Dark Galaxies, Fuzzy Dark Matter, and an Alzheimer’s Breakthrough

Hello Internet + lineup (dark matter + Alzheimer’s)

0:00dark matter quantum mechanics that is what we are worried about in this particular scenario. It's it's a very weird way to think about it but it might be the case. So this new paper from UC San Francisco talks about links between Alzheimer's disease and exercise but the conclusion was not what I thought it would be. Hello internet. This is your captain speaking Lester Nar joined as always by my co-host and our resident PhD Krishna Chowdery who is today dawned in his Ferrari race kit. We will get to that later in the rundown. But on this week's episode, we have two great main stories. We're start off with an astrophysics story about dark matter and

0:41we're going to follow up with a look into the latest in Alzheimer's research. For those of you who may be new to the show, be sure to like, share, and subscribe. If you want to support the show, head to our website. We have a great donation portal. And as you already know, we're going to learn about the science from the ground up today because this is from First Principles.

Story 1 begins — candidate dark galaxy in Perseus

1:17So, we're going to begin today with an astrophysics story about dark matter. As we mentioned, specifically, scientists think they've found a really good candidate for a dark galaxy. This is a galaxy almost completely made up of dark matter, no normal matter. That could give us a little bit of a hint into what dark matter actually is, which continues to be an enduring mystery. So let's dive in on what the story is about. >> Yeah. I mean the central question that almost every cosmologist is trying to answer is what is the universe made out of? Okay. The current paradigm is

1:57something called lambda cold dark matter. This is the model for what our universe is. >> The normal atoms that we see are only about 5% of everything in the universe. It turns out that 24% of our universe is something called dark matter, which we have no idea of what it is. All we know is that it is like matter in that it interacts with gravity, but it's not like matter in that light does not care about it. The electromagnetic field does not care about it. So, it's not so much as dark as it is transparent in every sense of the word. Like a light beam will just go through dark matter and not interact with it at all. This is very weird because atoms love interacting with photons, right? Because they're

2:39electrically charged. >> Obviously, electromagnetism is a thing. And then finally, there's about 72% that is like dark energy, which we're not even going to get into this time. But just just to set the stage, right? Like we only we only know what 5% of the matter is in our universe. >> That's pretty pretty insane. >> I can you just repeat that really quick?

ΛCDM in one minute (5% / 24% / 72%)

3:00>> It's like 5% normal matter. That's the stuff that we're made out of. Adams, berons is what it's called. >> And the stuff we can see. >> Mhm. Yeah. And then and then 95% is stuff that we have no idea about >> and we're trying to figure out. Yeah. >> It's weird. Weird. It is quite weird. >> Right. And and what this particular story is about is something called a candidate dark galaxy. This is the second candidate dark galaxy, but it's the one that like we can we have a lot of information on because the first one was kind of, you know, um it's it's it's

What “candidate dark galaxy” means; Cloud9 comparison

3:34a massive dark matter halo that failed to form a population of stars. Hubble first identified it and now they've worked together with the ESA Uklid Observatory that's also in space to figure out some more stuff about it. It's got an extreme dark matter fraction, meaning 99.9% of this galaxy is dark matter. >> That's way off compared to the Milky Way, which is in the like '9s, 80s, '7s type of thing. Okay, this is this is really weird. Okay, >> we had actually previously done a story on on uh dark matter. >> Yeah. So, something called Cloud9. >> Yes, that's >> And that had no stars.

Fritz Zwicky + Coma Cluster origin of dark matter

4:14>> Yes. >> This one has a few. >> And so that was what they were. Is this the first relic which was the abbreviation? >> Um and so but this is different than Cloud9. >> This Yeah, this this could be a proper galaxy and there there are differences that we're going to highlight. >> Okay, perfect. >> Okay, so let's give some historical context. There was this guy Fritz Vicki. He was at Caltech. He coined the term dark matter >> in 1933 because he observed this particular cluster called the Koma cluster. M he observed a bunch of galaxies in the coma cluster and he figured out what the velocities of these galaxies were >> and he figured out that how fast these

4:54galaxies were moving around some central mass distribution >> it didn't make any sense they were moving way too fast he used something called the viral theorem which is a theorem in basic physics that connects kinetic energy to potential energy okay you can apply the varial theorem to the solar system for example And everything makes sense because the farther you are out, the slower you move. That's a the farther you are out, the more potential energy there is. So the less you have for kinetic energy, right? When he applied it to the coma cluster, those observations didn't make any sense. There was way more potential energy from gravity than what he saw. And so he said, there's got to be some dark matter

5:34there that we can't see that is providing the gravitational potential energy. There has to be a source for the pole that creates these orbitals. >> Yes. >> And normally we can see where the source >> of the gravity is. >> Yeah. Normally it's stars, right? And it'll be bright >> and you can't miss it. >> Yeah. >> So the thing that's weird about this is we're looking things are moving in very fast. >> Very fast. >> But the thing we can see that would drive that movement

Vera Rubin rotation curves (why we know DM is real)

6:03is not large enough to to to explain why they're moving so fast. >> Yeah. And and so there's got to be some stuff there that acts like matter in that it's gravitationally bound and it creates a gravitational potential, gravitational force, but it doesn't act like matter in the sense that there's like no light coming out of it. Right. >> Okay. And that's what's weird. Um there were other observations. Vera Rubin very famously solid solidified the concept in the 1970s with galaxy rotation curves. What you can do is figure out how fast stars are moving >> around a galaxy. Yes. And what you expect is something like the solar system. Again, the farther you are out, the short the smaller your velocity, the

6:45slower you should be going. That's why a year on Jupiter is like 15 years because it takes 15 years to go around the sun. That's not just because the orbit is bigger, like the circle is bigger. It's also that it's moving slower, right? It shouldn't be that long. But it's moving slower. That's why it's that long. If it was moving as fast as the Earth, it would be like 5 years or 3 years. And this is this is a similar graphic we talked about in our our Ver Rubin deep dive. We're looking at a graph now where we see the expected uh the expected path for the visible disc sort of dissipating on the x-axis. >> Yeah. And that would be like if it was it if it was normal if the farther I get away from the galaxy the the slower I

7:25should be moving. What we actually see using starlight and the hydrogen 21 cm line 1420 MHz is that actually you're just getting faster. So there's a bunch of invisible matter that is pulling on you that we have no idea how to see. Okay. >> So >> the question is what what is what is going on? What what is this dark matter? Well, the more and more observations we did, the better we got. In the 1980s, people discovered something called low surface brightness objects. These became known later on in 2015 as ultra diffuse galaxies. Okay. In 2015, we made a discovery using something called the Dragonfly Telephoto Array. Okay. This is

8:07a new array that was built by Yale and the University of Toronto. It's very cool. Actually, you as a camera guy, you're going to love this. So, each of these lenses, so it's in a honeycomb structure, right? And it's called Dragonfly because the dragonfly eye very much looks like this. It's a bunch of like honeycomb lenses. >> Yep. >> Each of these lenses is just the offtheshelf Canon 400 mm f/2.8 lens. >> Really? >> Yeah. It's just >> No way. >> It's there's they just they just bought a bunch of them, put them in a hexagonal array, and then they're compiling the light together >> to create these images. >> That's actually incredible. >> Right. This is something you can get like on Amazon, but they they put a

8:47bunch of them together. Now, the question is, why would you do this, right? >> Yeah. Yeah. Yeah. Traditionally, you never want to use lenses because the bigger the lens is, one thing you get is aberration, which is the idea that the bigger the lens is, the weight of the lens is going to >> deform the lens, >> right? Because glass is not like >> completely metal. So, there's some fluid property to it. So, the larger you make the glass lens, it's going to sort of like deform because of the weight of its own >> gravity or the other way around. And so

Dragonfly Telephoto Array (Canon 400mm array) + ultra-diffuse galaxies

9:18you're going to get image distortion that way, right? But what these guys did was a bunch of small lenses put together. Now what is the advantage? Now the advantage is for mirrors, right? If you have a mirror lens, which is what modern telescopes all use, they all use mirrors. The light has to come down and then has to get sent back up because it's a mirror. So it's going to bounce up and there's a secondary mirror that then bounces the light back to a detector. Usually it's like warranted in some way or something. But at the end of the day, you need a secondary mirror. >> And that mirror needs to be held up by some kind of scaffolding. In the James Web Space Telescope, it's hexagonal scaffolding. In the Hubble Space Telescope, it's diamond shaped

10:00scaffolding like in a square 90°. Which is why when we see the images that these telescopes produce, >> the mirror structure or the architecture actually dictates. For example, I remember when uh James Webb first came out, certain objects had a certain star pattern or or like a certain visual representation that little like little Yes. that was different than Hubble, which is because of this diagonal versus hexagonal structure. >> Yeah. And those are called diffraction spikes. It it happens because of scattered light from that scaffolding. >> Yes. Well, if you want to image really really diffuse things, that becomes a really big problem.

10:40>> When when you say diffuse, what do you mean? >> Diffuse by really dim. >> Okay. Really? >> Like not a lot of light. Got it. >> As in like so little light that we're pushing the sensitivity of the CCD, the actual um charge couple device, the thing that is collecting the light and turning it into an electrical signal. Right? There's a bunch of silicon at the end of your detector. That's what makes up a charged couple device. And when the photon hits that silicon, it releases an electron and then you're like, "Oh, I got a little photon from there." Right? >> If you're if you're imaging really diffused things, even that the scaffolding, the scattering from that scaffolding is going to be a problem. So that's why you want to use lenses, but

11:22you can't use big lenses because of the aberration, like the deformation problem. So what if these guys just figured from the dragonfly telephoto array, what if we just use a bunch, right? like we used like 20 or 30 small lenses >> and the compile >> and we compiled them together. Okay. And it's it it was like relatively cheap I can imagine, right? Because >> you're not like making custommade lenses. This is just Canon 400 mm f2.8 that you put together. >> Calling up Carl Zeiss. We need a custom giant 13 meter. >> And he's like, uh, how about $10 million? Like, no. So, so this was this was a really cool thing. And in 2015, they came out with these extreme ultra

12:04diffuse galaxies. Dragonfly 44 is an example. >> This is a galaxy. You can see the faint fuzz in the middle. >> Yes. >> Not a lot of stars. Yes. >> And from what we can tell from how these things are moving around when you do spectroscopy and you figure out the Doppler shift. So you figure out how much is the red shift in this direction, the blue shift in this direction. From there, we can figure out how fast these things are moving around. >> Yep. And then you back calculate how big does the galaxy have to be? It's like dark matter dominated. 90% of it is dark matter. 95% of it is dark matter. So it's this ultra diffuse galaxy that seems to have a lot of dark matter. Okay. It's been hard to catch before because usually with normal telescopes,

12:45the scattering takes that away. But now with this new dragonfly telephoto, we can start imaging really diffuse structures. >> That that makes it. So the the point is with the normal structure uh we need bright objects for it to work well. >> Mhm. >> Um and these sort of dark matter galaxies where it's predominantly this this thing we don't understand as that does not interact with light. It becomes very difficult to see it or to to image it I should say. >> Yeah. To image it. >> To image it. >> But with this lensbased architecture it lowers it increases the >> signal to noise. >> Yes. such that we can at least get >> something >> something. Yeah, exactly. And then this

13:26particular paper is candidate dark galaxy 2. It is taking these ultra diffuse galaxies to a whole new level. >> It's pushing that diffuse into now dark because now we're getting 99.9% >> dark matter.

Missing satellites problem (why we expect hundreds more halos)

13:42>> Okay. And that's really crazy. >> Okay. >> Yeah. >> So why do we even care, right? this intensive hunt for something like a candidate dark galaxy. It's motivated by two crises that are in the center of this cosmological model that we called lambda CDM. You know that thing I told you before about 5% normal matter, 25% dark matter, the rest being dark energy. I think it was this one. Yes. >> Yeah. Exactly. So this is the model that we have today. Yes. >> Now fine >> if if that is the model, right? There should be consequences of that model. There are certain consequences of the model that we do not see. >> The the point being we have an understanding of the universe. >> Yeah. And we're like this is our best

14:23guess. And then there's always people who are like but that doesn't really make sense for these two reasons. >> Got it. It doesn't it doesn't it it doesn't compute. >> Yeah. It doesn't compute. And these are the two things specifically that do not compute. The first one is the missing satellite problem. Okay. these dark matter simulations. When you use that original 5% normal matter, 25% dark matter, blah blah blah. You press play on a universe, you set it up in the computer, and you press play. Simulations predict that there should be about 500 orbiting dark matter sub halos around something that is the size of the Milky Way. >> Okay, >> the idea is for every big halo of dark

15:03matter condensing, right, dark matter is still gravitationally bound. So the universe is expanding. There's a bunch of dark matter. There's going to be tiny fluctuations where there's a little bit more matter here, a little less matter here. And so it's going to start clumping because gravity is going to take that over dense region is what we call it. And we're going to start clumping that way. Now around those big clumps, there's going to be smaller clumps. Around those smaller clumps, there's going to be even smaller clumps. There's like a power law structure. And >> what we expect is there should be a lot more small clumps, right, >> than what we see, >> right? That is the missing satellite problem. >> We we we see big a couple big clumps. >> Yeah. >> But the cascading the second and third order clumps we should be seeing are not

15:44present. >> Exactly. And that's what we're seeing over here. Um the the triangles with the error bars, that's observation. Yes. >> The lines are theory. And you can see that for >> larger >> increasing mass, we're fine. >> It So when the mass increases, the observation matches the theory. Yes, but for these smaller clumps, these satellite galaxies, there's a massive discrepancy orders of magnitude >> because because of the power law, we should be seeing significantly more smaller dark matter uh clusters or halos >> than the large ones. But the the rate of increase of the number from observation

16:24is very like the quite low. >> Yeah, there's like there's like a point where it's just like we just we just see way too few. Yes. Okay. So the question is, I mean, maybe maybe these smallest halos are present, but they're completely dark, which is why we don't see them. There's no stars. So like, what are you what are you going to see? >> It's a bunch of cloud nines from our from our star dark matter cloud. >> Exactly. And so Cloud9 was trying to solve the similar problem. And this is kind of solving that similar problem too. If this is truly there, right? >> Okay. >> Now, the second, and this is something that Cloud9 was not doing. So this is novel novel. It's called the cusp core

Cusp–core problem + why a no-star halo is a clean test

16:57problem. >> Here's the idea. Dark matter does not interact electromagnetically, right? There's no like, you know, electrons, they want to they want to move away from each other >> and protons want to move away from each other. Even when we have atoms, these atoms, you can't just compact them. >> But if you have a particle like dark matter that is only interacting with gravity, >> that means that the closer you get to the center of the gravitational potential, the density should just get higher and higher and higher because gravity is just clumping everything together. There's nothing that's pushing out like when we think about like like hydrostatic equilibrium and things like that but there's nothing pushing out. So the closer you get to the center the density of dark matter should be higher

17:38and higher. That is not what we see. We see instead of a cusp which would mean in the center we get like a little pointy. >> Yeah. >> We get a core type where it's like at the center the dark matter density sort of flattens out. >> Interesting. >> That's weird. >> That is weird. If dark matter is purely gravitational, why would it just >> flatten out? >> It it could should be continuing. >> It should continue to get bigger and bigger as it gets closer and closer to the center. The density should be higher and higher. That's not what we're seeing now. Okay. The defense that the um lambda CDM people make, like the guys who make the cosmology model and they're drinking the Kool-Aid, right? They're like, "No, well, there's something called berionic feedback. This is the

18:18idea that towards the center of the of the galaxy, there's a bunch of supernova explosions. >> Okay. >> Okay. And that supernova explosion is going to send shock waves in spaceime. And those shock waves are sort of going to diffuse the center of that galaxy. So even though the normal dark matter would create a cusp. Yes. Right. Because there's so many explosions happening in the center of the galaxy that's going to sort of flatten out. Right. Kind of makes sense. >> Right. That could work. >> Okay. But this is why this particular galaxy is significant. This galaxy has no stars. Basically no stars. >> So there's no supernova expos. >> Right. Right. >> Right. Which means that this should definitely have a cusp. >> I see. >> That is why it is worth studying.

18:59>> So So the CDG2 >> Mhm. >> it's interesting because >> Yeah. And by the way CDG2 CDG means candidate dark galaxy for for those Oh >> no. My Siri is >> I'm trying to figure out what someone is, one of the intelligence agencies is apparently trying to hack our phones at the moment. >> Um, but this is interesting. So, there's currently a cosmological model >> of normal matter, dark matter, and dark energy. Like what is the composition of the universe? >> Yeah, >> that model has two problems that we just talked about. The missing satellite problem and the cusp core problem. the the folks who are in defense of this

19:42lambda >> CDM >> CDM believe that supernova explosions is the reason that we don't see a cusp. >> Yeah. >> Sort of like like you know blurring stuff. >> It moves around. It moves around rather than aggregating it. Let's let's say for example, right? Um the problem is observationally we've now seen something which is the CDG2 that has a very few amount of stars which means very few amount of no supernova explosions. >> Yeah. No one's dying cuz no one's alive. >> So how do you so so no? >> Yeah. So so so this this galaxy can now offer a test to your lambda CDM. >> So and this is actually an interesting

20:22point about understanding the intersection between theoretical >> sciences and experimental sciences. And the experimentalists are constantly in the real world looking for >> ways to test ways to test. >> You need some sort of stability blah blah blah like something that can be consistently >> measured. >> And we have something now. >> Yes, we finally found something. And we haven't done spectroscopic analysis because this we we just found it. >> That's why it's in that's that's why it's interesting because now we can we can get time on the James Web. Subaru Mellraight. >> And I just want to know because getting

21:03time on these instruments is >> is everyone there's very few of them. Everyone thinks their thing that they're studying is the most important thing in the world. >> But this seems to be because of the importance of dark matter generally as a concept >> a good candidate to be able to quote skip the line. >> Yes. >> Okay. Exactly. And so I want to make a case for why this is important, right? And that's what these people did in this particular um paper, right? The

How you find dark galaxies: globular clusters as tracers

21:30challenge is how do you defeat sky glow? Sky glow being dark galaxies, these these types of dark galaxies that I mean there's no there's no stars. So what are you going to look at? Right? They're super faint. The signal is often so faint that there's a statistical noise limit when it comes to like your CCD, your charge couple device, the sensor in the back of your camera that is like looking at this thing. There's a noise, right? It's going to catch stray photons and that noise is going to be higher than the signal that we're getting. So, we can't just like look at the thing. Right. >> Right. So, how do we actually find this thing? Well, >> this is for earthbased >> even. No, this is even Hubble, dude. >> Oh, interesting. >> Yeah. >> Interesting. >> Because the Hubble has a CCD on it. >> I see what you're saying. >> And even though it's in outer space,

22:10like that thing has a noise. And then there's there's something actually called zodiac light. >> Okay. >> There is dust in the solar system, right? And that dust is going to scatter sunlight into your detector. Wherever you look, there's going to be dust from the solar system. And that is going to have a faint glow. And if the thing that you're looking at is a smaller glow than literal dust in our solar system. So what? You're we're going to put a Hubble like outside the solar system. We're not doing that. And then wait like 4 days to get signal back. Like well, first of all, getting out there is like, you know, it's ridiculous. So, so, so we need to figure out other proxy ways to to look for these ultra

22:52low density galaxies, these candidate dark galaxies. Okay. >> Okay. >> And we you the proxy that comes up is called the globular cluster. These are some of my favorite objects. They're dense, ancient, they're bright, and they're point-like sources. They're way, way smaller than galaxies. They're like these old old relics of galaxy formation. >> Okay. Not to be confused with the acronym relic, just to be clear. >> Yeah. Yeah. Yeah. Not to be confused with the acronym relic. There's so many of these that are around our Milky Way. They're really small. Okay. They're going to be like hundreds of solar systems to thousands of solar system. Really small. >> It's really small. Like like and the density of stars in that in that like

23:34region >> is is insanely small. like you'll have like like you know the closest star to us is like four light years away over there. I think it's like on the order of light days. >> So it's like it's like comparing the population density of Korea Town in Los Angeles to like Pikipsy New York or something else. >> Yes. Yes. We are we are that New York. >> Yeah. Yeah. We are we are Pikypsy New York but these galaxies look like Korea Town. >> Yeah. Yeah. Yeah. I You shouldn't say New York because whenever I think New York I think New York City. It's like So try let's try like uh >> Pikipsy is like the middle of New York State, but let's say let's say like I don't know where that is.

24:15>> I only know New York City. >> There's only one New York. You could say like >> you know Nebraska, >> right? Right. You know Omaha. >> Yeah. Omaha. Nebraska. >> Om. Okay. Fair. Fair. Exact. Let's do that. But that's an that that's a that's a helpful reference point to so these things are really small and so imagine and these globular clusters they're actually really important for um just history astron astronomical history um hero Shappley at Mount Wilson at the Carnegie Observatories which we're going to have a special coming out soon. >> He mapped that the sun was not at the center of our galaxy by using these globular clusters. He observed a bunch

24:55of globular clusters, their positions, >> how far they were relative to the galactic disc, and he could figure out we're actually off center. >> Uhhuh. >> Right. We're not we're not in the center of this galactic >> Yes. >> um these globular cluster distributions. So they're really tiny little specks around the around giant massive galaxies, right? And so they become really nice for observing a dark galaxy because the galaxy couldn't be there. But if we observe a bunch of globular clusters, there's got to be something that is gravitationally binding the globular clusters in that area. >> This this vaguely reminds me, and I know I'm always making these stretch connections. When we talked about the

25:36radius of the proton story last week, >> the idea that you cannot directly observe it because it's so small. And so we have to use these like second and third order >> reference points in order to like compute the value. M and similarly because we can't because dark matter does not interact with light we're using the proximity and the movement and acceleration and angular momentum and etc etc of these globular clusters to derive understanding of this dark matter. >> That's exactly right. Yeah. And so this is what they finally did. They they looked at archival data from the Piper survey which is of the Perseus cluster. There's a giant galaxy cluster in the

26:17Perseus constellation. Okay. And um

Re-analysis with DOLPHOT reveals a 4th cluster (data-analysis matters)

26:20there was this survey by Hubble Space Telescope. When they looked there, they reanalyzeed that data with something called Dulfot, which is different from DaoT. It's like a a way to analyze data from Hubble Space Telescope, taking into account like the engineering of the Hubble Space Telescope, what it's capable of, things like that. And they revealed in this Perseus cluster, they revealed a fourth globular cluster >> that they hadn't seen before. If you look over there, there's three blue dots. And then there's a red dot. >> That red dot is a new globular cluster that they hadn't found before. And from that they can be like now well wait if there's a new globular cluster there's four four being in the same small region

27:02is really unlikely unless there's something in the middle there that is causing them to clump. >> I want to make a brief note here which is this important kind of note about how the importance of data analysis. Yeah. because you have a raw data profile and this is the same raw data profile. >> The same raw data profile of like what had happened and it's just literally you there's no interpretation on that raw data yet. >> However, depending on the parameters and how you filter >> that raw data, there's a whole spectrum of different conclusions you can make depending on what you're looking for. And so the idea here is that they change the parameters. >> Yeah.

27:42>> And by ch which is not changing the fundamental data. No, the raw data is the same. >> It's like if you have a camera with a 400 mm lens and you're trying to look at something that's really close to you, it's just going to look like a blur because a 400 mm lens is trying to look at things far away. But it thinks, but if you change to a a 30 whatever 24 millimeter lens, all of a sudden >> you can see the thing because you change the way in which you're trying to analyze the information. >> Well, I want to be cautious there because what you just talked about is actually different raw data. >> Ah, okay. >> Right. If you're switching out lenses, the raw data is different. >> Okay, that's fair. >> Right. Here, I do want to make a distinction because this is this is important. >> Here, we're doing the same we got the

28:23same photo. >> Yes, that's a good point. >> But we're now putting it through >> like like you know on on your computer on Photoshop. We got like Photoshop the newest version versus like the old version. And now we can do like, oh, contrast this way and like and like the you know the you know the color thingy and you're like oh I want to I want to do this color more and this color less. That's a very that's a really that's the data analysis. If you switch out a lens that's raw data you're you're switching out of engineering. >> No that that's that's a very fair point. The the the when the Photoshop analogy is correct because >> yeah this is they did like dull fat Photoshop instead of Dowot Photoshop. >> It's a different Instagram filter in >> Yes. And they're like, "Oh, there's four

29:04instead of three." Yeah. And Exactly. >> Okay. Yes. >> Exactly. But they did Instagram on the Hubble. >> On the Hubble. Yes. Okay. Yes. Yes. Yes. Yes. >> Yeah. And then and then they found this fourth one, right? And that was critical. And when they stacked, so then they they got their ESA colleagues. >> Yep. >> Um at the European Space Agency. European Space Agency has something called Uklid.

Hubble + Euclid stacking reveals ultra-faint glow; DM fraction ~99.94–99.98%

29:24>> Okay. >> And they used the Uklid image of the same spot with the Hubble image from the Spain same spot. stack them together >> and now you could reveal a same the same ultra faint diffused glow in between those four globular clusters. Those four globular clusters that were there, there was actually a little bit of light >> coming from the middle. So there are a few stars, not that many, >> but there's a chance. >> Uhhuh. Okay. And this is by cross referencing data sources, >> you can get a little more uh >> uh fidelity. >> Exactly. >> Okay. Yeah. And so with that diffuse glow now you can start fitting models and you can say okay what is the amount

30:07of stellar mass >> that's in that diffuse glow right in the middle right the amount of light that I'm getting is very little but from that I can make an estimate given how far I know it is like how far away it is um the Perseus cluster we can be like okay that's about 10 the six >> um suns so it's about 2 million suns okay it's not a lot But from what we know about the number of globular clusters and how luminous they are, we can also back calculate the amount of dark matter that should be there. >> And the dark matter is something like 10^ the 10. >> Mhm. >> So 10 billion suns.

30:49>> So we've got 10 billion suns worth of dark matter and only 2 million suns worth of normal matter. Meaning that the halo mass fraction, the amount of dark matter versus normal matter is 99.94 to 99.98%. >> This is where we get the 99 number. Like that's >> the size of the dark matter halo is about that of the large melanic cloud, which is one of the it's the biggest satellite galaxy to the Milky Way. You can only see it in the southern hemisphere, but the number of stars there is like a thousand times less than what we see in the large Magelenic cloud. M >> okay. So this is this is like a really really tiny tiny fraction of what should

31:30be there. There's not that many stars. There's like one in a thousand stars of what it should be. >> Right. And that's >> and that's and that's in that goes in direct that experimental data >> goes into direct conflict with the Lambda CDM. >> No, no, no. It doesn't it doesn't necessarily go into direct conflict because the lambda CDM does suggest that there should be a lot of dark matter galaxies. It's just now we found one. >> I see. >> Right. And now this this is where we get into how how could this be possible. Right. Okay. >> Now the first hypothesis is something called the failed giant hypothesis which is something like Cloud9 which we

32:11covered in a very recent episode. Yes. The idea is you've got some massive halo that's destined to be a galaxy

Formation ideas: failed giant vs ram-pressure stripping

32:17and then early star formation makes the globular clusters around it but then star formation somehow abruptly just like stops and so you just get a bunch of dark matter a few stars and these globular clusters around >> if I recall there's like some threshold where it's right in the sweet spot where it it will maintain versus continuing to >> generate though the weird thing is this is bigger than that threshold though >> I did check >> and this 10 the 10 um solar masses. So 10 billion solar masses is larger than the threshold that we had over there. >> So it shouldn't just be like cloud 9. >> Yeah. It's not like cloud 9. Otherwise, it would be a relic and there'd be no stars. There's still a few stars. Fair, right? Um the other thing could be like something called a quenching mechanism. There's something called RAM pressure.

32:59So this is a photo of the Perseus cluster. You can see a lot of the um starlight is coming from these large galaxies and the blue sort of haze is from the dark matter that we've mapped out because of all of the velocity measurements that we've taken. And one can imagine that in this really vibrant cluster, you had this early dark matter halo with a bunch of dark matter and it was like it had a bunch of cold gas is what we call because cold gas is like the gases are moving around so it can clump in and form stars. So, it was destined to be a galaxy kind of like the large melanic cloud, but maybe like some other galaxy just rammed into it and then stripped it of all of the star

33:40making material. So, now it didn't have any star making material and it's just stuck being like this dark galaxy. >> That's actually interesting. >> Okay, so there's these there's these like theories and you know more testing and more observation is going to tell which one is correct. What I'm really excited about is what I was telling earlier, which is this cusp and core model, right? Like which one why why do >> why does dark matter not form a cusp, >> right?

Next step: spectroscopy → cusp or core?

34:08>> Mhm. >> We need to do more spectroscopy. We don't have that data right now. All we have is phototric data. What we'd have to do is figure out the >> the spectra to figure out the red shift and the blue shift and all of the really nittygritty details of what that dark matter clump looks like. Because what we'd expect given that there's no stars is there's no smoothing out. >> And because there's no smoothing out, right? Then the >> we should have a cusp. We should have a higher density of dark matter in the center and we should have low density out in the in in the in the outside. We shouldn't have this flattened profile where in the middle there's just a bunch of dark matter, but there's no real big

34:49over density right at the center, right? there's no like clump of dark matter that's super dense in the center. So that is that would be the case if we accept the current models of lambda dark matter which is lambda CDM which is the idea that dark matter is some kind of particle. It doesn't interact with itself unless >> it's through gravity >> and now we get a cusp. If we see the cusp great >> right >> okay if we don't see the cusp what does that mean? So, so what we're saying is as we go out and get more data like spectroscopy data, >> um it will further clarify is there a core or is there a cusp?

35:29>> Yes. >> And if there is a cusp, >> then we're good. >> Great. This lambda CDM well done. >> Yeah. >> The experimental data is now further validating the theory. Mhm. >> If it continues to be a core, which is kind of what we it appears to be. >> Yeah. It appears that like >> it appears that way. >> Yeah. >> Then we got to go look back at the drawing board a little bit. >> Yes. Then we got to be like, what exactly is dark matter? Is dark matter really just like another particle like a proton or an electron. That fine. The only thing that's different about it is it doesn't have any charge and it doesn't interact with electromagnetic field. It doesn't interact with light.

36:09Right? That's that's the normal thing. And that's like like naively I would expect that it's just another particle but it doesn't interact with light which is why we see its effects across giant time sc giant scales of the universe but not like you know nitty-gritty stuff. >> This one is pretty cool. There's something called fuzzy dark matter. This is very different from lambda CDM. Okay.

Fuzzy dark matter + galaxy-scale quantum pressure intuition

36:30>> Okay. Fuzzy dark matter means that dark matter is not tiny little like particles, but it's fuzzy in the sense that it's super light, meaning billions of times lighter than the electron. And what does that mean? If it's billions, and that's and that's some of the lightest stuff that it's going to be lighter than quarks, electrons. It's it's at the order of like even lighter than nutrinos type of thing. M >> it is so light that if we were to calculate the quantum mechanics of a particle that's that light the wavelength the debrogley wavelength because you know remember in quantum mechanics every particle is a wave and blah blah blah blah blah well the larger the particle is the shorter the debrogley wavelength

37:11okay so electrons have a wavelength that's like around like ultraviolet x-rayish this thing at 10 the minus22 electron volt mass would have a wavelength that is hundreds of light years big. >> Oh my god. >> Right. Its quantum wavelength if you were to create a quantum wave out of it would be the size of several >> like it would be like a big chunk of the universe. I mean sorry of the galaxy. >> Right. This is so so the wavelength the wavelength would be larger than the distance between them. >> Right. >> Okay. >> Now what does that remind us of? That reminds us of Bose Einstein condensates.

37:55>> Yes. >> Because in a Bose Einstein condensate, the wave the waves of each particle >> pile on top of one another to create a single quantum wave. And that's why they don't like sort of clump together. So what this particular theory is saying is there's some kind of inherent quantum pressure. >> Yeah. Yeah. >> That's at the scale of galaxies >> that is creating like a core >> with a flat density. You've got a Bose Einstein condensate of this fuzzy dark matter particle that is at the size of a galaxy and that's why you don't get a cousin. >> Yeah, that's so >> it's so fascinating. So this this could be, you know, this um candidate dark galaxy could be a proving ground

38:35>> for something like fuzzy dark matter. It's very interesting >> from my layman's perspective. I like I like only because we've so recently talked about uh Bose Einstein condensate. So, I have a mental model of what that actually means. >> And so, I'm like, "Yeah, that sounds right." >> Yeah, that sounds about right. That would be so cool. Imagine a a particle whose quantum wavelength is the size of a galaxy. >> That's unbelievable. That's unbelievable. >> Very cool. >> So, I mean, so this is this is quite a big deal in, you know, in that you I'm curious. I would be curious to see >> how other folks and institutions begin to sort of digest and interpret the

39:16paper. But I think like we were talking about earlier, there are certain things that happen when it happens. It's like, okay, all the big guns point over there. >> Let's point over there. >> Um, you think it has the potential to reach that threshold? >> I think so. I mean, I mean, we know where it is. >> Yeah, we know we know like sort of the base spectrum spectrometry data. So, I can I can imagine there's going to be a lot of follow-up studies >> on this from all the top telescopes and institutions >> cuz why not? >> Yeah. Yeah. It's a it's a testing ground for like particle physics in some sense. >> Yes. >> Right. >> Yes. >> Which is very cool. >> This is very very cool. This is very very good. This this was out of uh the astrophysical journal letters.

39:58>> Um and we have our institutions here. This would have been because I want to make sure we cover who did the paper for this. >> Yeah. Yeah. Yeah. >> It was uh I'm just pulling this up now. >> Yeah. So is in astrophysical journal letters and it includes oh several institutions. Yeah, it's usually it's usually several institutions. >> There's a lot of >> but if we want to name a few, University of Toronto is a big one. >> Yes. >> Um yeah, University of Toronto is one of the big ones. >> Yes. And this was in collaboration >> and Yale because like you know it's the >> the these guys love that ultra >> dark you know diffuse stuff. University

40:40of Santa Cruz, California, Santa Cruz. >> So, yeah, >> big shout outs to all the team. This is a really interesting story. I mean, I I still continue to be so fascinated by dark matter. I know friend of the pod Dan Gilman has dedicated his life >> and he he did a postto at the University of Toronto. So, I wonder if um he knows some of these guys. >> I'll tell him to I'll tell him to send >> share it Dan share it with the the team and if folks want to give us any follow-ups, comments, or corrections, please let us know. Great astrophysics first story on dark matter. kind of continuation on our Cloud9 story which was fascinating and well worth the watch if you haven't seen it to get a little bit deeper into that use case where the cluster does not reach the level where

41:20it can produce stars and so it remains sort of this like empty >> the the relic cloud 9. So, we're going to move into the rundown now. And for those who are new to the show, we usually cover several main stories where we do a big deep dive, but the rundown

Rundown begins — Rubin alerts, magnets, Rapa Nui, baby chicks

41:36allows us to talk about other breaking science research that we find interesting. We're going to start with a follow-up to a story we covered in episode 4, one of our early stories about the the Verse Rubin Observatory. Um, it has now launched its real-time discovery engine. So, this is now about 8 months later. Yeah. And it's a pretty major update. >> Yeah. Yeah. This one's insane because the Very Rubin Observatory is unlike any other observatory um out there because it's this giant thing where astronomers cannot get time on it, right? Usually usually when there's an observatory and you you're like, "Oh, like in the last story, hey, I want to look at this candidate dark

42:17galaxy in the Perseus cluster. I need four nights." And that application is going to go to whoever is in charge of that telescope time. And they're going to decide from all the applications who is going to get time to look at what. The Vericy Rubin Observatory is an observatory that is just taking a photo of the entire night sky every three nights. That is its job like clockwork. Um fully automated and that's all it's doing. It's not focusing on particular things. That's what it's doing. And what that means is it's going to find a lot of data every night. there's going to be

Rubin real-time discovery engine (800k alerts night one)

42:51new data and what it can do is historically look at what did that position in the night sky look like a month ago, a year ago, 10 years ago because all of the archival astronomical data of maybe the Hubble pointed at it at some point or maybe the James Web pointed at it 6 months ago. It can compare images between what it saw that particular night and what other telescopes have seen that particular night and if there's anything different send out an email alert. And so on its first night >> that this real-time discovery machine was operating it sent out 800,000 alerts. >> It found 800,000 candidates for new

43:33stuff that scientists could look at. Th this is so this is incredible and we we talked about this when we did the story because the technology behind uh it is unreal. Yeah. In terms of the amount the volume and the it's like going from CRT 480p TVs >> to 4K. >> Yeah. And and and so that's in terms of resolution. The main thing is it's going from photographs to movies which we talked about. So we're seeing like motion. We're seeing we're seeing for example asteroids >> cuz the asteroids will move and so that's going to be a motion transient that the Vera Rubin can catch. We're seeing supernova which are going to be

44:15oh there was no light bulb there and then now there's a tiny faint light bulb on that galaxy. That means a star exploded and we've got about you know 2 or 3 weeks to catch that star exploding. So all these transient events, these fleeting celestial events, that's what the Ver Rubin was out for, right? That's why um it's called the Legacy Survey of Space and Time, >> the LSST. That's the program. It's for 10 years. It's going to go along. These alerts are going to be insane. Um >> it's expected to generate up to 7 million alerts per night. >> Yeah. And we're already on track because with the first night it was already

44:55800,000. >> I mean this is incredible because there's not that many astronomers out there, >> right? >> Okay. In the world and those astronomers don't have a lot of time, right? >> So this is really the advent I think of decentralized astrophysics research. Like now we can have citizen scientists for example. If you're like in a high school or like you know some advanced program where you want to do scientific research but you don't have access to data this is an incredible resource now to do small little projects if I don't know you want to get ahead in your college application this is something you could do you could go on the Vera Rubin website and find data and look at some of these alerts and be like hey

45:36which alert do I want to like dig a little bit deeper into things like that. Um, you can also, I mean, this is going to be huge for machine learning and AI because handling 7 million alerts a night. No human can do that, but creating a machine learning pipeline that can, you know, sparse through all those alerts, figure out which is the one that actually might require human attention from somebody with a PhD in astrophysics and so on. It's going to be huge. I I I this is this I think when we did our end of the year episode for season one, I brought up this uh the episode four is one of my favorites because I still just think >> technically from technical execution, largest camera lens in the world. um the

46:19amount the data pipeline build like being able to actually process and make available immediately that volume >> within 2 minutes I think >> is is a significant technological challenge in and of itself >> uh and then the open sourcing of it >> right out of the box meaning there's no gatekeeping there's no this and that it's it's it's a really I think it's a beautiful testament I think to the spirit >> of what science yeah >> in the modern era can look like. There's understandable reasons why not every system is like that. >> Yeah. And I mean this is this is something that I think the US taxpayer should be very proud of. >> This is you almost entirely a US effort

47:00by the US National Science Foundation and the US Department of Energy >> DOE baby. >> Like this is this is really cool that our taxpayer dollars are going into something like this. >> Very cool. First story. Um we're going to move into our story number two. Our story number two. This one is a material science and AI story as scientists have used AI to help them find new types of magnets uh which can be used in an innumerable number of industries. So what what exactly is going on with these new magnetic materials? >> Yeah. So finding rare earth free

AI finds rare-earth-free magnet candidates (materials at scale)

47:35permanent magnets i.e. permanent magnets that don't require these rare earth me minerals that like we have to mine for that are really hard to find that maybe we have to make a deal with China with and so on and so forth like that's crucial for reducing the manufacturing cost and the supply chain vulnerabilities of electric vehicles and electric vehicles if we can make them cheaper easier to make they can be the future right not just electric vehicles I mean batteries go in everything right medical devices renewable energy systems And what this team at the University of New Hampshire has done is use an AI system to extract data from thousands of scientific reports and train a model

48:16that will predict the magnetic behavior of materials of novel materials, right? And you you have like now a searchable database of something like 67,000 magnetic compounds. And out of those they found 25 new candidates that will remain magnetic at high temperatures. >> Very interesting. Yeah, because the the hotter you make a magnet, the more jiggling there is, which means that the local alignment between your magnetic elements. Like in iron, for example, the iron has like a veence electron that creates a gross magnetic field around the atom. Why? When iron becomes magnetic, like when you take a paperclip and you put it around a permanent magnet, that paperclip becomes magnetic,

48:58right? Because all of the iron atoms sort of >> align to be make a big magnetic field. Well, if you heat up that paperclip, you're done. But if you put that paper clip in the freezer and you come back hours later, it'll still be magnetic. The point is, we want to find magnetic materials that remain magnetic at high temperatures because that will give us industry applications, right? And that's what this is doing. And it it vastly accelerates material science. >> Um, you know, because millions of time consuming lab tests can now be churned through with this machine learning algorithm. and you can find like a searchable digital resource that you can then be like okay I want to find a

49:38particular magnetic material that does this and this you got something >> this is this is great so this was out of nature communications would it be >> so it the I mean obviously the thing that comes to my mind is >> it makes me think of like a magnetic version of alpha fold >> um yeah kind of is it is >> it's like a material science version of of of it like directionally speaking details um I I think it's it's it's still a little far from I give you a novel material and you predict this is still there's all of these um there's all of these thousands of scientific papers that have already come out with a bunch of stuff. How do I centralize that and create an understanding?

50:18>> So so rather so it's more so taking the existing data and making it uh making someone be able to iterate on ideas based on existing data. Mhm. >> Alphafold was really sort of saying here's all possible protein structures. >> Yeah. From all of the sequences that we know, even the ones that we haven't found, >> which is a different thing cuz that's generating net new data. >> That's like versus just interpreting past data. Okay. So, there's a there's a difference there. >> Yeah. But it's it's getting there, right? We're getting close. >> That's very cool. Very very very very good. >> Yeah. >> Story number three is a climate science story with a bit of anthropology. >> Yeah. Uh there's new science about what's actually what actually went wrong

51:00on Easter Island. >> Yes. >> What went wrong? >> So Easter Island is very famous, right? These this is Rapanui. It has those um giant statues of um these like heads with torsos that are facing out into the ocean. And for the longest time, the

Easter Island update (climate stress + adaptation, not simple “ecocide”)

51:20popular narrative was that the society collapsed there because of self-inflicted ecoside. Meaning they cut down all the trees and then by the time that the European colonizers came, there was a decimated population that was like almost starving and blah blah blah. Okay. Um that turns out to not be entirely true. >> Shocking. shocking that the European colonizers um made up a >> a story about the history. >> Yeah. A story about the history that is cons conducive to them and says, "Oh, the natives didn't know >> what they were doing, >> what they were doing." Okay. It turns out from new sediment core data, what they did was they studied hydrogen

52:01isotopes of plant leaf waxes. >> Okay? >> And these things preserve sort of a climate history across climates, across centuries, right? So by studying the the hydrogen isotope data there, what they actually found out was that the annual rainfall there dropped sharply by almost 24 to 31 in during the mid6th century. And this is right before sort of the European explorers went and like saw what was happening, right? And so what's happening is instead of like entirely collapsing due to deforestation and conflict which is the original theory what probably happened is the Rapanui

52:42people adapted to this severe climate stress and they shifted their rituals their power structures and their ceremonial centers to respond to that climate stress. Right? And it emphasizes more like human resilience and that the history of Easter Island is way more complicated than what this overconumption narrative that the colonizers had pushed. >> Surprise, surprise. >> Classic, very fascinating. We could probably I'm sure some people would be like hydrogen isotopes of plant leaf waxes being able to be analyzed over centuries in and of itself. I would love to understand that. >> Yeah. Yeah, that's a good one for a deep dive

53:22>> because it's it's >> that that's so curious to me. >> Yeah. >> But clearly is hugely impactful >> in both this like historical climate science related research but also mapping it in this case which is what's kind of interesting to an anthropological >> Yeah. I mean it's it's a technique that's like very similar to when people do ice cores. Have you heard of that? Like when they they barrel like a ice cylinder in an Antarctic glacier. And because Antarctic glaciers are like so old, layers of that are going to give you a recorded history of the climate going back almost millions to tens of millions of years. Right? So similarly, plant waxes like leaf waxes are stuff

54:03that are very old because it's not like >> it's it's not like a plant in that it was just born. It's like it created the wax and then it's the wax has sort of persisted across centuries. It's kind of like um in Jurassic Park when you have the resin, right? Resin is a plant residue. It's not wax, but it's it's a type of residue that can last through the ages. Similarly, leaf wax is something that can last through the ages. So, it gives us like a a a diary. Yes. >> Of what the climate was like in that locality if we can really analyze it at that atomic resolution, which is what these guys did at the Colombia climate school. FA fascinating. This was a paper in uh communications earth. Um, again, I

54:46think we'll continue to see as scientific tools and instruments and the funding that goes to support those endeavors, uh, continues to grow in capability, uh, we will be illuminating more about our as much about our past, >> yes, >> as we will about where we are going as a species. Our fourth and final story in the rundown. This one's a weird story about psychology, language, and baby chickens. >> Yes. and I don't know much about it on purpose. >> And before we get into it, >> okay, >> I would like to take our audience on a trip to Bahrain and ultimately Italy.

55:26>> Okay. >> We just ended Formula 1 testing in Bahrain. >> Yes. >> The season opener is happening in Australia at the end of this week >> for F1. >> For Formula 1, of course, I'm talking about Formula 1, the greatest sport um alive today. And Ferrari is back, baby. Ferrari is so back. This is going to be our year, dude. We've been We've been We haven't had a driver's championship since 2007. We haven't had a constructor's championship since 2008. I'm not blaming anybody. I'm just saying this is this is the reality of the situation. But we've got Charlotte Clair, who I'm I'm personally a fan of

56:07Charlotte Clair. Lewis Hamilton, one of the greatest of all time. So, also congratulations to Charles. I think he just got engaged. >> He did. He No, he just got married. He got married in um in Monaco. He's a Monagas native Monaco driver. Most most drivers like go there for the tax reasons. He was born there. >> I was born in it. >> Yeah. Yeah. He was born in it. Exactly. Um Ferrari had a great testing in Bahrain. Charlotte Clair was one second clear of Mercedes George Russell. >> Okay. And I know I know everyone's going to be like, "Oh, during testing everyone is like, you know, they're putting artificial masses on their on their car and they're like gaming the system to to

56:48not." But a second is crazy. And what gives me what gives me hope is that um Hos, which is an American team that is a engine customer of Ferrari. So Ferrari's engine is the same engine that's in the Hos in the Hos car. H is also doing really well during testing. So that's like it's like a you know in science we look for coincident data >> to to suggest that there's an underlying invariant. >> And as a scientist I'm just saying Ferrari is bad. Okay. And dude we're we're we're we're going to run away with it. It's going to be amazing. >> Can I ask a quick question before we continue? Which is did did the wing thing get figured out? I

57:28>> No, no, no. Ferrari has Ferrari has an amazing wing. I don't know if we're going to >> the upside down thing. >> Yeah. Yeah. So Ferrari has a wing that flips. Yeah, >> when it goes from um a curve to a straight, we don't have DRS this season because so for for those who are also listening um Formula 1 is all about the formula, right? Which is the regulations that the Formula 1 gods produce, saying these are the rules for how you can build a car. And then every car manufacturer looks at those rules and says, where can we read between the lines and create little efficiencies that others might not? Mhm. >> And Ferrari figured out that there's nothing against the building a rear wing

58:10that will entirely flip 180° depending on what part of a racetrack you're on. Because during straits, you want low drag cuz you just want to go straight. During curves, you want high downforce because you want more grip so that you can like take corners faster, right? And those two can be can have different wing configurations on the rear wing based on the aerodynamics that you get and so on and so forth. And most people just have a flap that goes like this and then goes down. This thing is going >> is turning around 180°. And all these other teams were like, "Yeah, we thought of that." I know you didn't. You didn't think of that. Okay. Okay. You saw it and you're like, "Oh, we better like

58:51cover our No, you didn't think of that. Ferrari thought of that." That's That's right. Anyways, I'm going off about um Ferrari because this next I'm going to bring it back to science. >> Okay. Okay. >> This next rundown study, >> yes, >> is from the University of Padua in Italy. >> Okay. >> It was out in the journal Science. >> Okay. >> And that's the connection to Italy.

Boba-Kiki effect in 3-day-old chicks (cross-species mapping)

59:14>> That's it. I just wanted to go on a rant about how Ferrari is going to win. >> But um dude, this this paper is pretty insane. >> Okay. because it didn't make any sense to me why it should be possible. Have you ever heard of something called a Boba Kiki effect? >> I only know about Kiki. Do you love me? Are you riding? But I don't know about the boba. >> All right. So, let's let's pull up the the overlay. >> Okay. Okay. >> All right. So, you see the two um you see the two shapes. There's a green sort of blobby shape. >> Yeah. >> And then there's a blue pointed shape. >> Okay. And just for for listeners, so we're looking at an image that currently Can I say this? Yeah, go ahead. Okay. Okay. >> So, it currently has what appear to be

59:54two baby chickens or birds of some kind or ducks. >> Um, yeah. >> In the middle and these two shapes at the top right and left hand corners. The one on the left is a green blob. The one on the right is like a blue star kind of thing. >> Star kind of thingy. Pointy. >> All right. So, which one is Boba and which one is Kiki for those shapes? >> Oh, Boba's obviously the blob and Kiki is obviously the >> Okay. The thingy. >> Why did you say that? Oh, because in my head like I mean like this is I don't know like >> Kiki is like >> Yeah, it's like it's pointy and boba is is blobby. >> That's exactly >> like like phonetically

1:00:34>> Yeah. And this is the Boba Kiki effect. >> Oh, is it really? >> This that's entirely what it is. >> If anyone said the blue was was boba, you're crazy. >> No, literally nobody does. But here's what's crazy. You know, you would think that, okay, maybe we're we're you know, we grew up in a western society. we are we are um exposed to western language. Yes. >> And so perhaps that has something to do with it. >> I that makes sense. >> No, they have done research with multiple language groups. Most famously the Tamil language group which is in India, South India. That is an entirely different language group from um English. English comes from the Indo-Uropean family. >> These guys are Tamils. They're not even Hindi cuz Hindi and you know Sanskrit still comes from the Indo-European

1:01:15family. The Tamil language comes from the Dravidian family which is native to India. It is completely separate from Indo-Uropean family language and they still believe that the left is boba and the right is Kiki. You go to other cultures entirely, Boba Kiki works. You go to infants, Boba Kiki works. So it was thought that this Bobaki effect meaning Boba goes to blobby rounder things and Kiki goes to poer pointier things is something that is innate within human language. Okay. Across human languages. And so far so good until this paper. This paper played recordings of humans saying either Boba or Kiki to naive 3-day old chicks and

1:01:58presented them with two panels that had food behind it. >> Okay. >> And the chicks knew which one was boba and shut. >> These are these are three-day old chicks >> that had the Boba Kiki effect. >> I don't believe it. No, it but it works. >> That's so meaning there is, you know, >> this is not a language thing. >> It's not a language thing. Yeah, >> this is this is like >> there is something inherent in the sound of Boba versus Kiki that maps onto the architecture of physical objects. >> Okay. And and you can you can start thinking like why? And actually, it doesn't seem that weird anymore when you

1:02:40start thinking about the physics of sound >> and the physics of boba and kiki. >> The word boba and the sound b. That's lower frequency. You associate that with bass. Yep. >> Right. >> Kiki is like the ting like the the high frequencies. >> Yeah. rounder objects when sound goes through them, the higher frequencies are going to sort of get absorbed >> around the the rounder frequencies are going to get deflected and so so on and so forth. The sharper an interface is on the physics of an object, right? >> The the sharper the sound that is going to come out of that object just because of the way sound is. Right? The sharper

1:03:21the thing is, the shorter the wavelength, the shorter the wavelength, the higher the frequency, the more we get that ding sound. >> Right? It's literally just for analysis is what what we would think if someone's a electrical or sound engineer. It's it's like the forier analysis of sound around the physics of objects. The smoother an object is, the bassier the sound that comes out. And so even chicks, their brains, three-day old chicks, their brains, there's some innate physics calculation that's happening within them that's saying the higher frequency has to come from a pointier object. >> That is so >> It's so weird.

1:04:01>> Unbelievable. But the way that you explained it does >> Yeah. Then it kind of makes sense. Makes sense. >> What what I'm what I'm surprised about is that a 3-day old chick has neural mechanisms within its brain >> to make that distinction. >> That's this is so invariant across um anyone that has brains, I guess, because this is not even a mammal. >> Right. >> Right. This thing diverged very very far away from us hundreds of millions year of years ago before the dinosaurs. >> There's a there's a deeper architecture here. Yeah. There's there's a deep not architecture, there's a deeper insight um about why that would be true. Yeah. Not only because of the cross species level uh similarities, but also that you

1:04:45don't need >> brains that have developed >> Yeah. >> for a long amount of time to have some sort of like, you know, learned >> understanding of the physics of the universe. >> Yeah. Yeah. They've just made this association because like over the years this is what's true, right? The B sound comes from rounder objects and the C sound comes from pointier objects, right? And it's like even like a a short a small neural network that we train with reinforcement learning, for example, would come up with the Boba Kiki effect, >> right? Where let's say the input of this neural network was the sound and then the output was a decision between the two. through reinforcement learning, it would learn that the lower frequencies are the rounder things

1:05:25>> that >> but the fact that like you know we did this going all the way back to our common ancestor between birds and mammals which is 100 million years plus before the dinosaurs >> and it's still true >> and it's Yeah. Yeah. And it's it's across all these animals, right? I think that's so cool >> that is >> to think about and also what a weird experiment to run. >> Yeah. >> Right. Like who who was like, "Hey, let's test Boba and Kiki on Baby Chicks." >> Who knew Drake was so attuned to Frontier Science Research when he made that song? >> Mhm. >> Whatever that song's called, Do You Love Me? Uh, four great rundown stories.

Story 2 begins — exercise reduces Alzheimer’s risk (mechanism)

1:06:04>> We're going to jump now back into our main story, our second story of the day, >> which is around Alzheimer's research. >> Yes, this one's very cool. So this one is a health and medicine story about Alzheimer's disease and exercise. Yeah. >> So scientists think they figured out exactly how exercise reduces the risk of Alzheimer's. It reports a new paper uh in cell by a group out of the University of California, San Francisco uh in this BAR aging inst uh research institute. And so we've we've actually done a lot I mean we just did last episode we did something on ALS. >> Um we've talked about Alzheimer's research. I think multiple times. Yeah, multiple times.

1:06:44>> Multiple times on the pod previously. >> Um this is clearly, you know, something that impacts millions of people, families, siblings, uh co-workers, etc. >> And so there's there okay there's a link between the disease and exercise. I'm curious what that link is. >> Yeah. And I mean it's not a it's not like that surprising that exercise is good for you, >> right? This is something we've known. But what I love about this study is we've again it's about the mechanism. It's the Lego blocks that are coming together. >> There's a pathway from exercise to the biochemistry of Alzheimer's that these guys have traced. >> Okay.

1:07:24>> From the from me running >> to me not getting Alzheimer's or getting Alzheimer's later. Okay. >> And that biochemical >> string of Lego blocks is what is super interesting to me. Okay. Okay. Very we potentially have a discovery of the pathway by which this is happening. >> Yes. Because if we have a discovery of the pathway the biochemistry of what this is happening then perhaps we can get the benefits without the exercise. >> Right. Right. Right. Right. Because we understand the me the mechanistic >> Yeah. >> dynamics >> that are driving the process. And and this could be huge because then for people who maybe can't do exercise but have Alzheimer's, this could be a big thing or for those who want to do

1:08:06exercise but get added benefit, you know, a boost. It's it's huge, right? Okay. So, we're going to start with the human brain. 80 billion neurons. >> It's the most complex object in the universe >> and stuff can go wrong because it is the most recent of our organs to evolve, right? Most of the mutations that create human beings have to do with brain development. And what that necessarily means is the brain is going to have a lot of stuff go wrong because it's still in the phase of the genetics is trying stuff out. >> We're in the beta version right now. >> Yeah, we're still in the beta version even though the beta version is doing really well. Like but but the no one's turned off the the code um like edits.

1:08:47>> Yeah. Yeah. It's like no one's frozen the main, >> you know, GitHub. People are people people are still merging branches on >> main. It's like it's like no no no we're doing like can we can we stop? Yeah. We got some bug fixing to do. >> Yeah. Yeah. Exactly. And so the bug fixing is what's crucial and it happens a lot later on in life. And for a long time the brain is sort of studied in isolation. There's like a you imagine like the brain is an island and we think about the mind and everything that's happening in the brain is isolated to the brain. But very recently that has

Liver-to-blood-to-brain pathway + “exerkines” framing

1:09:16progressed to be the brain, the mind and the body are connected and we need to figure out a holistic understanding of the entire system. Right? And that's where this research comes in. They've identified a biochemical pathway that links the liver to the blood to the brain and it shows the systemic benefits of exercise and how they are transformed or transferred to the brain without actually having to do exercise. That's the idea. >> Oh my goodness. Okay. >> And it's it's able to it's able to >> trace the entire story from exercise to brain benefits. Okay. The mechanism is from a particular enzyme that I'm only

1:09:57going to say once because these biochemists and organic chemistry people. They love just making the the world's longest word. I think that's what they're going for is like Guinness Book of World Records. Anyways, here we go. Okay guys, it's called glycosal phosidal specific phosphipase D1. We are going to refer to this enzyme throughout our story as GPLD1. So whenever you hear me say that word, that is the particular enzyme that we have discovered in the liver and all of its benefits all the way up to the brain. >> I just want to take a brief note to to

1:10:38to distinguish that although it has all of the same letters with a couple of extras. >> This is not ompic. Is that what you're saying? >> It is not a G, we're not talking about GLP1. This is GPLD1. >> Exactly. >> So those are different things. >> Those are very different things. GPLD1, this is not ompic. And what it does is act on blood vessels surrounding the brain to repair age related damage.

GPLD1 (not GLP-1!) and the BBB repair idea

1:11:01>> And that's what we're going to get into. >> So th this is like the this is like a cleanup crew that goes in to fix stuff. >> Mhm. But the way they fix stuff is very interesting. Okay. >> Okay. And that's where the you know, Lego blocks comes in and that's what I like about it. All right. So let's talk about Alzheimer's real quick. Um, most

Why amyloid isn’t the whole story; vascular hypothesis

1:11:18common cause of dementia, 60 to 70% of all cases of dementia are Alzheimer's. Globally, there's about 50 million people living with dementia right now in 2021. And if we look at a 2050 projection in the future, right, people are living longer. And if people are living longer, these age related diseases are going to go up. And so by 2050, there's projections of 100 to 150 million individuals globally. That's nearly triple of today living through these symptoms. >> Okay. Black Americans are two times more likely. Hispanic Americans are 1.5 times more likely. Women make up approximately

1:11:592/3 of the US Alzheimer's population. Yeah. >> And nearly twice the risk, lifetime risk versus men at the age of 65. So there's a lot of these disparities that go on. We still don't know much about why. I mean, you can imagine economic and social issues, diet issues with like black Americans and Hispanic Americans versus white people who just I mean, they're in a different economic bracket on average, right? So, these on average statistics comes through and it's it's a big like crisis in medicine to figure out what is going on and how do we help it. >> It was first identified in 1906 by by Dr. Alloys Alzheimer. He was a German doctor. um he identified these

1:12:40extracellular amyoid plaques in patients with dementia and since then we've been studying these things called amyoid plaques. The idea is you've got this accumulation of a protein called beta amyoid. Okay, there are these peptides and that's the primary cause. You get a bunch of amyoid precursor proteins. So this is whatever is the precursor to amyoid. They denature in some way and they get together, right? And when they get together, they create these plaques of amioid and those plaques then get in the way of all of the neural machinery.

1:13:20>> I get I get it's there's a highway and then these these roadblocks just >> Yeah. start autogenerating and blocking the flow of stuff. >> Yeah. And there was genetic link, right? because early onset familial Alzheimer's um was caused by mutations in this protein in the gene for this protein. >> Got it. >> So you can imagine okay that seems like something that we should target. >> There was a big discrepancy between targeting this pathway and actual clinical advantage. Okay. >> Okay. The drug failure rate the drug failure rate from 20 2002 to 2012 is something like 99.6% 6% and even today

1:14:00like this is a headline from last year. Nova Nordisk shares plunge after the Alzheimer's drug trial fails to hit the key target. >> Interesting. >> Okay. So there's so much research that is going into Alzheimer's, but when we try stuff in animal studies, when we try stuff in a petri dish, and then when we go in and try to fix the ammyoid crisis in the brain of patients, it doesn't quite work. There's only a weak correlation. It turns out the more of the research we do, there's only a weak correlation between the total plaque load in the brain and cognitive severity because the more and more that we've been able to do brain scans of healthy patients, >> we figured out that some of these healthy patients >> have a myoid plex. >> So that can't be the whole story.

1:14:42>> It's not it's not a single source. It's not the single source of the problem. Yeah, >> it it may be part of it. >> Yeah. But it can't be that oh a myoid plaque one direction causes Alzheimer's because there's plenty of people that have amyoid plaques that don't have Alzheimer's. Okay. >> Furthermore, there's a timing problem because the amyoid plaques, they start accumulating like 20 years before the actual symptoms. So by the time you've got the symptom and you're trying to treat this thing, it's like the forest fire is already starting. >> Yeah. Yeah. Okay. Yep. That tracks. And so there's been recent approval of certain um drugs, but it only shows a modestly slow decline in early stage patients, but it's not a cure. It

1:15:23doesn't reverse the effects of Alzheimer's right? >> And sometimes it carries risks like brain swelling and hemorrhage. So this amoid paradigm is really under pressure, right? And we need to find a new way to figure out what's going wrong in Alzheimer's brains and is there a way to reverse the effects? not just stop, not just slow down, but reverse. >> Because I mean, even just looking at this this data, the drug failure rate that you mentioned of 99.6% was across, you know, almost over 200 programs >> um in a in a 10-year time span. So, it's not like there if it's from a lack of trying. >> No, no, it's like we're trying, but like the fundamental issue we're having is maybe the philosophy with which we're

1:16:03going about >> right this. >> If if you're the Dallas Cowboys and you're like, why haven't we been to the playoffs in 30 years? It's like you're looking at everything but the source of the problem which might be you know anyway or if you're Ferrari. >> Yeah. >> Never mind. That's a that's a source for another day. >> Yeah. I don't want to get into that. We're doing well. It's going to be fine. Anyways, so we need to change the way that we're approaching this problem. Right. >> Okay. So now let's think about the brain from a mechanistic sense. >> Okay. >> The brain is only about 2% of the body mass. >> Mhm. >> But it uses up 20% of the body's energy.

Blood–brain barrier 101 (tight junctions)

1:16:38>> Okay. All of that glucose, all of that oxygen, 20% is going to our brain. Okay. It's got about 400 miles of capillaries in our brain that is supplying energy to the neurons. >> Mhm. >> Now, all of that energy has to get through from the blood into the neurons. >> Right. And it has to go through something called the blood brain barrier. >> Mh. This is an incredible plumbing system that we have in our brain. Okay? It's different from everywhere else in the body. Everywhere else in the body, normal blood vessels, they carry blood. Okay? And they've got these like

1:17:19gaps. >> Mhm. >> Where stuff can go in, stuff can go out from the blood. So, you know, hemoglobin can go through, oxygen can go through, glucose can go through, little tiny metabolites can go through from, let's say, our muscles to the blood back and forth. >> The the junctions in the piping are loose. >> Mhm. >> Literally, >> there's there's like gaps. >> Yeah. >> In the blood brain barrier, the junctions are what we call tight junctions. >> Okay. >> Okay. These brain endothelial cells, which are the the cells that line the blood vessels in our brain, have tight junctions and they're so tight that 99%

1:18:01of large molecules and 95% of small molecules do not make it through. Not even glucose can go through unescorted. >> There are literally things called transport proteins that have to take glucose from the blood >> through a special kind of passage into the brain. >> It's like a clean room. It it the the analogy I think of it's like it's like flying regular blood vessels are like flying back in the day where there's no TSA. >> Yeah, dude. >> And then and then but the blood brain barrier is like TSA. >> Yes, it is. It is exactly >> take your shoes off. >> Yeah. >> Got to scan your bags. >> Yeah. Yeah. >> Random search. >> Random search. Come to the back room. >> Yeah. Exactly. Yeah. Exactly. Exactly. The brain is like because because the

1:18:44brain is like arguably the most important thing for our humans. Like there's a lot of security >> between what's coming in and what's going out. >> Yep. Which makes which makes sense. >> Totally makes sense, right? >> Um and >> if these bloodb brain barriers fail, if these junctions fail, the tight junctions, >> if they become leaky like with normal blood vessels, then that's going to lead to non-specific leakiness. We're going to get inflammatory compounds that come in. We're going to get cytoines. That's going to drive neuroinflammation and that's going to exacerbate any Alzheimer's related pathology. I see. I see. >> So now there's a new hypothesis here. Not the amyoid hypothesis but the vascular hypothesis. This is the

1:19:26hypothesis that says systematic vascular disruption >> caused by aging caused by high blood pressure. That's going to loosen up the capillaries in our brain. And that's actually a primary driver for neuroderadation. >> Okay. >> Yeah. So the the the idea is the the we have two sort of sort of uh thought processes here. One is this a myoid plaque accumulation >> is the driver of this neurodeenerative stuff. Then the other idea is that the the degradation of the blood blame barrier on on the capillaries the 400

1:20:07miles of capillaries in your brain >> is like the inciting incident that then enables a whole slew of derivative stuff like that is >> it's a precursor to the amoid stuff. >> Got ah that right got yes right like it's not just that the amoid piece is not connected at all. No, but it's not the first >> it's not the first thing. And perhaps there's something upstream that we can tackle, right? That we can put a dam on. >> Okay. >> Okay. And usually this stuff is a hurdle, right? Because the blood brain barrier structurally prevents most drugs from reaching the brain. Right. >> Right. We've we've heard about that like, oh, this thing can cross the blood brain barrier and then like the valuation goes up by a lot because it can cross the blood brain barrier.

1:20:48Right. People are trying to do like nanoparticles. You might have heard of like nanoparticles that go through the bloodb brain barrier. You can also do like sonic disruption where you like like send like high frequency sound and that sound disrupts the bloodb brain barrier. But it's like well if we're trying to fix the bloodb brain barrier, we don't want to, you know, like it's like a >> catch 22 type type situation. And this particular paper is leveraging that hypothesis to actually treat the brain by targeting that blood brain barrier itself. You're not going to pass through it. We're trying to repair the the plumbing. Yes. There. >> Yes. >> Okay. Yes. >> Now let's get into um this particular paper in this particular lab at UCSF. So

1:21:28they first started out by doing this technique called heterocchronic parabiosis. The idea is you take uh old mouse and young mouse and you join up their vascule, their blood. >> Okay. >> And so it's kind of like that thing on Netflix where like you know the old guy is getting like the >> Yeah, it's what Peter Teal's doing, right? Yeah, like the young person's blood. Well, this is like the the genesis of that. Okay. >> Okay. They did this in mice first. And the old mice that were exposed to the young blood showed neurological rejuvenation. So, new neurons, better memory. It was actually reversing the effects of dementia.

1:22:08>> Very interesting. >> Okay. >> The introduction of new blood was actually basically having neurological positive benefits. >> Exactly. And so now you can start thinking, well, okay, maybe other interventions could make the blood young. One of the easy things is aerobic exercise right? >> Like cardio. >> It's a potent intervention. It delays cognitive decline. And maybe that cardio and exercise is doing something else to the blood that is causing that blood to rejuvenate. >> Okay, I just want to pause really quick because I I'm tracking. So the idea is >> this uh uh heterocchronic parabiosis is this idea of a younger >> rodent

1:22:49>> uh has better the blood is >> the blood is better >> it's better. >> So if you're >> and it's like reversing all this like >> stuff. >> Yeah. >> And so if you put that younger rodent blood in an older rodent it makes the older rodent better.

Parabiosis → “young blood” → exercise signals (exerkines)

1:23:00>> Mhm. >> Like that's the first kick. >> Yeah. That's the first and that we know that's true. >> That Yeah. You did the surgery, you make them share blood. Boom. >> It's done. Yeah. >> And so when we look at an activity like exercise, so the point is it's indicating that oh like the blood has an impact >> on neurological >> Yeah. >> maintenance >> and growth and degradation. >> Exactly. Yeah. >> And exercise impacts the blood. >> Yeah. >> The more you exercise, the more it keeps it sort of let's say youngish. >> Exactly. And actually the exercise impacts the blood in a very specific way because there's signaling molecules that are released by our organs during

1:23:41exercise. They're called exorces meaning like these signaling molecules. Exer meaning it's coming from exercise. And examples are like Iin which stimulates brain derived neurotropic factors. Neurons live longer that way. There's more synaptic plasticity. There's clusterin which reduces neuroinflammation. So all of our organs are releasing this. The muscles release stuff. The heart releases stuff, the kidneys, the liver releases stuff, even the bone releases stuff into our blood. >> And these are all signaling fl factors that facilitate like interorgan communication like if the brain wants to coordinate with the liver, the liver wants to coordinate with the kidneys and so on and so forth. And all of this is happening in our blood. And so the key >> insight here is that well

1:24:23>> how do we make the blood younger? It's because of these exores which are >> released during exercise. >> Mh. And what Sam what Saul Va's lab in UCSF did in 2020 is they reported in science that the exercise was actually bec the the effects of exercise were happening because of this liver derived enzyme GPLD1 whose name I'm not going to say >> but that is the that's the subject of this particular paper. In 2020, they discovered this enzyme. Th

2020 discovery: GPLD1 from liver improves cognition but can’t cross BBB

1:25:02>> this is Can we just hold on this for a quick second because I think this is important. So what you're saying is in this in this 2020 paper >> Mhm. >> in science >> we had this understanding that when you exercise all your body starts sending these extraines and >> they're sort of like bene like the benefit of that is it it helps sort of sustain uh it's like a almost like this maintenance process. Yeah. for the body where it keep it keeps you at stable state for lack of a better way to put it. >> We we know that it comes from exercise which drives a process that the body enacts >> the discovery that there is a singular protein that comes out of the liver.

1:25:43>> Yeah. >> In this in this 2020 paper >> that is improving cognition >> is is improving cognition out of all the extra kinds. >> Yeah. And other ones could be as well, right? But this one particularly found this that is like the signal is way high. of all the impacts, this one is the one that has the most and is a is a key for uh focus and attention. I just I just think that's a kind of a crazy idea in and of itself. >> Yeah. There's there's an enzyme coming out of the liver. >> Yeah. >> There is an enzyme coming out of the liver that is improving cognition demonstrably right >> in older mice when they do exercise. >> Right. >> Right. >> That's fascinating. Okay. Okay. >> Yeah. And that was in 2020. There's a

1:26:23problem though. >> Okay. This enzyme GPLD1 is massive. >> Okay. >> So, it's not making it through the bloodb brain barrier. >> Uhhuh. Okay. >> So, what exactly is it doing? And that's why there's a giant, you see the question mark, BBB, and then there's a question mark and it points to the hippocampus. Like, yeah. What is going on? How is this massive GLD1 enzyme >> that cannot cross the bloodb brain barrier? I mean, even glucose can't pass the bloodb brain barrier, let alone an enzyme of this size. So, how exactly is it making my smarter and their and their brains bigger? >> Because it's not going directly to the source apparently because it can't pass >> because it can't it can't. So, what is it exactly doing? And that is what this 2026 breakthrough is about. Okay. It's

1:27:05unpacking the mechanistic reason how that GLLD1 is affecting the brain. Okay. It's not actually going through the bloodb brain barrier. It's maintaining the bloodb brain barrier itself. >> Ahu. >> Okay. >> Okay. It's like it's like the gates

Mechanism: GPLD1 cleaves GPI anchors; TNAP as the key BBB target

1:27:22themselves, the TSA >> is what this thing is maintaining. >> Okay. I Yes. Okay. Very interesting. >> So, here's the mechanism. GLDD1 is something called a phospholipase. That's what the P the the uh D is there somehow. >> And what it's doing is its function is to cleave something called the GPI anchor. >> These are anchors that tether proteins to cell membranes. The cell has an outside. >> That's the cell membrane. And there's a bunch of proteins that are tethered to the cell membrane. You can imagine a boat on a dock, right? The boat is not actually attached to the to the dock. There's like rope that attaches a boat

1:28:02to the dock and a bunch of like those >> like sailor knots, right? >> Yes. >> That's a tethering protein. >> Okay. >> And what this GLD1 is doing is attacking that tethering protein. Okay. >> Now, why is that a good thing? >> Yeah. Like why? >> You would think that like if a protein is tethered to the to the cell membrane, you want to keep it there. >> Yes. >> Right. >> It's exactly what I think. >> That's that's what you should think. If there's too many proteins though on that cell membrane, what's going to happen to your dock? >> Uh the dock's going to break down. >> The dock's going to break down. If there's too many boats, that's why there's like you got to like pay for slips. You can't just like free-for-all dock to your to the dock, right? Yeah,

1:28:43>> if there's too many boats, the dock is going to float away and then and then it's just chaos. >> So, you've got to have a way to have slips and like untether boats that are causing the dock architecture to rupture. That is what GLD1 is doing. Okay, they they >> they got a target. >> This target must be on the bloodb brain interface. And then they used actually transcrytoics to figure out that there's this specific protein called TNAP tissue non-specific alkaline phosphotase. This is a boat that docks to the membrane. Yeah. >> Okay. >> And if and this thing on its own has its own like job, right? >> Okay. This particular T-nap protein, it

1:29:24involves promoting bone mineralization. And in the brain, it does a variety of other stuff. Okay. But in the brain, if it's anchored to the to the cell membrane, if there's too many of that anchoring, then the cell membrane of your blood brain barrier is going to start breaking down and then all of a sudden your tight junctions start becoming loose. >> That was the hypothesis. Okay. Okay. Okay. >> And the key action was that if you circulate this GPLD1 drug, which is something that cuts, it's a scissor for the ropes that tether the boat. >> Then what that's going to do is act as a molecular scissor. It's going to cut that anchor. The TNAP is going to go off

1:30:05>> and you're going to maintain the blood brain barrier structure >> and that is what is key. So to recap, >> yes, >> you exercise. >> Yes. >> The liver produces this GLD1 enzyme. Yes, >> that GLLDD1 enzyme that goes into the brain. >> Yes. >> And it maintains the bloodb brain barrier by cutting >> the anchor between this T-nap gene and the >> cells that maintain that bloodb brain barrier >> because the the T-nap the T-nap at too much volume begins to degrade the actual structure of these tight joints. Yes. in the bloodb brain barrier and they're not

1:30:46currently for whatever reason self-regulated. There are no slips. >> Yes. Yeah. I mean, yeah, that the the self-regulation is the GLLD1, >> right? Right. Right. Which >> that's what the front the function, >> right? >> GLD1 is kind of like the boat uh I guess the dock like >> Coast Guard. >> Yeah. The Coast Guard. It's like you don't have a >> Right. But the currently, at least in the context of this study, >> Yeah. The only source uh for the GL GPL D1 >> is exercise >> is the liver during exercise. >> That is in the context of what we're looking at. That is the only way we can generate enough coast guard people to patrol the docks to maintain the

1:31:27structural integrity of the blood being barrier which can then be a driving force for the uh the um >> uh the plaques we talked about earlier. >> Yeah. Because now when when when you know random stuff gets in that maybe starts denaturing the plaque proteins which causes amoid plaques which causes >> yes >> Alzheimer's okay this is >> it's like from the top down and I I think it's worth it's worth really like lingering on this for a bit right we've created a molecular cascading mechanism >> from exercise to preventing Alzheimer's >> right there's now a link between the two >> yes >> and it it's literally exercise the liver produces a specific

1:32:10exine which is this GLD1 that then goes to maintain the bloodb brain barrier once the bloodb brain barrier is maintained >> Alzheimer's gets reduced because maybe that is the key issue >> right it's the the key issues potentially there's no TSA preventing whoever wants to come into the airport getting into the airport >> exactly >> um and then what we're looking at when we had the amyoid hypothesis was a specific specific type of person that was getting through at the airport but not >> and like causing problems inside the terminal, >> right? >> But that was it's potentially just a single use case versus understanding the systemic structure. >> Exactly. Yeah. And let's just go through some of the experiments that they did.

1:32:51So they did alkaline phosphate labeling of TAP. Tap is this thing that the the boat that is docking too often. Yes. To the bloodb brain barrier. And you can see on the left there is old animals. There's lots of TAP >> along the blood vessels. And on the right is an old animal that does exercise, >> the amount of TENAP is drastically reduced. >> That's incredible. >> So there you can see just very much in the staining of this particular enzyme that we're getting a lot. They also did novel object recognition and Yaze, which are these behavioral tests for old mice, okay? Like if if the old mice have dementia, they can't do well on these where's the cheese type of task. But you could actually recover and reverse

1:33:34cognitive decline >> by treating old mice with GLD1. You could treat old mice with GLLD1 without having them do exercise. And then there was a profound reversal in the deficits in this particular behavioral maze, right? So they could find the cheese faster and all that other kind of stuff and remember where the cheese was. And just to be clear, what we're what you're saying here is um you can artificially

Cognitive tests improve with GPLD1 treatment (maze/recognition)

1:34:00introduce GPLD1 >> into these aged mice who already have an accumulation of tenap happening. >> Yeah. >> And it still works when it's not necessarily coming from exercise >> because it's the same enzyme. >> Yeah. >> So it doesn't necessarily it doesn't have to come from exercise. >> Yeah. It's like this this singular benefit from exercise we can replicate by just injecting. >> Right. and then we're good to go. >> And we're seeing that even when the you don't have to do it preemptively or preventatively. No, >> you can do it even. >> This is already an old mice that has Alzheimer's like symptoms. It's bad at the Y maze, >> right? And you you artificially insert it and it's got it gets better. >> It gets better. It finds the cheese,

1:34:41>> right? So, this is huge. And for blood um blood brain barrier repair, right? So, let's look at the difference between how much leakage is happening >> when you've got an aged mice >> that has leakage versus one where we treat it with GPLD1. On the left, you can see the the pink is all of where the leakage happened. Okay? It's all over the hippocampus. This is a slice of the hippocampus. On the right, most of the pink is on that bloodb brain barrier in the corner. there's no huge spots in the middle of the neurons because there's not that much leakage because you've treated them with GLD1, right? So,

1:35:22>> I think this is this is really really cool. >> This is this I mean so it's cool for a couple of reasons. It's again understanding, you know, it's like think what you talked about like connecting the biochemistry processes of the body >> to these neurological like long-term >> symptoms >> symptoms >> and being able to trace that story

BBB leakage decreases with treatment (repair the “plumbing”)

1:35:45uh which again we only talked about one of the exorind which is this GPL there could be more >> there could be more so again if there's a fundamental >> that's interesting thing independent of the specifics of them looking at Alzheimer's direct like as as their initial use. So and then the then how they then um you know ex tried to test this in mice. >> Yeah. and saw the >> they did all the right things, right? And when we compared it to the amoid uh experimentation and why it didn't work, it like it it clearly works or it doesn't.

1:36:26>> You know what I mean? Exactly. Like like you have a myoid plaques in people and they're fine, >> but when you introduce this GLPD1 into it, it it's not like they just continue being ineffective at the Y maze. >> Exactly. Yeah. And and you know, one could say, oh, this is um this is in mice. There's obviously going to be issues right? >> Always. >> Always. >> That's what's cool about this. >> Okay, that's what's cool about this. So, TAP like Okay, so >> how did they implement GLD1 in mice? >> Yes. >> They've got this procedure where what you effectively do is you increase you you put the gene for GLLD1 inside a viral vector or like in some kind of plasmid chunk of DNA. You stick that in

1:37:06a mice. Mhm. >> Um, you increase the blood pressure such that the liver gets infused with this piece of DNA and then the liver makes >> the GLD1. Can't do that in humans. >> Okay. But here's what's cool. >> The TAP >> is already something that we've been worried about. >> Okay. >> Okay. That TAP gene, the boat that gets anchored, that's already something that we've worried about. And there is a small molecule drug called SBI425 >> that is already approved. >> Okay. to inhibit tenap. >> Okay. >> The boat. >> Okay. Yeah. >> It's already getting in the way of the boat. And it's successfully been used in pre-clinical studies to prevent blood

1:37:46vessel calcification because remember TAP has to do with calcification and bone meation. So like sometimes like there's calcification like calcium deposits that happen in blood. And so this particular drug was developed >> in order to circumn that. But it's like, well, this thing doesn't have to cross the blood brain barrier because what it could do is go into the blood and inhibit the boats from anchoring.

Translational path: TNAP inhibitor SBI-425 mimics effect orally in mice

1:38:12>> Mhm. >> Right now, this is a we're we're taking our understanding from the study. Yes. That it's about boats anchoring to the dock and too many boats anchoring to the dock that is then >> creating the physics of the dock to not be tight and be loose and now there's leakage. Well, what if we just like messed with the boats, >> right? The team, >> right? >> And that's what they did. >> Okay. >> And the oral SBI 425, it successfully mimicked this GLD1 gene therapy. >> Oh, in the mice. >> Wait, wait, wait, wait, wait, wait, wait. This is actually probably the craziest piece of this. So, so in the mice, we had to do gene therapy in order to to instigate.

1:38:54>> We have to do this like weird stuff that we can't do in humans. But now it's like what if we just feed the mice this this drug >> right which is already >> approved >> approved for a different use case. >> Yeah. Yeah. Some some random you know calcification. >> But the point is it has the same it it creates the same exact GL GPLD1. I keep saying GLP1 GLD1 >> uh uh generation enzyme generation. >> Yeah. >> Just take and it's already an oral format. >> Yeah. It's already in oral format and it's mimicking the gene therapy. Right. Maybe it's not doing what the gene therapy exactly is doing, but effectively what do we want? We want to maintain the bloodb brain barrier. That

1:39:35is our end case. >> Yes. >> Right. Yes. >> And by whatever means necessary. Well, we already have >> SBI 425 that's been approved for other stuff. >> Yes. >> And it's doing it >> right. Right. >> So now this is now a bioavailable pill. >> Yes. >> That can treat Alzheimer's. >> Yes. purely by targeting the bloodb brain barrier, by targeting this vascular interface between the blood and the brain. Maintaining that, >> it circumnavigates this historic bloodb brain barrier drug delivery problem. We don't have to worry about does is it going to get through? Because what we're worried about is actually maintaining it in the first place. >> Yes. We want to make sure that in the analogy we brought up about TSA in the airport um or the boat and the dock in

1:40:16the Coast Guard, uh we want to make sure that the Coast Guard is always patrolling the dock. >> Exactly. Because what's happening is Coast Guard over time people retire and then there's not enough we staffing issues >> and and then all of a sudden these teen apps are just having a field day kind of like the Chinese fisherman people off the coast of Chile and all over the place that are just totally >> exactly >> unregulated. But if you had some Coast Guard there and then there's no less TAP and then >> and the final thing I want to bring up about this paper is they did do their homework with actual human tissue cuz so far it's always in been in um >> in mice in in particular mice.

Human tissue evidence: TNAP ↑ in Alzheimer’s; GPLD1 ↑ in active elders

1:40:54So in humans they looked at autopsied cortical tissue of older adults with Alzheimer's dementia >> the TAP protein levels are elevated there >> which means there's a lot of these TAP boats that are anchored to the dock >> of the blood brain barrier >> the GLD1 levels were elevated in active healthy elderly humans. So those who are active and healthy >> the Coast Guard was up. So already we're seeing these like large scale correlations, right? And it's hope that this could actually work perhaps in humans. I I thought it was a very cool study because they they did a comprehensive research in mice. Yes. And

1:41:35they and they didn't leave the human the human part >> you know >> because the point is the insights that were derived from the studies in mice are uh in terms of uh autopsy data from past Alzheimer's patients both those who were active and inactive >> basically mimic the levels >> both in this uh the t-ap inhibitors being more present >> uh on on the blood as well as um exercise increasing these this enzyme level. >> Yeah, this one that helps to maintain the structure of the blood vessels and the tight junctions of the blood beam barrier which again seems to be as a

1:42:16competing theory to the amyoid hypothesis a potential uh it's like a lower it's like lower in the um stack >> uh of processes because you want to basically attack a problem as close to the source as possible. >> Exactly. um the amiioid plaque is not the source. >> Yeah, it seems it's not. >> It seems it seems it's not, but this might be >> getting closer >> uh to what it's going on. >> And also just an interesting conceptual way to think about how to look at like neuro uh degenerate neurodeenerative diseases. >> Exactly. As a holistic thing. As a holistic, right? It's like this thing involves the liver, the blood, and the

1:42:57brain. It's the brain is no longer like an island that you have to think about. Right. I mean it's it's I think the future is looking very bright for this particular study. Of course I don't want to do commentators curse but the GPLD1 which is like you know your Coast Guard that thing targets all kinds of boats not just TAP. It's targeting all sorts of GPI anchored proteins >> hundreds of others. So there those could be targets for new therapies, right? Um you could have a antip I mean I anticipate there's going to be a surge in peripheral TENAP inhibitors, not just this >> SBI425. There could be other stuff, >> right? >> So yeah, and I think that's an important point that the the SBI 425 already being

1:43:40an approved drug like is huge because it sort of shows the the drug development >> scaffolding. Yeah. to get to an end result even if you don't use exactly like you you can at least reach like see what the pathway is to success here around how to approach it because you know there's there's the research but then the whole drug discovery process from stage one uh stage one or phase one to phase four quite complex you have early stage drug discovery where they're just trying to figure out like the active ingredients then you give it to the chemists because they have to figure out how do you actually deliver it right >> to the patient is it introvenous is it oral dosing what is the halfife life, all that kind of stuff. Then you get into sort of early pre-clinical trials.

1:44:22>> Yeah. Where it's like, okay, let's make sure it doesn't kill the >> bug, right? Then you get into clinical trials, FDA approval. I mean, it's it is a very complex and long thing >> and this is already halfway through. >> It's already So that's a huge deal. >> Uh fascinating story. >> Yeah, I thought that was very cool actually. >> It's very cool. And I think what's been so interesting in this journey on the show for me, right, having the benefit of just having you able to really again having a first principles general science understanding is not super common unfortunately. Um, and the ability for now like me almost 30 episodes in to be able to make these connections across stories because there's fundamentals about the process

1:45:05of this early stage research or frontier research that track. Yeah. I mean, we talked earlier about the proton piece and how that's like you how measurement not of direct observation, but indirect observation is a concept and it can manifest Yeah. >> at the when you're zooming in really really small or when you're zooming out really really far. I mean, this is a really really cool story and great work. This again was from um UCSF. >> Yeah, UCSF. They do amazing work there. >> Lots of Nobel prizes. >> Oh my god. Fascinating, fascinating story. And again I just want to make a quick sort of caveat here. The point of this show is to talk about breakthrough in frontier research from first

1:45:45principles. There is a large delta or a big gap in between a research paper coming out and it having >> you know over-the-counter or prescription medication available to people. And so just to be very clear about that point >> we very much understand that one there needs to be replication. There's a whole process of implementation. >> There's a lot of stuff that can go wrong in the middle. >> And so the middle is there's a lot there. >> It is it is about kind of the beginning of the story, but it's important just to be clear that that that is that is the case. >> Yeah. >> Um Yeah. >> And so, you know, I know some folks, yeah, this is still early. Yes. >> Yeah. >> But isn't it cool that we figured out

1:46:26this Coast Guard boat docking thing? >> It's the concepts and the ideas are really and understanding how we got there. Yeah, >> is the point. >> Yeah, that's the point. I mean, >> um because there's so many things you can get from this. So, in any event, um >> fantastic. We started with dark galaxies again. We space and then neuroscience, some great rundown stories. I want to do a couple of quick housekeeping notes before we wrap up for the day. Number one is if you are listening to this episode and you've made it all this way, >> congratulations because some of this

Wrap-up + how to support + Dave Chang/Netflix tease

1:47:02stuff is really hard to gro when you're just only listening. And if you are curious and want to know more, we have a video podcast. So you can watch this on YouTube, on Spotify, and soon coming soon because we just got the announcement email, Apple Podcasts is introducing videos. >> Oh yeah. >> Uh later this spring. So if you are an Apple podcast listener, you are on the precipice of receiving video as a first class object just like we have on YouTube. Spotify, it's not our fault. That's just they haven't implemented it yet. So that's exciting. We do a ton of graphics and visuals that help really understand what's going on here. So please take a watch. If you are

1:47:43listening on Spotify, Apple, a fivestar helps us significantly uh to get this out to more people >> and some comments. and some comments, too. So many of you talk about how this is your favorite podcast. Go ahead and comment it on every show that you feel that way. Share it with a friend. Bring it to lunch at your journal club if you're at a lab. Um, it's really helpful for us. We make this show freely available. There's no subscription content. There's no exclusive back-end content. It's all the episodes and all the clips are available for everybody. And so if you want to support us directly and become a patron, you can go to ffpod.com/donate. You can do a one-time donation or you

1:48:25can do a monthly donation. That's what helps us do all the editing, get all of the equipment, uh do travel for some of our special stories. Uh maybe for those who have reached the end of the episode, maybe I'll do a little tease here that you know, you only will know, well, you might have seen it on our Instagram story but >> we have uh recently had the uh wonderful experience of getting to be interviewed on the Dave Chang show with David Chang, the famous chef and owner of the Mumaf Fuku restaurant group. Uh, being able to have food made for you by the one of the goats is uh, >> yeah, it wasn't like, oh, we got takeout

1:49:06from Momaf Fuku. This is like Dave Chang is making me >> I can't say actually see what he does. You have to have the last episode. >> But that should be coming out on Netflix, >> our first our first Netflix placement as well as Spotify um, >> by a week, probably a week from when this episode comes out or a week or 10 days. And so keep an eye on our socials at FFP Pod for uh when that comes out. It's a great interview. You get to see us in the real world outside of the studio. We had a great time. Big shout out to Dave's team. Um David Meyer, uh Christian, you guys are fantastic. We had a great time and we look forward to you inviting Krishna when you guys need

1:49:48the science of food or food science expertise. >> I will do anything for food. >> Friend of the pod. Shout out to Dave Chang and and co. Really appreciate that. My name is Lester Nar join. >> Hold on. >> Yes. Uh the comments

Comment prompt: best alternate meaning for “GPLD1”

1:50:04>> I want you guys to give me an alternate >> um GLD1. >> Oh, >> what does it stand for? >> Yes. >> Because the one that we have, which you know, just for all time sake, I'm going to try glycosal phosphotidylonol specific phospholipase D1. I would like something better. Thank you. >> That was a great call. Yes. So again, listening all the way. You'll get a little couple secret nuggets in here. Send over the comment. Keep an eye out for us on Dave Chang. And we may or may not start to do some inpod giveaways. And so, you know, you might want to listen cuz you might be maybe someone

1:50:44who gets some special merch or who knows. >> Who knows? >> Who knows? >> Bringing it back. I'm Lester Nari joined as always by my co-host and our resident PhD Krishna Chowdery. We really appreciate you guys so much especially those who listen all the way to us ranting for two hours. We tried to do two stories this week to make it less. >> Yeah. And it's still two hours. >> It's still two hours. >> We got to work on that. >> We will be back next week with our next episode. This is from First Principles.

1:51:23Hey,