This moment, Alex's mile swim, it means everything to him and to all of us.
At Evernorth Specialty Services, we help providers make sure patient care doesn't get interrupted, especially for rare, chronic and complex conditions.
Now Alex and his care team have his hemophilia under control with life-saving medication and care management from Evernorth's specialty pharmacy Acredo.
So Alex can be the last one out of the pool.
Evernorth Specialty Services, because every moment counts.
Visit evernorth.com slash specialty to learn more.
What's up?
Adam Grant from Work Life, a TED podcast here, and I want to tell you about something exciting.
You're watching Every Dollar, and Walmart Business helps you stretch each one.
From office supplies to snacks and cleaning gear, you'll get everyday low prices, plus easy bulk ordering and fast delivery.
And with tools like spend tracking and multi-user accounts, staying organized is simple.
Save time, money, and hassle at business.walmart.com.
It's free to sign up.
I find this not only refreshing but at some level astounding.
Nature.
Welcome back to The Nature Podcast.
This time, a soft implant that grows with the brain.
And restoring damaged artwork with the help of AI.
I'm Benjamin Thompson.
And I'm Nick Petrichow.
Researchers have made a soft brain implant that can be installed into an amphibian embryo and grow with its developing brain, monitoring its neural activity.
It leverages this unique morphogenesis process of the brain development and enables the stable recording and stimulation of the neural activity process over the entire time course of the development.
That's Jie Liu, one of the team behind this amphibian implant.
He believes this is quite the achievement, in part because making any sort of brain implant has historically been quite hard due to the brain being quite soft.
So you know, the brain is a very soft, like a piece of the tofu, you know, and the traditional electronics is very rigid.
Then you put them into the.
Any movement of the brain can cause electronics to cut the brain at a micrometer scale.
It will also trigger this kind of immune response from the brain tissue that, over the time, it causes degradation of the neurons at the implantation site.
It also causes the growth of the immune cells that is surrounding the implantable probe.
So that over the time they will form this kind of a glial shell surrounding the probe that isolates the neurons from the implantable probe, that you will quickly lose the single-cell activity after you implant the probe into the brain for just a couple of weeks.
The tiny damage caused by a hard implant as the brain moves can cause inflammation, which then leads to immune responses, eventually making the implant ineffective as it's surrounded by glia cells.
In fact, when people have brain implants to treat things like Parkinson's disease, the implants need to be moved every few weeks to avoid these issues not exactly ideal.
So Jia has been investigating soft implants that could move with the brain, and in previous work he's shown that this prevents inflammation.
And we found that once they become extremely flexible, this kind of mechanical mismatch will be reduced.
There's no this kind of drift of the probe inside the brain when the brain is moving.
And also there's no immune response, or at least like a very minimal immune response, from the brain to the implantable probe.
This does still cause a small amount of damage when it's implanted.
It's a bit unavoidable when you put something into a mature brain.
Also, researchers are keen to know more about the developing brain, how it grows from a handful of connections between neurons to the billions and then trillions that humans have as adults.
The developing brain though is just that, developing.
It starts off as a flat sheet of cells, before folding into a tube and then eventually becoming the organ we all know and love.
This dynamism would mean that any would-be brain implant would have to stretch and shift with the brain as it grows, a problem that Jia has been thinking about for a long time.
I remember, since I was in high school, I read a report that said that actually the 3D brain is coming from a 2D stem cell sheet named the neural plate on the embryo.
And this neural plate will force the food into a neural tube and, with further expansion and folding, fold into a 3D brain structure.
So I'm thinking that instead of just make a device flexible, if we make them highly stretchable and miniaturized, then we can integrate those device into the neural plate at the early development stage.
By exploiting the 2D nature of the early brain, Gia figured he could lay a small stretchy device onto this sheet of cells, which would then get wrapped up and distributed throughout the entire brain as it develops.
Some years, a PhD and a lot of work on tiny bioelectronics later, and Gia was ready to try out this idea.
He and his team landed on a stretchable mesh less than a micrometer thick, embedded with even tinier electrodes that could be used to measure any neural activity.
They first tried it out in brain organoids, tiny structures akin to brains made from stem cells.
As the organoids grew from simple cells to a brain-like structure, the mesh stretched and flexed and the electrodes ended up being distributed throughout the brain.
With that shown, the team moved on to a living animal.
They chose the clawed frog, Xenopus laevis, partly as it's a well-studied model of development, but also for some other reasons.
The embryo development is in the solution, so that we can easily manipulate that.
If we do anything else, how to maintain the viability of those embryos is a challenge.
But for frogs, it's pretty easy.
Now, of course, there were challenges.
The team needed to maintain the temperature carefully to ensure that the frogs developed at a steady rate.
And putting the implantable mesh onto the tiny frog embryo brains required a special set of skills.
The first author, Hao Sheng, is a student with a very steady hand.
You know it's a very unique skill capability that I cannot do it, but he's really good at this.
Like can align those very flexible device with the embryo neural plate, since the sensor part is already embedded into the frog.
So when the frog is grow, the device will grow together with the frog, The mesh integrated with the brain becoming enmeshed across multiple brain regions.
And it didn't appear to disrupt development.
The implanted tadpoles, that the team dubbed cyborg tadpoles, also seemed to have no ill effects.
So we did so many different tests.
Actually, it's a very interesting experiment that the frog can sense the color of the background.
So you have this background switch between white and black.
It always goes to the white area.
And if the brain has been damaged, then this kind of capability will be unparalleled.
And then we found that it doesn't change it at all.
They also ran some other developmental tests and used genetic sequencing to see if there were any traces of stress.
The cyborg tadpoles seemed normal.
So they then looked to see what they could discover by having this brain implant.
By using this technology, we found that we can enable the stable recording of the neural activity in the embryo brain.
And we found that it's very interesting that when the neural plate first falls into the neural tube, they have this global synchronized activity.
It looks like this kind of neural plate needs to have a synchronized activity to guide the development.
And then gradually they decoupled.
So, whilst the early brain had very synchronous activity, over time the different brain regions developed their own unique activity.
Also, the speed at which neurons fired changed.
At first, it was very slow, but as development progressed, it got much quicker.
This matches what has been shown in previous studies of the frog with conventional hard implants, but the new soft mesh allowed Jia and the team to be less invasive as they studied it.
The team were then curious about what else their soft brain implant could reveal, so they looked to an animal known as an axolotl.
This South American amphibian also develops in water, allowing the team to relatively easily insert their device.
They also have an ability to regenerate parts of their body.
What happens in the brain during such regeneration was something that Jia was interested in.
Once they grow into a tadpole, then we cut its tail and ask it to regenerate.
We found that the brain activity goes back to its early development stage.
So this is maybe a very unique phenomenon reason why this creature have this regeneration capability and the brain seems to participate a very important role in this regeneration.
So that we want to test this idea.
We have another like this kind of exotical tadpole.
We cut its tail and we stimulate the brain.
We make the brain even more synchronized together.
We found that this regeneration speed goes up.
It can accelerate the regeneration capability.
This is a routine procedure to understand regeneration in axolotls.
A small part of the tail is cut before growing back and they don't show signs of pain or distress from this amputation.
Flavia Vitale, a neuroengineer who wasn't associated with the new study, thinks that there's a lot to discover about regeneration like this, and this new device could allow researchers to do just that.
Now we have a potential opportunity and a tool to start investigating those mechanisms, and could be true for regeneration, but maybe also for plasticity and, who knows, even ageing.
In general, Flavia was excited about the potential applications of the new device and was impressed with the meticulousness of the study.
It is a really rigorous journey from fundamental engineering principles to engineering new materials that can grow and deform in contact with biological tissues and then, from These initial investigations using fundamental engineering principles, the idea of taking these devices and making them grow inside the body and follow the natural brain development and embryo development.
I thought it was extremely fascinating.
Ultimately, Gia hopes to use devices like this one to help people with neurological conditions.
But before we get to their use in humans, Flavia thinks there are a lot of ethical implications that need to be considered.
Are we going to perturb or interfere with their development or are we going to guide their development in states like that were not healthy or were not intended right?
And also the ethical implications of manipulating these embryos while they're still forming.
I think we have to very seriously consider Devices like this.
One could help understand the brain as it develops.
Frogs and axolotls can already give us some insights into how this complicated process progresses.
And Jia is working on translating this technology to be used in mammals and maybe eventually in humans, after a lot more testing and ethical considerations.
So that is even like a more challenging task that we need to address.
But with the technology reporting, this paper is a very good start that we have this foundation stretchable electronics that could be directly translated to the neonatal or developing brain.
That was Jia Liu from Harvard University in the US.
You also heard from Flavio Vitale from the University of Pennsylvania, also in the US.
For more on cyborg amphibians, check out the show notes for a link to the paper.
Coming up a new method that could drastically reduce the time it takes to restore damaged paintings.
Right now though, it's time for the Research Highlights with Dan Fox.
One of the largest known exoplanets is evaporating into space and will break apart in the next half a billion years.
HAT-P67b is a planet a little more than twice the size of Jupiter, but with a much smaller mass.
That makes it one of the lowest density planets yet found.
And astronomers have wondered exactly how it has managed to stay intact for so long without being evaporated by the radiation of its host star.
And so researchers used the WIYN telescope in Arizona, combined with data from other observatories, to calculate the planet's mass, which they estimate is less than half that of Jupiter.
This estimate helped the scientists understand more about the planet's past and future evolution.
They found that the host star only recently became active enough to start irradiating the planet, explaining why the planet is still relatively intact.
But they calculated that HAT-P67b will evaporate away within the next 500 million years.
Read that research before then in the Astronomical Journal.
Cockatoos in Sydney, Australia have pioneered an innovative way to stay hydrated.
In 2018, researchers noticed a cockatoo in a recreation area in Sydney using its body weight and feet to turn on a twist-handle drinking fountain and then gulping from the resulting stream of water.
To see whether the behaviour had spread, researchers filmed a drinking fountain in the area for 44 days in the following year.
The team estimated that around 70 of all cockatoos in the flock tried to work the drinking fountain during the study period.
But of the 525 recorded attempts, only 41 were successful, suggesting that the technique of balancing using your weight and drinking is tricky for cockatoos to master.
Drink in that research in Biology Letters.
Next up on the show, a new way to help restore old paintings with the help of AI.
Now tucked away in the basements of museums around the world are a huge number of donated paintings that will never be replaced.
A major reason for this is that these artworks have suffered decades, perhaps centuries, of damage caused by things like exposure to light pollution or fluctuating temperatures.
These can cause paintings to crack or flake, and restoring them to their former glory is often a hugely time-consuming process for conservators.
But this week A Paper in Nature presents a new technique that could speed up a particularly labour-intensive part of restoration, known as in-painting, literally the painting in of damaged areas.
But this method doesn't rely on paint taking a damaged painting, using AI to help work out what the damage or missing areas might have looked like, then printing these areas onto a removable transparent mask that gets placed over the artwork, fixing it up.
This method was developed by Alex Kashkin from MIT in the US, a mechanical engineer who also enjoys restoring paintings.
Alex tested their system on a small painting attributed to an anonymous artist known as the Master of the Prado Adoration.
I wanted to find out how it went, so I gave Alex a call, who first explained a bit more about how long conventional painting restoration can take.
It depends on the amount of damage.
Like the longest it's taken me for any given work is closer to a year.
And of that, it's a couple months to get it ready.
And then the rest of that is just each of the tiny losses that are present in painting them by hand.
But there's some noted cases where restorations take over a decade, sometimes over 50 years, because of the amount of damage that's present and the intricacy of that damage.
Conservation is a science, but the in-painting portion is artistry.
And one thing that is becoming more and more used, as I understand, in repairing paintings is maybe getting AI to help conservators assess damage, things like that.
Yeah, there's lots of very interesting ways it's being applied.
The visual reconstruction of damaged art has been something that's been going on for decades.
But increasingly in recent years, different models have been used to predict damage.
Things like deterioration in different artworks.
They've been used to classify them stylistically and understand some various aspects about them, like the texture profile or what kind of brushstrokes are used.
And all of that information can really assist conservators in doing their job.
And so here we are.
Then we've got this situation where AI is used increasingly so in painting, restoration.
You are someone who enjoys restoring paintings and you're a mechanical engineer.
And so you've kind of put these three things together in your palette, if you'll allow me.
Tell me about your thought process behind this paper.
Basically, the background story to it is I have this assortment of skills and knowledge that most conservators do not.
A lot of the techniques that we have in mechanical engineering.
They can be very applicable to art, but no one had really ever applied them.
So, in this paper, essentially what you've come up with then, is a way to lay a mostly transparent film with some areas of the painting printed on it.
And, when overlaid, this fills in the holes seen in the painting underneath, so that the painting looks repaired.
Maybe you can tell me about the painting you have restored.
It's from the Netherlands in the 1500s, right?
Yeah, I wanted a painting that was so heavily damaged that conventional conservation.
No one would really do it because of the complexity of doing it.
And in this case this is a very late 15th century, possibly even past the 1500s work of an oil on panel painting that I attribute to the master of the Prado adoration.
And it's a few figures in a barn, basically an infant lying on the ground and the Virgin Mary in a blue robe looking over.
And it has a lot of damages to intricate features.
It has damages to complex color gradients and it has a very large number of damages, many tens of thousands.
And And.
In this study I only correct.
I think it was around 5000 of them, but that's enough to get it back to a state where you look at it and to a viewer it seems very reasonable and a lot of things have been restored and you can interpret the work directly, as opposed to before, when there's things like an entire human face missing.
And maybe you can lay out the steps you took to restore this painting then.
So what I've done is I've taken the painting, scanned it, used various AI techniques to construct a restored version of it.
And the way these work is they basically identify damage in a certain area and then, based on the context, extend the context to fill in whatever loss is present.
And that's pretty simple to do to derive a fully restored visual version of a painting.
And then the part I find very interesting but is very nuanced and gritty, is then deriving a mask that infills the damages.
Basically, what are the areas that we actually want to cover up in the original?
And that includes most of the damages, but that's subject to a number of constraints that are tied with the physical realities of doing so.
But once we've found this infill mask, that infill mask is fabricated in full color on this multilayer transparent laminate.
And it's a very thin laminate.
The transparent regions are just 30 microns thick.
That's way thinner than a human hair.
And then that laminate is applied to the painting and that conceals the damages in a way that doesn't optically interfere with viewing the surviving parts of the work but does fix up any areas that really distract from the viewing experience.
And so the printer prints out printer ink rather than oil paint, it's just printer inks, and fortunately, over decades in industry, color reproduction has reached the stage where we can do so with many pieces of conventional equipment really achieve color reproduction to the degree that's necessary to restore paintings.
Tell me, how long does this take from start to finish, compared to how long it might take if you were to do it in a conventional method?
Maybe Yeah, so the part I think that is most meaningful to conservators is basically how much time is needed in a lab space to apply the mask.
And in this case, that was only around three and a half hours.
And relative to conventional inpainting.
You know the activity that I enjoy doing.
I came up with the rough metric for it that in this case this painting would have taken 260-ish hours to inpaint all of the damages to the degree that I did.
So once you've applied your mask to the painting, what does the painting look like?
Because you talk about in your paper how it's not a perfect fix.
No, it's definitely not perfect.
And that's reasonably fine.
If you look closely, you can pretty easily see that there's a textural difference.
Printers don't really replicate the texture or the visual look of oil paint that well.
Although, if you go to a museum and you look at a painting, often if you pay close attention you'll see, oh wow, a lot of this is inpainted.
And generally in conservation, we value that inpainted areas are easy to distinguish because that helps the viewer see what was there originally, what did the artist paint, and then also to appreciate that oh, the conservator tried to make this a more cohesive visual experience.
But they're not trying to basically pretend like what they did is equivalent to what the artist did originally.
And speaking of what the artist did originally, it's rare we get to talk about aesthetics on the nature podcast.
Sure, But of course you've used the AI to identify the damage and used your own knowledge to try and fill in some of these damaged areas.
Of course, there is the chance that this isn't what the artist intended back in the 1500s.
Oh, yeah, definitely.
And the good news is that this kind of laminate that's applied as a mask that's very easy to remove.
You peel it off and it's made with materials that are dissolvable in conservation grade solvents that really don't run a risk of damaging the paint layer.
And if we have new information, what the artist would have intended, we can just apply a new mask.
And indeed in this study I tried a number of other variants to see how different parameters would affect the way it looks after the restoration.
And the version that's left there is the one that I thought was most reasonable.
And of course, your paper is out now.
It's only been a few years since you've been working on this.
What's the long term stability for this method?
Because of course, you're trying to protect the painting from damage and maybe give a sense of what it looked like when it was originally done.
Is this a long term fix, do you think?
So I ran an experiment to just artificially age the laminate film, not the one that was actually applied, but just a little sample.
And to then evaluate basically under reasonable conditions how long this would be expected to last.
And my expectation is it should be able to last a century.
But also this is the first time digital methods have been hybridized with physical ones to restore a painting in such an aesthetic sense.
And my expectation is that new developed techniques, especially ones using nicer hardware, that's going to enable us to do these kinds of restorations with an even greater degree of certainty that they will last a long time.
And Alex, you've shown N equals one and this is a potential method to do this sort of thing.
But what does it not do?
People are very excited about AI with this.
And that's very fair.
It's very useful.
But there are two caveats to current AI things that I have not resolved.
The first one is that most AI models do not scale to very large image sizes.
And the other bigger part, and this is down to training data.
We don't really have a good corpus of data of paintings that have been restored well and the before and after images of them at very high resolution.
So it's very difficult to capture the qualitative difficulties present in conservation.
This is going back to the questions of why are we correcting a certain damage and is that damage bad enough to correct?
And what about conservators?
Because, of course, we cover a lot of papers that use AI to a greater or lesser extent.
And people are concerned that here's another one that's going to ultimately potentially replace a lot of what I do.
Have you spoken to any conservators about this?
And what have they told you?
I have.
And to most conservators it's pretty obvious that what I've done does not replace a conservator in the slightest.
So there are so many areas that conservators work in, and most of them are tied with material stabilisation.
We don't have robots that can take a painting, evaluate how sound its substrate is or pass it through the necessary examinations to determine if wormholes on the inside of the support for it are problematic.
My work just expedites the portion of the work where a conservator is just trying to bring back a work to a normal state.
And that's not painting new art.
That's something that I hope conservators will appreciate.
I've met some that appreciate it.
I've met some that are, you know hesitant, but these are all questions that i assumed would be controversial and i assume the field as a whole will move forward and try to leverage what tools there are to ensure that conservation can happen on more and more works that need it.
Alex kashkin, there to read their paper.
Look out for a link in the show notes, where you can also find a link to a video showing the process in action.
Finally on the show, it's time for the briefing chat, where we discuss a couple of articles that have been highlighted in the Nature Briefing.
But why don't I go first this week?
I've been reading an article in Nature about a new kind of metric and integrity index, it's called that could help make it easier to spot institutions that are chasing conventional publishing metrics.
So integrity is something that we've talked about on the podcast a lot before, and ranking things as well.
And this is doing both.
Yes.
The idea behind this is to get a sense of where institutions may have been chasing quantity of papers over quality.
So, as we know, many universities to determine their rankings, they look at things like number of papers published and amount of times those papers have been cited.
And that can be a useful measure for all sorts of reasons.
But it can also be a bit of a double-edged sword, because it could mean that some researchers or some institutions could just chase these big numbers without really thinking about the quality and may not always do it in the most ethical way.
So basically then, by flooding the literature corpus, then you can potentially game your position in the rankings.
Exactly.
And, as we've talked about as well on the podcast many times before, there are paper mills out there as well.
And these are organizations that try and sell authorships to researchers on sometimes real papers, sometimes entirely fake papers.
And that's one way in which people can try and boost these rankings.
So the idea behind this is to try and shine a light on where stuff like this may be happening, so that the institutions and other researchers can put a bit more of a careful eye on these places.
Well before we get to the rankings themselves nick, maybe we can talk about how the list was pulled together.
What were the metrics involved?
So this ranks institutions by the proportion of papers that are published in delisted journals.
So you have listed journals in things like Web of Science, and these are usually listed based on a lot of measures of quality, including effective peer review, adherence to ethical publishing practices all the sorts of things that we want in science.
So a delisted journal has not been listed on that for not reaching those criteria.
The other thing that takes into account is how often papers from this institution have been retracted and, as we know, papers can be retracted for good reasons maybe researchers spot a mistake or something like that that they want to correct in a scientific record.
But papers can also be retracted due to misconduct, things like data manipulation.
And who's done the legwork here pulling all this stuff together?
So this is a metric that's been put together in a preprint, so an unpeer-reviewed paper by an information scientist at the American University of Beirut.
And essentially what this does is it categorises institutions in five different ways, going from low risk to red flag.
And red flag is where you have a high proportion of these papers being published in delisted journals and having a lot of papers retracted as well.
And what has the list thrown up then?
So they've looked at 18 institutions located in Saudi Arabia India, Lebanon and the United Arab Emirates.
And the reason that they've looked at these particular places is they have seen what's been described as an extreme publication growth.
So they've had a lot of publications recently and they've climbed the international rankings as well.
So by going through this list, they sort of rank these different universities.
And I can give you an example of what they found here.
So the Lebanese University in Beirut, for example, has seen a 908 increase in the number of publications authored by its researchers between 2018-19 and 2023-2024.
And that is much higher than the average increase of 17% across all other institutions in Lebanon.
And by doing this analysis, they ended up putting this university in the red flag category.
Now I should say as well that Nature reached out to the institution, and a spokesperson from it said that the surge in publications was attributable partly to a pilot program it launched in 2022 that supported 130 researchers.
And the increase in retractions of the number of publications in delisted journals occurred during a period of rapid growth in research outputs.
And they do agree as well that having a score for research integrity is a good idea in principle.
But they say that this particular one described in this preprint lacks institutional context and relies on generalized assumptions.
And what about more broadly?
What are other folk in this field saying about this work?
So in the article, they spoke to a couple of different researchers.
One of them said this is a good first step and should be used by the ranking agencies.
And they were the founder of India Research Watch.
Another researcher, a data scientist, also from India, said that it offers a strong and timely correction to a system that often equates research excellence with sheer volume.
Well research, integrity and measures thereof is something that we will continue to cover on the podcast because it is getting more of a spotlight.
I think it's fair to say.
But let's move on to my story this week.
And it's one that I read about on the Associated Press website based on a paper in Science.
Now, I talked about conservation earlier.
This is a very different sort of conservation.
This one involves large mammals, specifically rhinos, and research into how effective conservation cutting off their horns is to prevent poaching.
I see because I think that has been something that's been proposed in the past because poachers want rhinos' horns, so I guess cutting them off would make good sense.
But has there not been a lot of research and understanding of how well this actually works?
Yeah, that's right on all fronts there.
Rhinos are poached for their horns, which are hugely valuable... for use in traditional medicine.
And poaching is a real issue.
In South Africa in the first three months of this year, 100 rhinos were killed and it has caused populations to collapse.
And, as you say, cutting off a rhino's horn is one way to combat this, to reduce the incentives for poaching.
Now, horns are made of keratin, of course, a bit like our fingernails or our hair.
And cutting off a rhino's horn is relatively straightforward.
It takes about 10 minutes.
The rhinos are sedated.
They're blindfolded.
They have earmuffs put in, I've read.
And the horn is cut off with maybe a power saw or a chainsaw.
Doesn't hurt the animal.
Needs to be done every few years.
And this has been done for a long time as an effort to try and reduce poaching levels.
But as you've alluded to there, evidence of its effectiveness was sparse, maybe until now though.
So what has this study revealed?
Is it actually effective?
Yeah, so this research looked at 11 South African reserves between 2017 and 2023.
Now, this is an area of the world where a huge percentage of the world's rhinos live.
And what the team behind this work did is they compared data from eight reserves that did dehorn rhinos with three that didn't.
And the results showed that dehorning over 2,000 rhinos resulted in a 78% reduction in poaching.
So evidence that this approach is indeed effective.
And that makes a kind of logical sense.
No horn, no reason to poach.
But I wonder, does it have any effect on the rhinos?
Yeah, that's a good question.
And that has been something that there has been some pushback on from some quarters, right?
A rhino has a horn function for a reason, maybe defence against predators, maybe competing for territory.
And there has been some research on this, with maybe some showing that territory size is reduced, but others showing that there's no adverse effects on things like breeding or mortality rates.
So it seems like the benefits of dehorning outweigh the costs.
But it has to be said, Nick, this won't end poaching.
It turns out, the rhinos are still poached for the stump of their horn.
That is still valuable, but it does appear that this method is effective and the research also shows that it appears to be maybe more effective and cheaper than other methods that are currently in play to try and halt poaching.
But the researchers say, of course this isn't a panacea.
This really does need to be done in tandem with other things like law enforcement, to really help these animals, whose populations have crashed around the world, maybe get back to healthy numbers.
Well, it's always good to have evidence-based approaches to conserve and protect animals.
Thanks for that one, Ben.
And listeners for more on those stories where you can sign up to the Nature Briefing to get more like them.
Check out the show notes for some links.
And that's all for this week.
In the meantime, if you want to keep in touch with us, you can follow us on Blue Sky or X, or you can send an email to podcast at naturecom.
I'm Benjamin Thompson.
And I'm Nick Pettichow.
Thanks for listening.
This moment, Alex's mile swim, it means everything to him and to all of us.
At Evernorth Specialty Services, we help providers make sure patient care doesn't get interrupted, especially for rare, chronic and complex conditions.
Now Alex and his care team have his hemophilia under control with life-saving medication and care management from Evernorth's specialty pharmacy Acredo.
So Alex can be the last one out of the pool.
Evernorth Specialty Services, because every moment counts.
Visit evernorth.com slash specialty to learn more.
Fall is here and Nordstrom has the latest styles and hottest trends, thousands of options for under 200.
I'm talking about plaid outerwear, oversized sweaters that are cozy, Western-inspired pieces, prairie dresses, and you know I love a skirt.
Nordstrom has free shipping and returns, which is why I love them, and they have fast options like buy online, pick up in store.
Shopping at Nordstrom is always easy.
I love that Nordstrom is always high quality and kind of guides me in what's trendiest for the season.
And they have the top brands like Veronica Beard, Favorite Daughter, All Saints, Coach, Rag & Bone, Tory Burch.
So shop now in stores or at nordstrom.com or download the Nordstrom app.
