fixing known errors and you want to talk
16:07about doing stuff if it's more synthetic
16:10or you want to be pretty new stuff in
16:11the challenge there is really figuring
16:14out what are you going to put in like
16:15what is the spark that you would put in
16:17there's still huge amount of time it's
16:19still a huge amount of work to be done
16:21to figure out you know what am i
16:22changing what what actual
16:24new stuff to address particular
16:27properties they knew three billion every
16:30million letters that's the letter that
16:31each one of us has a sort of unique 3
16:33billion letter code this is two
16:35megabytes of difference between us yeah
16:38we dealt what is worth roughly the
16:40fraction those letter to be after even
16:43know what they do how many of those
16:45things do we understand so if you're
16:48talking about the protein coding genes
16:49you know which is a small fraction we we
16:53know what a lot of those proteins do it
16:54is still a lot we don't know the real
16:56sort of dark matter and and biology is
16:59understanding the couple transcription
17:01with transcriptional regulatory logic so
17:03this is the gene is the what but the way
17:06are in the wind like when I made this
17:07gene when we make this protein that will
17:10we're about time of my life I'm a child
17:12or adult as a developing embryo there's
17:14an incredibly complicated state machine
17:16they're very complicated
17:18dynamical system that is encoded in
17:21these promoter regions these non
17:23transcribed regions they're their
17:24regions we give you DNA don't encode
17:26proteins but they encode a kind of
17:28logical kind of developmental code and
17:30so a lot of what your the your body plan
17:33is encoded and this no regulatory logic
17:36is just promoter stuff
17:38and we are getting better it's sort of
17:42reverse of it beginning to read with
17:44some of that stuff does but I'd say our
17:45understanding of that the
17:47transcriptional logics still fairly
17:48primitive you can't go in and just look
17:50at the code and say oh that's gonna
17:51cause me to have a second phone you know
17:53it's gonna it's gonna change the Hox
17:55gene expression pattern has to cause my
17:57legs to look a little differently and so
18:00much of the code that dictates how
18:02complicated bodies work he's really
18:05stuck inside that non-coding on you
18:07what does he take first students to
18:09understand to move back to no I mean
18:13what are the steps to understanding the
18:16yep I mean where there's a lot of work
18:20going on in the stretch right now and
18:22particularly when you're talking about
18:23this transcriptional logic in the
18:25context of state development when you
18:27just just he's in Crete and when you're
18:29talking about how two bodies develop
18:30from a single cell into a complicated
18:32machine a lot of the work that that's
18:35happening right now is it's benefiting
18:37from the fact that we have amazing
18:38microscopes nowadays prepared what we
18:40did in two years ago and you can
18:42actually look in like a zebrafish or a
18:44small developing mouse or you can
18:45actually watch all the cells in the body
18:48up to a certain point every single one
18:50of them we can begin watching how they
18:51have to divide grow and you gossiped use
18:54all these impressive fluorescent
18:56proteins that you can so you can in this
18:59research models you you can splice these
19:01proteins in you can watch and help
19:03particular parts this comes Kennedy just
19:05crucial that wouldn't come and come on
19:06and off like fun around her time so you
19:08can actually begin to just watching it
19:10or begin taking measurements of this
19:12dynamical system in action and that and
19:15it's sort of hide stupid wait will
19:17really allow us to begin getting the
19:20datasets we need to sort of reverse
19:21engineer what's going on with that
19:22Bodmin we talked earlier
19:25all of us talked a little bit earlier
19:29needing data just like we access to
19:33large datasets we need to to be able to
19:36see what everyone's genome is we can
19:38compare it like learn what these things
19:41in this country right now maybe keep on
19:44walking us through yeah surah says I
19:46love to I mean if you're trying to
19:48figure out what you know it is we need
19:49data to do that can you explain to me
19:51why and then what that process looks
19:53like the county accent yeah so if you
19:56think about like the G like genomics in
19:58general it's kind of like the most
19:59perfect machine learning problem you
20:00could come up with you know it's like
20:01literally a what hot encoded vector like
20:03given to you like you have four
20:04possibilities new position and like
20:06that's just crazy right it looks like
20:07Nature gave you a machine learning
20:08problems like here do this but but the
20:11problem is like you know because we only
20:13have you know three billion base pairs
20:14information per person in watts and to
20:17really get sekolah kind of them power
20:19out of that you need on the order of
20:20maybe tens of thousands at minimum
20:22hundreds of thousands kind of and
20:24maximum genomes to work with and we
20:27really you know today in the world I
20:28would say they're probably about a
20:29million fully sequenced that exists on
20:32the spread across many different
20:33countries and agencies in the UK they're
20:35about 70,000 that are kind of publicly
20:37available to researchers there or not
20:38it's hard to get that doing here
20:40the US has about 400,000 veterans that
20:42they've taken blood samples from but
20:44only about 2,000 of those have whole
20:46genome sequenced and the the UK also the
20:48u.s. also has approximately 10,000 to
20:5040,000 kind of whole genomes that are
20:52kind of locked up in this database
20:53called DB gap where if you apply for it
20:55and your professor you can kind of get
20:56access to it but I think a really big
20:59problem is you know like this is really
21:00a machine learning problem you know and
21:02you can actually throw kind of basic
21:03image kind of recognition nets at do you
21:06know and immediately get like the
21:08state-of-the-art results in biology
21:09which is pretty crazy and you can do
21:11this with any kind of net that you can
21:12think of for current etc for most kinds
21:14of problems you could be given biology
21:15which is which is a really amazing fact
21:17but but all the researchers that you
21:18talk to you in this field of machine
21:20learning who are passionate about
21:21biology maybe have kids with diseases
21:23they're like but the data is just not
21:25that like I know it's there but like to
21:27access it I would have to write forums
21:28and talk to administrators and I don't
21:29want to do any of that like you know
21:31we've need to do my models and you go
21:33get the data nobody wants to do that so
21:35it's pretty crazy because you know in a
21:36place is like trying everything for
21:37example you're seeing this amazing
21:38openness to just give data to everybody
21:41like I said to the US Senate to whoever
21:42has the technology to analyze it and in
21:45the u.s. it's kind of like no we have to
21:46be very very cautious and careful and
21:48that's great but it also is kind of I
21:50think putting us at a disadvantage to a
21:51lot of other places why hasn't it been
21:53working why miss country people so
21:56reveal their genetic code morning would
21:58you put your genome online today like so
22:00I would totally the insurance thing
22:03which I had to get bought off I mean
22:05technically technically and I believe
22:07it's true there's actually a law was
22:09passed 15 years ago from its insurers
22:11from using genetic data for making
22:13but it is generally a worry you know you
22:18know here is that like your insurance
22:20company sees that you have genetic
22:21predisposition to I don't know some
22:23crazy disease for example and they just
22:26don't want to give you insurance do
22:27another job right now there's a law
22:29against that but it's allowing it
22:30appears maybe that's perhaps they'll
22:33find a way around it that's not gonna be
22:34enforced you know how can you prove that
22:36the using that information it's there's
22:40genuine concern in America because of
22:41the way that we insure ourselves and get
22:44health care and pay for health care that
22:46makes people very leery about just
22:49putting up their genome online and the
22:52gym is really only a part of it I mean
22:54it's important part of it but you just
22:55have a string literally just a code it's
22:58it's only modern use but what you really
23:00want is to have both of the genomic
23:02application as well as a lot of the
23:04basic a lot of basic lab tests and
23:07hopefully longitudinally over time like
23:08the Nellis innovation history like the
23:10how the whole family's
23:12genomes and basic thinit types that we
23:14say in biology you have the genotype is
23:16just literally just the strand but what
23:18buddies encode and then these peptides
23:21the color you're scanning the color you
23:23care your eyes no do you have these
23:25diseases you know what they puts the sum
23:29cleanse your body part and what you
23:32really want you're going to apply
23:33machine learning models is to have not
23:36just the general but also a lot of facts
23:38about the patient as many as we can you
23:40know it's pretty levels other
23:44everything that a doctor might have
23:46ideally have some sort of anonymous
23:49phone with that to be able to begin
23:50doing their supervisor anyone it's more
23:52you were just saying that I mean if you
23:53had all that information you can run it
23:55through a computer now and see some
23:56pretty interesting results it's crazy
23:58what it can do with like the crack
23:59misinformation possible like like today
24:01people are working with on board of like
24:02thousands publicly available genomes I'm
24:04like their interpretive like basically
24:06everything that you can get from these
24:08folk will be available too so it's like
24:09with for a high precision with nuts and
24:10so it's not hard to think of like all
24:12these this you could start to approach
24:14better so we need with that yes we need
24:16more we need more data what are some
24:18workarounds means to the data access we
24:21have lots of data read access to the
24:24well there's two there's two barriers to
24:26write and one reflects the other so
24:28privacy is a very strong cultural value
24:31more so in the West than places like
24:34China and so we need policy systems that
24:38reflect that cultural value just to be
24:41able to you know pass snuff or existing
24:44in the wild one of the ways in which
24:46that value is coded right now is that in
24:49order to mess around with people's data
24:52you need to you need to be able to argue
24:56that there's some tangible benefits to
24:58that patient right so when things are
25:01purely exploratory there's a lot of
25:03resistance to that because it's hard to
25:04articulate I'm putting some people's
25:08privacy at risk they're taking some
25:10exposure I can't articulate a specific
25:12benefit until I actually you know run my
25:15nets and figure out what the upside is
25:17right so we need a way of encoding that
25:21benefit to people that in a more
25:24abstract way so that we can create a
25:27sort of you know policy framework that
25:30says there is a huge advantage to
25:33everyone including the people that are
25:35that are part of those databases and
25:37that that allows the researchers to go
25:42in and start looking at things in a way
25:44that justifies the privacy exposure once
25:46you do that then it makes it possible to
25:49develop the technical systems to
25:52actually make the communication possible
25:55right like now all these different
25:58countries and agencies they've been
26:01growing organically and so they put all
26:03of this database all this data in just a
26:07homespun systems that are protected
26:10because there's no there's no policy
26:13impetus to do that large scale scale
26:15kinds of share why don't you have that
26:18then then that creates an environment
26:21where it makes sense to actually do the
26:22technical work to put it all together
26:24why is there no push for this problem I
26:29think there's like no cultural push
26:30there's no government push why people
26:31not pick me about this in this sort of
26:33serious way I think people like thinking
26:35about it in a serious way for sure and
26:38it's but it's that's mostly local level
26:41decision-making right especially in this
26:43country where you know the people who
26:47have to build coalitions in order to do
26:50that sorting in the UK they have a
26:53National Health Service and so there is
26:55a there is a like top-down organization
27:00that has a that can that can
27:07information and make sure that it's used
27:10appropriately in the u.s. though there's
27:13always the concern that okay if we get
27:16if we choose anyone to sort of be the
27:19one true place that everyone is stored
27:21it that person or that organization
27:23might derive some preferential benefit
27:26and then the people who get gave them
27:29their data lose that so figuring out a
27:31way to solve that coordination problem
27:33so that everyone's our interests are
27:36aligned is very difficult and a very
27:37political people are absolutely working
27:40on it but it's not something that
27:41happens quickly especially not in this
27:43country it seems about you're saying
27:45Lauren you were afraid that we were
27:47falling behind for example in China
27:48because really trying to hope we will
27:51collaborate with us to do things right
27:52about it's just weird to see like that
27:54you have to Kate like like everyone
27:56wants to do this thing but then like
27:57everybody just can't quite agree in
27:59Savelle like we don't do it for like
28:00five years and then they're like quite
28:01behind a lot of different measures I
28:03think another thing is that you know
28:05like when we sequence the genome like
28:06the early like 2001 or so I think a lot
28:08of folks like you know their covers on
28:10nature and science and people saw that
28:12the genome had been sequenced right and
28:14so obviously genomics was a real thing
28:15we'd have like many diseases cure and we
28:18didn't immediately after that fact and
28:20so I think a lot of folks you know if
28:21you took something on the street we said
28:22we have this new CRISPR therapeutic and
28:24it's greatly be like yeah I read about
28:26that recently like maybe but he said
28:28genomics they'd be like Holden we do you
28:30know like why why did now why me etc and
28:33so it's hard to get people I mean it's
28:34amazing to like we've got four hundred
28:35thousand veterans or something like yeah
28:36sure you know take my information like
28:38hey I've done a lot of my country I'm
28:40gonna do some more and then other folks
28:42I think it's a different the different
28:44approach to it it's interesting that you
28:49so the practicum Aquino Merrill activist
28:53991 sent home to the year
28:57lost any money it is cracked into
28:59thousands and there was this wave of
29:01press that said you know we would have
29:03curious it's from every since there are
29:05like covers with people's faces on them
29:06are going to understand aging what about
29:08what makes Wyatt pack maybe just like
29:10what is it what is can you believe a
29:13spares like what do you do it like just
29:15like given like a TCG for like three
29:16Billy position okay there's nothing you
29:18can do with that I don't like you could
29:19look for patterns but like like even
29:21without machine learning at that point
29:23I've been really hard to just
29:24unstructured like it unstructured
29:26fashion find patterns and so now we can
29:29kind of do that but I mean to do great
29:31yeah work I mean if I gave you the the
29:35binary compiled version of - saying
29:37Laura - I just gave that to you and said
29:41well I would give it to Aaron
29:45take you two years to really understand
29:49all the interesting structure in order
29:53and unfortunately with the kind of
29:55genome we don't have the source code
29:57wasn't included unfortunately so and
30:01it's even harder because the system
30:03that's running it is not as
30:05deterministic it really is you know the
30:08way that 500 cells work which is work
30:10and there's a bit of a statistical
30:11nature to it and very subtle one
30:14and ultimately a lot of binding
30:16interactions between proteins are on
30:18mechanical in nature and they're very
30:19very hard to forget to straight route
30:22for stimulation so it's really very hard
30:25to to just pull out and implicitly the
30:28interest at logic that it's occurring
30:30the thought internet and systems inside
30:32cells even if you isolate yourself to
30:34just talking about a small ok the
30:36systemic you proteins and I can gather
30:37people quite hard to predict things from
30:39you just the code and so another
30:42challenge you know is that the code is
30:43not not it's necessary you know we're in
30:48this critical thing to do but you've got
30:50it to do quite a lot of experiments on
30:52top of having it toads really begin
30:55figuring out how that stuff works I
30:57would say there's actually there's e
30:58there's even a deeper problem to this
31:00which is that the metaphor of thinking
31:03about it as code learning on a computer
31:04is slightly wrong because if I look at
31:08the Linux source code I can be said to
31:11have understood it if I can tell you
31:13what each piece is for I need what the
31:17intent of the programmer who wrote it
31:18was when they wrote that that piece but
31:21there's no writer from our genetic code
31:24there is no intent it's an evolved
31:27system and so we can impose our own
31:30interpretations of what things are for
31:32but that doesn't actually map to the
31:35underlying intent from some from from
31:38from an original program and so thinking
31:42about it that it has to have that level
31:44of abstraction and it just doesn't work
31:46there's parts in there that are you know
31:49they're essentially chaotic from our
31:52level of interpretation right and so we
31:55need a way of making predictions about
31:58what will happen in without getting
32:01drawn into this this is how the code is
32:06understanding human consciousness level
32:09version of what is that to happen oh
32:11well we got a couple minutes left we've
32:15kind of outlined the big huge goals
32:17controlling human biology and being
32:19aging hopefully one day
32:21in the short term and needed Club
32:26be this problem of data we have a lot of
32:30it we don't know how to necessarily
32:31figure out what everything does we don't
32:33have access to two different two
32:35different pieces there's this difficulty
32:37in understanding how to read it if it's
32:41you know again we're about applying our
32:43own language to this system that really
32:46it's not maybe it's not you know
32:48perfectly analogous to computer
32:51programming or something what are those
32:53signs for hope is like it seems like a
32:57what do we have to be excited about yeah
33:00so I mean the big thing that's happened
33:02the last few decades really been better
33:05computers you know and really amazing
33:08imaging sensing technology so again the
33:10fact that we have computers and fast
33:12enough to really punch through the
33:13massive amounts of data to order like
33:15much as Potomac but also brain imaging
33:19and from just watching just watching
33:23themselves in action using the first
33:25reporters or just gathering is very
33:27large you know image to do datasets that
33:30sort of our lens into what's happening
33:31is like those in fact we have these
33:34sensors that can do this in a better
33:35sense adjustable photon you can just
33:37stream this information in terabytes a
33:40minute into GPUs and actually crunch
33:42data reduce it again looking for real
33:44patterns in it I think the fact that we
33:47have such amazing tools for measurement
33:49and they keep coming out I mean we have
33:52sequencers that are incredibly cheap to
33:54run right now in fact with man of course
33:56you can buy even buy a thousand dollar
33:59sequencer it's all right now actually
34:01run it yourself it's very impressive
34:03these things are they only gonna get
34:05better and better when you're seeing
34:06nénette' where technology might soon
34:07allow us to actually directly sequence
34:09proteins and to do you know to do
34:12surveys what's inside a cell very fast
34:15way these things are gonna they're gonna
34:16keep coming out you know so a lot of a
34:18hard Tech is invading biology and really
34:21giving us many many new viewpoints and
34:23lenses and it's making data much more
34:25cheap and much more reliable and regular
34:28people what we've ever had before
34:30biology so just our science into the
34:33system is growing very fast that's the
34:36thing that gives me a great deal of hope
34:37and actually being able to understand
34:39what's going on inside cells and to be
34:41predictably change them
34:46yeah it's just from the aging aspect of
34:49it my two oven where I come from like
34:51you know in the 2006 2009 era kind of
34:54when you were doing Aging in a lab
34:55you're really taking rooms and they were
34:56like if we change one gene like what
34:58happened to this room like if we change
34:59it to jeans oh like that that's a big
35:01deal I think I put two things together
35:02steal one that lives and we're getting
35:04you know like 10x or so increases in
35:05lifespan doing that kind of thing I
35:07think it's really just to kind of see
35:08how that's evolved the clinic can kind
35:10of see a lot of companies coming along
35:11and are funding us and some of these but
35:13like a lot of companies are targeting
35:14the same pathways that we were working
35:16on kind of in that family or kind of
35:18that we see in different models of Aging
35:20and you know you know maybe most of them
35:21will feel like they made their about 50
35:22different ones that I would say are
35:23reasonable guesses but kind of like you
35:25have this feeling of hope that like you
35:26know maybe the first or period that
35:28actually affects lives but you know not
35:29like she lacks maybe more like 30% 50%
35:31max if you like a bite a bunch of things
35:33together maybe but maybe those first
35:34therapies that affect lifespan and are
35:36collecting aging are kind of already on
35:38their way and in the clinic and so I
35:40think I know what we're talking about
35:40today is more you know like if you think
35:42beyond just kind of the the single gene
35:44kind of focus this the the kind of
35:46limited view of biology today what we
35:48were what more dance could we get if we
35:50looked at things from systemically and
35:52that's a really exciting and totally
35:53unknowable kind of open steal right and
35:56that's kind of where we're headed so
35:57it's like an exciting time to be a
35:58biology in general I think I
36:01am very optimistic about it and that the
36:05you know we've had medicine for
36:07thousands of years but medicine has been
36:10only a top-down empirical pursuit thus
36:14far you try a bunch of things you see
36:17who gets better and who dies and
36:20you go with what works right you know
36:23when Epocrates was building a hospital
36:26they hung some stakes in different parts
36:28of the town and the one that developed
36:30maggots the slowest is where they built
36:32the hospital so they didn't understand
36:34what maggots were why that that worked
36:37but that it's just sort of that for
36:38works and so they went with it and you
36:41know we developed things like the
36:43double-blind study and
36:45and and the doing trials and made it
36:50more rigorous but fundamentally was
36:52still a talk Suz and now we're getting
36:55to the point where we can do
36:57biology and medicine and health on a
37:00bottom-up sort of way where we wear
37:03biology is getting to the point where we
37:06can really start to understand things
37:08from first principles and then not just
37:10deal with what works but construct the
37:13physiology and the anatomy that we
37:15actually want to have
37:18Thanks thank you guys offered the day
37:20with us and thanks for attending
