TWiV 1369: Florida Model Makes Measles Great Again

0:00This Week in Virology, the podcast about viruses, the kind that make you sick.

0:10From Microbe TV, this is TWIV, This Week in Virology, episode 1369, recorded on October 23rd, 2026. I'm Vincent Racaniello, and you're listening to the podcast all about viruses. Joining me today from Madison, New Jersey, Brianne Barker. Hi. It is great to be here and to see you all. In Madison, it is pretty nice right now. It is Kathy's favorite temperature, 61 Fahrenheit, 16 Celsius. And pretty sunny, but I think it's going to get pretty cold this weekend. I saw Kathy in person yesterday in East Lansing, Michigan at Michigan State University. Also joining us from Western Massachusetts, Alan Dove.

0:58Good to be here. And it is 58 Fahrenheit 14C here with clear blue skies and probably about a week out from peak fall foliage color here in Western Mass. Western Massachusetts. And joining us from Austin, Texas, Rich Condit. Howdy. um there's a discrepancy in my various uh weather apps but i'm going with 82 because that's kind of i was actually outside and that's what it feels like and it's sunny and dixon is here but he's kind of lying low uh on the he's kind of passed out on the desk it looks like at brare dixon he lay low yeah yeah it's a good day we've been having oh morning temperatures in the 50s, which is just awesome.

1:48By the way, I realize, I apologize, this is a digression, but what do you guys use for a weather app? This is a loaded question. I just use the app that comes on the iPhone. Same. Weather.gov. It seems to be fine. Okay. So, you know, eons ago when the iPhone first came out, I decided that the iPhone weather app was garbage and i started using the weather channel uh forever okay and i just got ultimately so annoyed with the ads that you know pop up in places where you can't avoid clicking on them right that i thought i'd check out the apple app again and it's amazing it is good yeah It's got everything in a really nicely – According to the Apple app, it's 28C or 82 in Austin, Texas.

2:45Is that what you said? Exactly. That's why I went with 82 because I've come to trust this app. I'm really enjoying it. It's great fun. And 82 is another palindromic one of you. Oh, is that right? Yeah. No, it's very good, especially when you have lots of friends and family in a lot of different locations. I can have all of everyone's location and see everyone's weather. If I want a forecast, there's another site that's sort of specific for some local things here that I use. But anything other than like how much snow is the blizzard going to give us, I use the Apple. It's amazing. You know, if you just keep clicking, it digs deeper and deeper and gives you more information.

3:34It's got the moon. You know, it doesn't have a tide table, I don't think. No, it doesn't. I haven't looked for that yet, but I got a separate app for that. Okay. Enough. Sorry. I'll have to give that a look. I have dismissed that for years. Yeah, well, I was doing the same thing, Alan. I think you'd dig this. It's pretty cool. It's pretty good. It's pretty good. Just don't stop. Keep digging around. Okay. Because, you know, everywhere you click, there's more information, and there's no ads. It tells you all about the moon. Like, right now, there's a waxing gibbous moon. I like that very much. Yeah.

4:05The next full moon is in three days. It tells you when there's sunset, UV index. There's somewhere it talks about allergy index, too. And on the weekly overview, you know, you're looking mostly at, you know, some sort of general overview. But if you look up in the upper right-hand corner, you can click on it and specifically get precipitation. Yes. And then if you click down on a given day, you can slide through a chart of the day and see what's going to happen at any given time. unbelievable they really thought this through I was going to use it as a pick of the week but now I just blew that yeah I just went to other apps and I just went back to this because it's pretty good it's all that I need and you can now do what are these called when you put them on your desktop oh widget you can make weather widgets and have all kinds of things in them I like that very much Well, if you enjoy these programs, we'd love to have your support.

5:09We talk about science mostly, but we – That was science. That's meteorology. That's meteorology for sure. Although meteorology is more of a predictive thing, right? You're not quite sure what's going to happen. Well, I mean all sciences are trying to – you're trying to find fundamental principles that will allow you to make predictions. Well, but weather says tomorrow there's a 50 % chance of rain. Right. Because that's about the best they can do. Because they're honest about their statistical models. I'm not going to – if you bring me a gel, I'm not going to say there's a 50 % chance that there's a DNA there.

5:42I'm sorry. You either see it or you don't, right? Well, no. But you can then use the data, the hard data that you get from the lab to make statistical inferences. And we're going to talk about two papers today that do that. Yeah. I mean, part of the weather is also the fact that it is currently 61 degrees. Yes. Yes. It's a fact. But there's a lot of prediction in weather, which in science you do make a prediction, but then you test it with an experiment. You test it with the weather too because tomorrow you check the temperature and it's not the one you predicted. It's just the day comes and you see if they're right or not.

6:19It's a natural experiment. If you get a storm or if you get snow storms. Actually, this makes me think about a fork in the pathway for me professionally years ago, where I was considering two different things. One was animal behavior, and the other was laboratory science. And I went with laboratory science because I had no control over the animals. They did whatever the hell they wanted. I could observe them, but they did whatever they wanted. I couldn't really set up an experiment.

7:00Good point. Okay, let's go into our paper. Anyway, no, support us. Yes, we talk about science here. We diverge now and then, but that's what normal conversation is all about. And if you would like to support us because we don't do ads on our programs, you can go to microbe.tv slash contribute. There are a number of ways that you can do that. Okay, we have two papers today. The first is a paper in PNAS. It is interesting. It's under the classification of population biology. Projected effects of removing school entry vaccination requirements on measles transmission in Florida. This is by Alice Overson and Abba Gumel from...

7:46On page six are the author affiliations. University of Maryland, College Park. University of Maryland, North Bethesda. And University of Pretoria, South Africa. Even though it's two authors, they have three affiliations. Well, we've talked a lot about measles here and on Clinical Update. And measles was eliminated in 2000 in the U.S., but now it's back. And many states are reporting clusters. Huge numbers this year. I mean, last year there were 2 ,100 cases, which is a 650 % increase from the 2024 count of 285 cases. But I think now there are over 5 ,000 reported cases in the U.S. with eight deaths.

8:30And so that is being fueled by the fear about the vaccine, MMR vaccine, which is being promulgated by RFK and his ilk. And so this paper is about what's been proposed in Florida. In September last year, the state announced plans to remove all vaccine mandates for going to school, including MMR for school attendance. And this has been justified by the governor and the surgeon general, not because they're anti-vax, this is what they say, but because they want people to have the freedom to choose whether they get vaccinated or not, which is really not the appropriate position because if you give people freedom, then you have outbreaks, and that's what we're seeing.

9:32And freedom, I'm all in favor of giving people freedoms, But you have to understand that there are limitations to freedoms which happen when your freedoms infringe on my rights. And so if your freedom to not vaccinate your child endangers my child, then that is not your freedom. I do not have the freedom to not stop at red lights because I don't feel like it because I would hit someone else's clock. And you're not allowed to smoke indoors because the smoke will harm other people's secondary smoke attempts, right? And you can't say, I feel like smoking and do it. You'll get thrown out. Okay? So it's the same idea.

10:12If you're not against vaccines, then it's about freedoms and it doesn't work that way. Right. Okay? There's also a long history that demonstrates conclusively that vaccine mandates work. And they're effective at reducing disease burden. I mean, we required kids to get vaccinated for school and we eliminated measles. And as you'll see in this paper, kids are the target audience for vaccination against measles, right? You vaccinate the kids and you really bring down the projected outbreaks. But this proposal has been discussed a lot. A lot of people are concerned about it, of course. But what this paper does is do a modeling study to try and figure out the impact of removing vaccine mandates in Florida.

11:00So the driving question here is, how bad could it really be? Right. How bad could it get? Yes. And in particular, whether – so what they're looking at here is what they call vaccine-induced herd immunity. Okay, V-I-H-I-T, VHIT. So they say this question is whether removal of the MMR vaccine mandate would cause Florida fall below the VHIT needed to sustain measles elimination. Now, in the past, people have made similar estimations many times, right, to estimate the level you need for vaccine-preventable diseases. is they always assume homogeneous mixing within the population, but that's not always the case, right?

11:53It's not the case. Mixing patterns differ. I mean, mixing between people who are vaccinated or not, right? Often you have clusters of people who are not vaccinated, but often they're mixed in and it varies across ages, demographics, social settings, regions, and so forth. So children, for example, have more close interactions than adults. So the effect is going to be different. And so they say it's important to incorporate this heterogeneity in any model that you do. So the measles control is very sensitive to small reductions in population level vaccine coverage because it's extremely infectious or transmissible.

12:40And recent modeling studies have done similar things about transmission and vaccination-induced thresholds. And usually in the U.S., it's between 91 % and 98 % you need to have vaccinated to have effective restriction of measles transmission. But nobody has looked at the Florida situation, how eliminating the mandate would affect transmission and so forth. So they have developed a Florida a specific age and contact structured mechanistic model to estimate the vaccination reproduction number and compute the VHIT and quantify the population level consequences of removing the entry requirement. And so they developed their own age structured model tailored to Florida's population, but it's fitted to measles incidence data from an outbreak in Gaines County, Texas.

13:39So they use real data to fit this model. And if you'd like, you can look at figure one and see lots of variables that you may or may not understand. I don't understand. So we won't go into that. We'll just look at the results. The cool thing is – I would say that the basis of this is I have seen the foundation for this kind of stuff before, where you generate numbers for susceptible people, vaccinated people, exposed people, infectious people, recovered people. And there's a dynamic relationship between all of those that affects the overall situation. And then this layers on the complexity of having two focused populations, children and adults.

14:32So it's just a more complex version of what, in my limited understanding, is a fairly standard modeling. Yeah. So people, so I am most familiar with the model as the SIR model. And with SIR, here they've added in exposed and vaccinated individuals. into SIR. SIR would be susceptible, infected, recovered. Yeah, R stands for different things, but recovered is one of them. And this is basically a more complicated SIR model. Right. And just for those of you who do want to look at this figure but don't have institutional access, this is unfortunately paywalled. Okay, so using their models, they generate a heat map of this vaccination reproduction number as a function of steady-state vaccination coverage for children and adults.

15:26And so this is a graph or a heat map, and the y-axis is child vaccination coverage. The x-axis is adult, and then on the right-hand y, it's this R, the vaccination reproduction number. And you can see that they have this star here, which is the current vaccination scenario in Florida. It falls below this line that you would need to prevent the outbreak space. The line is the V hit. That's the threshold. If you're to the right, if you're on the green side of that line, you're in a situation that would prevent transmission and really snuff the thing out. If you're on the red side of that line, you're playing with fire.

16:19So currently the vaccinated – Florida is currently in the yellowish part of the red side of the line. They are clearly on the wrong side. So they have 91.6 % vaccination coverage in Florida. And what's really nice about this is that they're doing both children and adults because you could imagine somebody could say, oh, we just make it about choice. You can choose as an adult. Let's not make kids be the ones getting vaccinated. And here they actually can separate out the importance of vaccinating kids and vaccinating adults. I also I was gonna say the other thing I really liked about this when I first read the title I was very curious about how they could predict how many people were going to get a vaccine if the mandate went away and how they were going to say this is how bad it could be because this is how many people won't get vaccinated and I kind of wondered how they do that and what I learned really quickly from this figure is it doesn't matter because we're already in the bread zone Yeah.

17:20Yes. And there have been about 150 cases of measles. So there's virus circulating. And their point is that if you reduce vaccination levels anymore, you're going to have an outbreak because it's going to push you even further away from the VHIT. And also, the way they're presenting the data in heat maps and the way their model runs, they don't actually have to pick a number of what percentage you have to fall below because you're getting this curve that says, if you're on the wrong side of that curve, you're on the wrong side. And the further you get down, the higher your chances of having a major outbreak.

17:59So then this is an interesting next paragraph. It begins with the word thankfully. Thankfully, figure two shows that increasing vaccination coverage in either children or adults leads to a substantial reduction in the vaccination reproduction model. So if you maintain adult vaccination at the current baseline of 92.5 % and you increase child vaccination coverage 1%, you get a 4 % reduction in the RV. And you can also do the other way around. You can hold child vaccination and increase adults. But when you consider per vaccine dose, you see differences. Vaccinating one child results in an average reduction of about 1 times 10 to the minus 6 in the RV, whereas vaccinating one adult, you get 10 times less, 3 times 10 to the minus 7.

18:54So on a per-dose basis, vaccinating children is far more effective in reducing the RV and hence the disease burden in Florida than vaccinating adults. And so they say the child-focused vaccination-focused control strategy would be very effective at reducing this RV. Like maybe a school mandate. Yeah. Not only that, but there's five times as many adults as kids. Yeah. So that's more vaccine doses. And the kids are all, you know, they're all going to school. Right. I wonder if you could corral them. I wonder if you could sort of look at similar data to this, because sometimes people talk about, you know, the idea that some vaccine coverage wanes over time in adults.

19:42And this shows us that vaccinating the kids and having the response in kids is of particular importance. So I also wonder if this tells us something about the fact that if it's going to wane, it's going to wane in the adults and the adult doses are somewhat less important. I just wonder whether you could address some of those conversations with a similar kind of model. Yeah. So anyway, currently there still are mandates in Florida. They haven't been overturned yet. Okay. So that's probably why we're not seeing huge increases in the outbreak. Well, that and luck. Yeah, right. Okay. So what's going to be the impact of the mandate removal?

20:26So they theoretically evaluate the potential impact. And they say, you know, national surveys have found that about 16 % of parents are not confident that MMR is safe for children. 17 % of parents believe that the risks of MMR outweigh its benefit. And 18 % of parents oppose school entry vaccination mandates for MMR. I want to stop here and just 16 % are not confident. MMR is one of the safest vaccines out there. Where are they getting this not confidence from? I'll tell you. RFK Jr. RFK Jr. Exactly. He says on talk shows that it kills you and it's horrible and it's worse than the disease. And they believe him.

21:14And Andrew Wakefield and all the, yeah, the whole PR machine around. Denny McCarthy, I think, involved. Oh, yeah, yeah. She jumped on that bandwagon. The experts, the scientists, the physicians, the public health people counter it, but it doesn't matter because they control the narrative apparently. Anyway, so they estimate that mandate removal could result in up to an 18 percent decline in MMR coverage for children, causing vaccination uptake for children to fall to about 82 percent. And you can just look at figure two. Yeah. And see what that would do. That's not even on their graph. It doesn't go down to 82%.

21:57That's right. It's deep, deep red. Yeah. And I kind of wonder, you know, they say that's worst case scenario. Is it? Because that means that all the people who oppose vaccines aren't getting vaccinated. But what about the people who are just sort of lazy? Yeah. I'm sure it would be worse. Yeah. Yeah. So that's based on those national surveys. Right. So, I mean, they could have taken the 16 % number, which would be not as bad, just the parents who are not confident it's safe. They took the 18 % number. But Brianne's point is excellent, that there's a much larger population of parents who just really haven't given the matter much thought.

22:34They get the kid the shots that the pediatrician says or that the school form says, oh, you got to get your measles shot. I guess we didn't get around to that. Yeah. Okay, so under this scenario, they predict Florida will have a pronounced measles outbreak within three to five years, and specifically the estimated peaks for these low vaccination coverage numbers that we talked about will exceed those of any year in the U.S. since 1977, and estimated infections ranging between 260 and 315 cases per 100 ,000 children. And they also find similar results in adults but with numbers about half the size at peak outbreak, 135 to 160 measles cases per 100 ,000.

23:27And when they do longer-term simulations, they get more sustained measles outbreaks. But the reduced size, they have smaller size and increased frequency. Because people have then been exposed to measles and are not going to get reinfected if they survive. Two things that interested me about this. The first was that if you lift the mandates and people quit vaccinating, you don't see the effect right away. It can take a couple of years. So I'm sure the anti-vax people can use that to say, well, it wasn't lifting the vaccine mandates because it wasn't coincident. But nevertheless, just in terms of the epidemiology, that interests me.

24:11The other thing that interests me is, as Alan mentioned, depending on what the vaccination level actually is, you get peaks.

24:24It's episodic in nature. You get a peak and then I guess that takes care of all the susceptible people for a while and then they reappear and you get another peak. Just like back in the pre-vaccine days. Yep. They also looked at mortality, measles-related mortality. And, of course, these outbreaks will give you substantial measles-related mortality.

24:53So that's a nice – they have a heat map of that as well. You can see that in figure five. Yeah, and you can see the time scale in that as well. So if it falls down to 0.82 in children, which is what they predicted, then the number of deaths, of course, go up with a number of cases. But it takes a while. So that's a heat map of how many children this policy change is likely to kill. Kill, right. Yeah. I mean, that really should be part of the discussion, shouldn't it? The consequences of giving you freedom. Yeah. That you die. Yeah. Or your children die. Your kid. Or somebody else's children die.

25:36I mean, what's sad here is that the kids have no say in the matter. Yeah. Right? Right. And maybe a lot of them would say, you tell them, here's your choice. You can not get vaccinated. You might die or you can get vaccinated. Nothing will happen except you will be healthy. What do you think kids are going to say? Anyway, they do some long-term dynamics as well. So under the worst-case scenario where you have this 82 % coverage in children, no other interventions. Their models predict or simulations show that persistent outbreaks with increasing frequency and decreasing amplitude emerge within roughly a decade.

26:15And Florida will experience these outbreaks over the next 50 years, and the virus will become endemic in Florida. What does that mean? It's always circulating and it's harder to eliminate. I mean, we introduced measles vaccine in the U.S. in the 60s. It took until 2000 to eliminate measles. It's a lot of work in all the states. Okay, so then they did a model where they assessed the impact of catch-up vaccination of adults who are 80 % of the population of Florida. and increasing adult MMR coverage from baseline of 92.5 % by 4 % or 5%, and you keep the child vaccination at 91%, can also reduce the reproduction number below 1%.

27:06So a catch-up adult program, raising adult coverage to 96.5%, could push Florida above the VHIT and maintain the state's elimination status, because right now it's eliminated in Florida, as long as the The children's coverage does not change, but that's probably going to change too. And that's a big ask. Yes. Because it's 80 % of the population, so it's a lot more adults than kids you've got to vaccinate. It's almost a million adults. And, I mean, come on, how hard is it to get adults to get vaccines that are recommended for them? People don't get their flu shot. They don't get their shingles vaccine.

27:50They don't get their RSV when it's recommended. They just – the chances that they're going to go get their measles vaccine to boost those numbers up that last 5%, it seems pretty remote. Yeah, they say the public health effort required to achieve such coverage would be substantial. Yeah. And it's also predicated on the assumption that childhood coverage remains at 91%, which they're saying is probably not going to stay there. And I want to point out that throughout this paper, they are really bending over backward to come out with the best scenarios they can. Like, let's come up with ways that this could be mitigated.

28:30Let's imagine that you could boost adult vaccination that much. Well, it would be okay if the child vaccination doesn't drop. they're not this is not some paper you can point to and say oh they were just trying to show that the the man the removing the mandate's a bad idea no they're really doing the work of the mandate opponents to the extent they can and saying the model still gives us this yeah they're not like you know fear-mongering with this they're actually presenting the data well and and going beyond Beyond that, they're presenting the most optimistic scenarios in many cases. They also point out that this idea of compensating with the adults only works if the childhood vaccination stays where it is now.

29:20If it falls below 90%, the adult catch-up is not going to work at all. So they've modeled it and they'd show that it could work under certain circumstances, but if the childhood vaccination drops, it will not. Right. So at the current childhood vaccination rate, adult catch-up will work, but as soon as the child drops, the child rate drops, it's not going to work anymore. Okay. So then they say, what about non-pharmaceutical interventions? What kind of impact might those have? And this would be the post-rash isolation of people and case-triggered pediatric mask mandates. Oh, gosh. We're kids. That's going to go over real well.

30:01So for the people who don't want to get vaccinated, you're going to tell them to wear masks. And this is actually an example of what I was mentioning, where they're bending over backwards to come up with best cases. If you look at the numbers they use for their modeling on masks, based on some data from other viruses, and we talked about this extensively in the pandemic, Yeah, masks work if we're talking about healthcare workers in a hospital during a flu outbreak who conscientiously wear properly fitted masks. You can get 80 % efficacy in stopping spread. And they're going to use those numbers in this let's put masks on the third graders, which is actually kind of preposterous that you'd get 80 % efficacy in that setting.

30:59But they're going to say, OK, this is the best you could do and see what happens. Have you seen kids wearing masks, young kids? Oh. They're hanging all over the place. They're just – Have you seen – I mean, think back to 2021. You'd go to the grocery store and 25 % of the people had them on as chin hammocks. I mean, that's adults. Yeah, adults would too. Yes. I know. Kids are even worse. Anyway, so they model – so first of all, even moderate isolation compliance brings Florida above the VHIT, okay? And even masking without isolation shows that you can get a reduction. If we assume 80 % efficacy for the masks, which, as I just said, is kind of absurdly generous.

31:47Yeah, it is high. And when you use both, post-rash isolation and masking, you get the most robust system of defense. But then they say, you know, it's financially and logistically difficult for people to take time off for work to isolate, get child care, and comply fully. So this is not realistic that it's going to be extensive enough to make a difference. And by the way, a lot of the people who are in favor of this are, you know, claim to be fiscally responsible Republicans and you're going to pull all these parents out of work and incur childcare expenses to isolate the kids when you could have given them a cheap vaccine and prevented this.

32:31What is a fiscally responsible Republican? That sounds like an oxymoron. Well, that's, I mean, that's what they claim. But of course, if you look at the actual numbers, the current party is not. I mean, they want to take health care away from 10 million people. Yeah, no, no, no. No, I'm not saying that they are. I'm saying that they claim to be. Okay. That's the claim, which if it ever held water, certainly doesn't anymore. Good Lord. All right. So they say it's important to remember that even if these non-pharmaceutical interventions work, the next time there's an outbreak, you're going to have to do it again and again and again.

33:09Whereas if you get vaccinated, it will protect you for a long time here. So vaccination is preferable to that. So these things cannot substitute for the immunity provided by a vaccine, right? School entry vaccination mandates remain the cornerstone of prevention, ensuring that each new birth cohort enters the population protected and preventing the gradual decline of herd immunity. It is the cornerstone, and that's why it's been used for so many years so successfully. So that's the story. In the end, this is a bad idea, and it's politically motivated because the one party has made as its basis having your freedom in all ways, including medical freedom.

33:59And I don't understand this because was this ever an issue before? I mean, I grew up, you get vaccinated. That's it. Vaccines work. You go, you get it. I don't know why it's suddenly an election issue. Well, in a certain population, apparently, actually, this would be an interesting study. Does that particular stance actually help you get elected? The assumption is it does, but I wonder. I think the only thing that matters is the economy. Yeah. I don't think people care about anything else. It's the economy, stupid. It's the economy, stupid. They don't care about science. I think there are pockets of people who do.

34:44Yes, of course. But many people do not care about any of these issues. They don't care about immigration. But I was just thinking previously, you know what? Even if you don't care about anything, why don't you care about your democracy? It is what has given you a good life. You want to throw it away? That I don't get at all. I know everyone is not against a democracy, but my gosh, this is what we're headed towards. I think the overwhelming majority of Americans actually have, in a lot of cases, pretty sensible views on a lot of this. And they vote for a variety of reasons that are not necessarily the reasons you think.

35:29Yeah, they see a TV ad and they get convinced. I don't think it's that. I think if you talk to people, they have more nuanced ideas about why they're voting particular ways. Alan, things are so screwed up right now. It's not possible that these elections can be so close. It makes no sense. If you see something not working, your gas prices are high because of a verkakta war and tariffs and killing innocent people, you like that? Vote against it. Why is it so close? But this is not a political channel. No, no, no. We'll move on from there. In terms of vaccination, it makes no sense to eliminate mandates.

36:11If you have a medical issue, that's fine. I have no problem with medical exemptions to vaccinations. But anything else, no, you need it to go to school. And if you have a reason you don't want to vaccinate your kid, then you have to homeschool him or her. That's it. It's the end of this. That's my feeling. But I was raised in an era where you just went and got your vaccine. It wasn't about because keeping the population healthy gives you more freedom so you don't have to spend your days in an ICU. Right? Yep. Okay. Hey, Florida, get your act together. The governor's not up for election, is he?

36:51I don't think so, or we'd be hearing about it. No, I think, I thought there was a Florida governor election. Let's look.

37:08Yeah, wait. There is an election, yes. Yeah, there is. Yep. Well, actually, DeSantis is finished. Right. Okay, so we got two other people. Two noobs. And I hope that they decide to get rid of the Surgeon General and reverse this. This is – anyway. Let's go on to the other paper. This is another paper in PNAS. Yes, this one is HLA class one escape drives the evolution of SARS-CoV-2 in human populations. And so the first author is Ekaterina Riomina and the senior author is Georgi Bazikin. And the affiliations are once again on a later page, which I am scrolling, scrolling, scrolling down to. Here we go.

38:04Russia. Yes, Russia. So one's an independent researcher. Then we've got National Research University, Higher School of Economics University, Moscow, Lomonosov Moscow State University, Central University of Moscow, National Medical Research Center for Hematology in Moscow, and the Russian Academy of Sciences in Moscow. So it's not only Russian. It's a pretty much Moscow group. Okay, so there's also a commentary by Mahan Ghaffari, which I may refer to also in PNAS. It's called The Other Immune Landscape of SARS-CoV-2 Evolution. So I just want to point out the way you submit papers to PNAS is you find a member.

38:48I used to do this. I think it's slightly different now, but it goes through a member. And this one was handled by Eugene Kuhnen, which means, so he's an academy member. and he considers it for publication. You have to send it for review, I think. I'm pretty sure you have to send it for review. I think that's true. It didn't used to be. It used to be that a given member could, without review, submit a certain number of papers a year. And we're talking about members of the National Academy of Sciences, which is an organization that exists to elect its own members. Exactly. It's not clear to me what the function of that organization is anymore because they don't say anything.

39:26Anymore, yeah. It used to have functions, but it is. They don't say anything about bad things that are happening to public health in this country. Anyway, I just want to point out one thing. So the manuscript was received by Kunin on July 1st, 2026. It was accepted on July 20th. Who would review this in two weeks? Okay, I'm just saying. I think that's awfully fast. You know, if you look at some of the nature papers we look at, it's like a year submitted and then accepted the next year because they make them do more experiments. I'm just saying. I'm not making any judgment. I'm just saying. Vincent, I'm impressed with your ability to not make any judgment, and I will try to follow your suit.

40:11No, this is no question that this is a contentious paper, okay? But I thought it would be a good discussion because the underlying issue I teach in my class and I want to get it right. Okay, so this is all about SARS-CoV-2. And as you know, when a virus infects you, you have two main adaptive immune responses. You make antibodies and you make cytotoxic T cells, right? And we know that SARS-CoV-2 has changed over the years so that the spike protein amino acids change so that they're not as well bound by antibodies. So the antibodies don't work as well. So that's antigenic evasion, right? Antibody evasion.

40:51But the other half, the cytotoxic T lymphocytes, they recognize a viral peptide that's displayed in a MHC molecule on the surface of the infected cell. We talked about T cell epitopes a lot. We had Alex, what's his name? Sety. Sety. That's the one that does the intelligence out in space, right? No, I'm just kidding. We've had him on to talk a lot about how the variants of concern don't have changes, mostly in T-cell epitopes, so the T-cell response is still there, okay? So that's what this paper is all about. The T-cell response targets linear epitopes on peptides. So there are a lot of T-cell epitopes, many more than antibody epitopes.

41:34And remember, the T-cell epitopes are presented in complex with HLA-MHC molecules, which are also polymorphic in the population. You know, your MHC may not be the same as mine. And so because of that polymorphism, that population-level polymorphism, it's always been thought that changes in T-cell epitopes have limited benefit because you may have a variant arising in you that changes a T-cell episode, But the people you contact may have a different HLA or MHC and it may not matter. Probably do have a different HLA. If you read this paper, they're throwing doubt on that, right? So we're going to have to talk about that, right?

42:21Uh-huh. And they're making the point here that the T cell variation is not as insignificant as everyone thinks. And one of the things is that these HLA types are often clustered. So you may be surrounded by similar HLA people and types in your population, right? So they wanted to do a study of – so they mentioned HIV-1, which I know Brianne has thought a lot about. So let me tell you what they say. Outside of SARS-CoV-2, HIV mutations in T-cell epitopes, I would say, amino acid changes in T-cell epitopes, are favored by selection in those geographic areas where the cognate HLA class 1 alleles are more frequent so that you have geographically local escape.

43:13And that makes perfect sense, right? Mm-hmm. You don't agree, huh? You can hold your objection. That's fine. Maybe it's wrong. Maybe it's incorrect. That's paper number 23. Uh-huh. And for SARS-CoV-2, it's thought that people are persistently infected with SARS-CoV-2 because they have some kind of immunocompromise. And they may be more – in them, the T-cell variants may arise and they may have more of an effect in them. So what they do in this paper – Can I give some more background? Yeah, yeah, of course. On sort of the immunology here. So as Vincent points out, we have different MHT class 1 alleles amongst the population.

43:58MHT class 1 is the most diverse gene in the human population. So we all have different versions of the protein and are presenting different peptides. and you could imagine a situation where you may have a mutation in the virus that changes amino acids such that they no longer can bind to the MHC molecule. That's usually what we think about with this type of mutation. And because it won't bind, then the T cell can't recognize it. And then the T cell can't recognize it. So we're basically sort of, you know, the virus is hiding from the T cell because its peptide is no longer being presented to the T cell.

44:42Now, wouldn't the virus have to do this with more than one peptide? So the virus would potentially, would maybe have to do this with more than one peptide. There are some situations where it absolutely happens with one peptide because that peptide is involved in a very dominant T cell response. And so the T cell responses are not equal. And if you happen to change a peptide that is involved in stimulating a very dominant T cell response, that could be sort of, you know, hitting the lottery in terms of evading this very dominant T cell response. This is a challenge because changing an amino acid in one of those epitopes is changes the amino acid in the protein and perhaps might make that protein not fold correctly or otherwise not work in the virus.

45:46Perhaps the virus will become less fit because of that change in the amino acid. And so the virus is less fit sort of in someone who doesn't have that T cell response, but in you, because of your T cell response, the fact that the virus is perhaps less able to reproduce, less able to replicate is okay because it's avoiding T cells. And in many cases, we think that this whole process, when slash if it happens, happen, it takes a while because it takes a while for the virus to make whatever changes are needed to get a structure that will actually work for viral replication and also avoid T cell responses.

46:38Sometimes you need additional compensatory responses or compensatory changes elsewhere in the protein to make the folding work. And sometimes it takes a while for all that to happen. And so when people have looked at this in the past, they often look in, say, with HIV, They look in SIV in animal models where they use an animal model that has a particular MHC class one molecule that we know presents a very, very, very strong T cell response or a very strong epitope that gives a very strong T cell response, a super dominant T cell response. So that there's a lot of pressure for there to be change.

47:23And we do see that happen in the case of, say, macaques that have that particular TSA response. We can actually rank different epitopes in SIV in terms of how frequently we see this. And we end up getting some interesting information about the plasticity of different proteins within the virus and how much those can tolerate changes. Um, but as, as Vincent said, the, the general idea here is that this can be really important on, in you as an, at an individual level. If you have a virus over time, um, that virus may pick up some mutations that lead to amino acid changes to avoid your T cells. And that's sad for you.

48:11Um, and it might only happen in acute, in chronic infection because it might take a while. Right. But if we actually then take this outside of the individual and go to a population level, it's unlikely that we're going to see this change persist. Because while the virus may have undergone some mutation and made some amino acid change in me that made it more fit in me because of my T cells, when that virus transmits to Alan, who has a different T cell response, now the fact that the protein is messed up and that the virus is less fit actually really matters. And now Alan is actually going to be able to eliminate that virus really easily.

49:00So in fact, that is what was more fit in me is less fit in Alan and is probably going to get killed out really quickly. And so we're not going to see any of these things persist at a population level. And so the idea has always been there's a paper I actually actually now that I think about this, it's really funny because if I had done it this year in class, it would have been today. that we actually look at one case where we do see these sorts of changes happening, which is in mother-to-child HIV transmission, where mom and baby have the same HLA type, and you can actually see adaptation to mom's HLA type and then transmission to the baby and the baby being really sick.

49:44But the idea has always been that this doesn't necessarily happen at the sort of population level. And so if you look at one of the papers that they cite here with HIV, they cited a paper that looked at two individuals and saw that it happened on the individual level in those folks, but not necessarily at the population level. They do talk about a big population level study that's pretty famous in the HIV literature as well, which I can say more about later. But this is sort of the big picture idea. So how is the antibody response different from that? So the antibody response is binding to just the antigen without your MHC molecule context.

50:30It's really about the three-dimensional structure of the antigen. And the three-dimensional structure of the antigen in me and the three-dimensional structure of the antigen in Alan is the same. And so if the virus achieves a new structure that can avoid antibodies in me when it moves to Allen, that structure is still evading Allen's antibodies. But because of the MHC being unique from person to person, the epitopes are a little bit different. The antibody epitope is the same from person to person. And that's why previously we've thought it only matters in chronic infections where it matters to you.

51:10hep C, HIV. But SARS-CoV-2 is not a chronic infection, except in immunocompromised people, and maybe it makes a difference there. But even there, there are only a couple of patients that have been looked at for chronically infected SARS-CoV-2. And by the way, if I've got all this straight, this is actually a good thing, right? Because the virus has, it's hard to get around the T-cell response. Yeah, it's hard to get around the T-cell response. And basically, if we think about this as a population biology issue, the fact that we have such MHC diversity means that it is not possible for one pathogen to get us all.

51:49Yeah. It might suck for you individually, but as a population, we're good. Diversity is an extremely important survival strategy. So folks, no one virus is going to wipe out humanity. People always say, could the virus wipe out... No, it won't because of MHC. Okay, so here's what they do. First, they have a data set of all the SARS-CoV-2 T-cell epitopes, CTL epitopes that have been shown to be immunogenic. Okay, they're actually working. And they're dense across the genome. So the mutations that have led to protein amino acid changes in SARS-CoV-2 over time. 69 % of all of those happen in CTL epitopes because there's so many CTL epitopes.

52:45Okay. So this is the database they have. So then, so the idea is a change in the CTL epitope can affect, it can destroy the peptide HLA binding, or it can destroy the TCR binding to the peptide HLA complex. So there are two different things that can be happening there, not just HLA, okay? So in both cases, they expect more frequent HLA alleles to increase the frequency of the escape mutations, right? If you have increased... There are some alleles that are more prevalent in the population than others, right? Some are very popular and others are less so. Okay, so they find that many of the SARS-CoV-2 amino acid changes disrupt presentation.

53:33And 14 % of those, that's 34 of the mutations of the amino acid changes that we see in variants of concern and 25 ,000 of post-variant of concern changes cause a loss of predicted binding to HLA. So these are predicted. Again, they're not experimentally verified. They're all predicted by looking at the sequences. An additional number, 54 and 41 ,000 for VOCs and post-VOCs were predicted to reduce binding by a factor of one and a half or more. So the other ones had a different extent of binding. And some of them would increase the binding as well to HLA. So you could decrease binding or you could increase binding.

54:23And 35 % of the variant of concern changes, 39 % of the post-variant of concern, and 34 % of the post-variant of concern that attained high population frequency, they impair epitope binding. So this is the characteristics of the database that they're looking at here. So basically, it looks like using these sort of predictions that the virus is getting more changes that avoid binding to common MHC class 1 molecules, just based on sequences that are observed and how those observed sequences are predicted to bind. Right.

55:06So, most post VOC, so post VOC, we had variants of concern for a while and then they just became variants, no more concern. Most of these EVs, these are the escape peptides, most of them are the escape amino acid changes. They have a low frequency in the viral population, 96 % of post-Delta and 98 % of post-Omicron variants. Never exceed the global frequency of 1%. But they say individual genomes may carry multiple escape changes, and the total effect may be non-negligible. And also, the number of changes per genome has increased throughout evolution. There's an interesting figure showing that, suggesting that they may have – they're playing an increasing role.

55:55Is it appropriate to call these escape mutations when we're modeling binding affinity? We don't actually know that an escape occurred. Right. Right. So one of the things that I think is really important here is that technically what they're showing us is they're showing us that the virus has evolved, which means that sequences have changed in frequency over time. Right. There are multiple reasons, whether whether we're talking about rabbits or whether we're talking about viruses, there are multiple reasons why evolution can happen, why allele frequency changes over time, one of which is natural selection.

56:36So they've shown us evolution, and they're assuming it's because of this escape mutation. They're assuming it's because of natural selection. And their sort of predictions are telling them why they're making that assumption, but they don't actually show us that there is a selective advantage here. Yes. So what does EV stand for? Where does EV? Where are you saying? Because now they've started to use it and I don't remember what it stands for. I think they're calling that escape variant. Escape variant. Okay. So that's Alan's question. Can you really say that? Epitope variants. Epitope variants.

57:13Epitope variants. Epitope variants I'll give you. Yeah. Epitope is fine because they are variants. They are variants of epitopes. And all they're doing is looking at the frequency in the population right now. Yep. And they say it goes up. Okay. Right. So then they go on to another study where they show throughout the pandemic, frequencies of the variants differed between countries, right? And also HLA alleles differ between countries. So they could then ask whether T-cell associated selection shifted the population frequency of post-Delta and post-Omicron variants. The reason they're doing post-Delta and post-Omicron is that the post-Omicron was an era of huge numbers of changes compared to Delta.

58:11So their idea is that the frequency dynamics in the population are shaped by HLA presentation. presentation. So they expect the viral escape, now they're using viral escape mutations associated with loss of binding to a specific allele will increase in frequency for more common HLA alleles, right? That makes perfect sense, right? You have a common HLA allele, you should see increased frequency of changes that prevent binding to that particular allele. So to do that, to address this hypothesis for each epitope, each T cell epitope and country, they identified a single loss driver mutation, and loss driver in quotes, that had the highest frequency among the changes leading to a loss in HLA binding.

58:58And the increased driver mutation is the one with the highest frequency among those with increased HLA binding. All right. So a driver mutation causing a loss of HLA presentation could be observed at widely different frequencies in different locations, again, consistent with the geographic variation. Although they don't ever tell you which HLA are in which country, right? Do we have that data? Those data exist in other places, and much of the information that they are using in this paper comes from the IEDB, the Immune Epitope Database. Right. And that information does exist in IEDB. Because that would have been nice to say, okay, in this country where we have this HLA, this particular D-cell hepatope didn't work.

59:53Okay? They never do that. It seems to me that the data, although actually in the commentary they lament that as well. I'm going to go back to the commentary at the end also. Okay. All right. So independent of the difference in frequency, lost drivers that allowed escape from frequent HLA alleles gained higher frequency compared to both lost drivers that allowed escape to less frequent HLA alleles. Okay. And they observe this across both Delta and Omicron and across countries as well, which each has its own HLA composition. But you don't see this pattern for epitope-matched synonymous mutations.

1:00:35Okay. So in other words, again, I'll repeat it. Lost drivers that allow escape from frequent HLA alleles gained higher frequency compared to lost drivers that allow escape from less frequent HLA alleles. Okay. So, okay, there you go. In some numbers, on average, lost drivers for frequent HLA alleles were 1.7 to 2 more frequent than those for rare HLAs, which were like 1.3. And they saw this shift in country-level data and also globally. So they're trying to say you get a higher frequency with the more popular HLA alleles. One thing that would be sort of interesting to see is in locations with the same common HLA type, do you see the same escape variation?

1:01:28Right. That would be good. They don't do that. Yeah. Okay, so the next section is entitled, escape mutations are favored in countries with a high frequency of the cognate HLA allele. So that makes sense. If the viral escape is driven by the HLA allele composition of the country, you should expect that the spread of the changed epitope should depend on the frequency of the LEAR, and it should be favored by positive selection where the HLA allele is higher. Okay. So they look at that and they find that the loss driver mutations tend to reach higher frequencies in countries where the cognate HLA allele is more frequent in human population.

1:02:08I mean, I think that's so far the best data that they have, right? You find the higher frequency of the altered peptide in countries where the matching HLA is higher frequency. And, okay. I mean, they're good p-values. They're not amazing correlations. Yeah. 0.05. What were you going to say, Rich? It's just, to me, that's, I have a fairly naive look at this, but to me, that's the whole paper. It is, really. There's a correlation between these particular variants. and the prevalence of a certain HLA type, period. And they restate that a dozen different ways in the paper. Yes. And do a few miscellaneous experiments to try and support that.

1:03:04So then the next part, they were looking at changes in the peptides that disrupt T-cell receptor recognition, right? So you have the peptide in the HLA, and then the T-cell has to bind to it to the peptide. But if there's a change in the peptide, the T cell receptor might be disrupted as well. Okay.

1:03:26So they say epitope TCR binding cannot be reliably predicted computationally. I don't know. Is that true, Brian? That is correct. Okay. So they have to do an actual experiment in the lab, a wet experiment. Okay. And they complain that it requires laborious generation of T cell lines expressing specific TCRs for each epitopes. So they can't do a lot of them. So they have to narrow their focus. I would personally never write laborious in a paper. I would say it's unreasonable or beyond the scope. Impractical. Impractical. That's good. Laborious is a value judgment because you can always find someone who will do it.

1:04:13No matter how crazy it is. So they make T cells expressing a specific T cell receptor for the known epitope, for the known T cell epitope. Okay. And they just do a few here. And they generate three T cell receptor transgenic lines. And they ask whether the changes that do not disrupt HL. So they have some changes that seem to be enriched in certain areas, but they don't disrupt, at least they don't predict to disrupt HLA epitope binding. So they say these are candidates for disrupting the T-cell receptor binding. In each of the three T-cell receptor line, they had three mutations with geographic evidence for escape, but not reduced HLA binding exhibited reduced avidity or complete loss of T-cell receptor recognition of the HLA peptide complex.

1:05:08So they say this supports the idea that for a fraction of these T cell epitope changes, the escape is mediated by disruption of the TCR. So I think it's a very small data set and it's suggestive. Yeah. So I've never seen this assay done this way. I've seen it done the other way to look at HLA binding. Yeah, right, right. But I haven't seen it done to look at TCR binding in this way. I mean, the overall shape of the graph looks like the graph I'm used to seeing for HLA binding. And I'm having trouble finding exactly how they got – like what machine they used to measure affinity. Like I don't – I'm having trouble finding that.

1:05:57But that's likely a me problem. All right. The last thing I want to talk about is they have this section, evidence for selection for T-cell escape in the general population transmission chains. So when does the selection happen? It may happen within host in persistently infected individuals, and this has been described before, but it could also occur in the general population, not just in persistent infections. So what do they do? They compile a data set of SARS-CoV-2 evolution in persistently infected individuals. This is defined as being infected for more than 21 days, and these are all cases in the literature.

1:06:43And there's a data set of 1 ,054 amino acid changes. This is one thing I have to point out. They always use mutation incorrectly in this paper, consistently over and over again. 1054 amino acid changes that were inferred to have occurred during persistent infections. 673 of these fall into the HLA class 1 epitopes. Okay. And then they try and look at country-specific frequency. But they say for most studies, we had no data on host HLA composition. So they use instead country-specific, because they know where these people were, of the cognate HLA alleles, and they didn't find any evidence for stronger selection in these persistent infections than they have before.

1:07:37So they say this is inconsistent with selection taking place only in the persistently infected individuals and instead suggest that selection occurs in the general population. So it seems to me that, all right, you couldn't find it here. So you're saying it must happen somewhere else. It's kind of weak, right? Right. That presumes that their hypothesis is correct, that the selection is occurring. Yeah. And this is where the problems really come in. So let me just – let me – before you go into it, Brian, let me just – I'm going to use the commentary just to review some of the things because it's really clear.

1:08:15It's actually clearer than the paper. So the authors collect 1 ,450 experimentally supported CD8 epitopes and predict how observed amino acid changes altered binding to HLA. Across changes in the known epitopes, 35 % of the lineage-defining mutations for VOC and 39 % post-Delta and Omicron were predicted to impair HLA binding. These high proportions are suggestive of selection, but not convincing enough on their own because T-cell epitopes cover most of the proteome, and some changes can increase fitness for reasons unrelated to HLA. That's the commentary. And then they say the stronger evidence comes from asking which HLA molecules the changes evade.

1:09:02HLA frequencies vary between populations if it correlates at some support. So if escape from a given HLA benefits, the virus mutations that disrupt this binding should do better where the HLA is common in the population. And this is what the authors find. The most frequent binding loss mutations associated with common HLA alleles were about 1.7 to two times as frequent as those associated with rare alleles. And many changes were more frequent in countries where the cognate HLA was more common.

1:09:43Okay, so a simple calculation, this is interesting, a simple calculation gives a better sense of the strength of HLA selection. If a mutation has a 10 % advantage in a matched immune environment, which it encounters in 20 % of transmission, then the average fitness advantage is about 2 % in the absence of anything else. That's enough to increase the frequency over many transmission generations without making a big sweep like the antibody resistance does. So unlike lineage replacements associated with antibodies, you have the whole variant replaced very quickly. HLA adaptation may therefore consist of small geographically structured changes.

1:10:24Okay. The paper also shows why reduced peptide HLA binding is not the host story. And this is the T cell receptor experiment. They say this was a small experiment, but the observation is compelling as it shows a population-level association predicted. And then, yes, so they asked, can selection during persistent infections alone account for the population-level HLA escape? Across four different data sets, they found that changes from persistent infections were no more strongly associated with common HLA alleles than those accumulated during sustained transmission. So persistent infections can't explain everything.

1:11:04Okay. So they say a potential next step would be to estimate fitness effects across different host populations and ask whether they vary with the local frequency of the cognate allele. So maybe fitness effects of the changes are actually what's going on. And finally, they do not replace antibody escape as a major contributor to SARS-CoV-2 evolution. They say this is just another component. And they say the next challenge is to test whether measurements in defined HLA and CTC receptor systems can predict the effects of individual escape mutations. And that's the functional assay that they don't have here.

1:11:44Anyway, so they're a little more positive. They're somewhat positive, but say that we need to have more experiments. Okay, Brian, now tell us what you think. So as I was reading this paper, I was cooking on all of the references where they talked about HIV because this was tickling sort of something back in the cobwebs of my brain that we had talked about a lot when I was a graduate student. And I specifically remembered a there was a paper that came out when I was a graduate student that tried to do the same thing for HIV and tried to show the same types of effects with HIV. at a population level kind of similar and came up with a similar kind of conclusion.

1:12:49It was published in Science in 2002. I found it.

1:12:57And again, realistically, what this is showing us is that there has been evolution. There has been a change in allele frequency over time. And it hypothesizes that that evolution is due to selection, like changes in fitness due to evading T cells. But it doesn't show us that that evolution is happening because of the natural selection driven by T cells. And in basic evolutionary biology, there are a few different reasons why gene frequencies change. that, you know, for a few different reasons why evolution can change. And some evolutionary biologists took the data in the paper that I mentioned and did some further analysis on it.

1:13:47And I will admit that all parts of this analysis I don't understand. But when I look at who has been involved in this, they are all people who I strongly trust. And basically what they found was that when you looked at the original data, all of it was due to founder effects in the viruses that were looked at in the population. And so founder effects or genetic drift is, in fact, one of the recognized mechanisms of evolution, of change in gene frequencies, but it's not natural selection. And so when they redid the analysis where they could actually track virus that was transmitted to people, and so they did what we called looking at transmitted founder analysis.

1:14:39When they looked at transmitted founder analysis to actually see if they could see selection happening or if this was a genetic drift thing, they found out that it was entirely genetic drift. um and so um and that was published in 2007 um and um a bunch of these analysis were done on a patient cohort in perth um western australia um and you can and all i remembered going into this was perth one scientist's name and something about founder effects but i i somehow managed to find it And then if so, that kind of took away a lot of this idea in the HIV literature. They also in these papers talk about one other group that has talked about this in the HIV literature.

1:15:27And that group and the reason why that group has some interesting data is because they look at a specific HLA type and a specific epitope and look at them in populations. And so instead of trying to look at this in a very broad level, they actually try to show that it happens with one particular HLA type, which is exactly the information that the commentary and that Vincent mentions, this group is missing. Right, right.

1:16:25So they don't have the phylogenetic context added, which these other researchers sort of indicated, at least in the HIV field, was critical. Yeah. And this includes people like the senior author on that paper was Betty Korber, who many of us are familiar with. And so I don't see any of that in the PNAS paper. and I don't see any of the specific work on individual HLA types and individual epitopes. And so without that, I'm not sure that this says anything more than the gene frequency changes over time. Maybe somebody should look at that. But what about this observation that certain T cell epitopes are favored in countries with certain allele frequencies, right?

1:17:18Yeah. And so that's – I think that that is the strongest thing that they have. But I think that without the phylogenetic context to show that it wasn't a matter of different viruses being introduced into those populations, that you can't fully make the conclusion. So we don't capture HLA type when we sequence virus. So we get a clinical sample. We could get HLA type from it, right? We theoretically could. I tried to figure out once if I could get my HLA type figured out. And at least one can send away to get part of it, but not very well. We could do it in the lab. I mean, we can do it in the lab.

1:18:04It can be a little complicated. When you took a nasal swab, you do the genome sequence and then do something else to get the HLA type and store those data together. And so that's what some of the other groups have done that showed it for the sort of chronic HIV situation. So what is the story with HIV? Do variants spread in populations with the right HLA? There are a couple of really, really, really dominant HLA types. Right. Where you can start to see some effect. But it's really only seen in those very particular, very, very, very strong HLA types. Yeah. So they cite a paper here showing that certain T-cell epitopes follow geographic HLA frequency.

1:18:55In B57 and B27, yeah. And that's 2009. adaptation of HIV to HLA class one. But then they also show that when they move into other people of a different HLA type, the change reverts. Yes. And so the group that did that paper looking at B27 and B57 is also the group that did the paper I mentioned about the mother-child transmission pairs, because those two HLA types have such a strong T cell response that goes away with this amino acid change, that it's a particularly good system of study. It's interesting that the change goes away in a different HLA type, which tells you that the change is a fitness cost, right?

1:19:41People have actually done the structures of these changes in the proteins and shown the fitness cost. So even in the case of HIV, which is kind of the ideal setup for this type of adaptation to occur, it is far from a slam dunk. Mm-hmm. And COVID-19 is a short-term acute infection. Yeah. Yes. But you have bigger numbers, right? You have bigger numbers. Anyway, I thought this – I don't know what – I mean, more has to be done. It's kind of suggestive. More has to be done. But, you know, I teach in my class that T-cell escape is only for the person, but that's too binary. I think it has to be softened up a little bit because there are some examples, like HIV is a good example of that, right?

1:20:30Yeah, but I still teach on that same pretty binary. Yeah, I do. Okay, then I will too. Because for the level of undergraduate students, that's the level that they're thinking about things. And that sort of helps give them a clear understanding of what's going on with immune evasion and fitness and MHC class one presentation. And you don't have an additional several hours of lecture time to explain the nuances that we've just discussed.

1:21:05Not that we took several hours on it, but we kind of rushed. And that's actually why I do like that mother-to-child transmission paper is it shows that in a very specific situation, it can really matter. But outside of that situation, I'm not sure that we see the same level of effect. Okay. Ellen, you just came up with a new term to evaluate scientific literature. FFASD, far from a slam dunk. from us.

1:21:43Let's do a couple of emails. Let me take this first one. It's short from Anthony. He sends us an article. First in Brazil, black-faced lion tamarins are vaccinated. Huh. Oh, these are cool-looking monkeys. I don't think I've seen one of these before. But what are they vaccinated against? Would it be like a Flavie? Yellow fever. Really? They get yellow fever? That's what it says at the top. Oh. Huh. They're an endangered species.

1:22:19Yeah. Yellow fever vaccinations. Huh. So what do they do, capture them? Yeah, they catch them.

1:22:30It's quite a striking animal. Yeah. It's golden with this black mane. It's really cool. That's neat. Looks like you want to be an orchestra conductor.

1:22:45They're pretty small. Look, they have a picture of a lady holding one. Yeah. Okay, that's pretty cool. Very cool. They look like something out of a Dr. Seuss story. All right, thank you for that.

1:23:01Let's see. Brianne, can you take the next one? Sure. Ari writes, Hi, TWIV team. Love the recent episode. A Cauldron of Bats. I happen to work in drug discovery at a company that studies aging, and for any of your listeners who are interested in learning more about the peculiar longevity of bats, I highly recommend Stephen Ostad's book Methuselah Zoo. Fun fact, humans also break allometric scaling expectations for lifespan, but most species of whales do not. It just so happens that I blog about books for fun and spent this summer looking at popular press books about aging specifically. I've got a summary of Ostad's book, along with several other popular treatments about the physiology of aging on my website, Unsequenced Life.

1:23:46Please keep up the good work, Ari. P.S. My first industry job after leaving an academic career as a developmental neurobiologist was assay design for point-of-care diagnostics for infectious disease. Throughout my undergrad and PhD training. I had never taken a course on microbiology or virology, and the microbe.tv podcasts were an invaluable resource for filling in that gap in my education. Cool. Cool. Very cool. Thanks, Ari. I put the book on my list. Alan, can you take the next one? Okay. Sage writes, Hello. It is a chilly 7C here in Western Massachusetts this morning. I just finished the most recent episode of TWIV while splitting firewood to heat my house.

1:24:28The story on bats had me enthralled. I teach microbiology at Greenfield Community College, and I recommend TWIV and TWIM to my students. Thank you for all the work you do. It's such a gift to hear experts discuss current research. Thank you, Sage. Do you know this Greenfield Community College? I don't know the college, but I certainly know the town. Greenfield is a little up in I-91 from me. Well, there you go. You got some TWIV fans in your neighborhood. Rich? Felicia writes,

1:25:27On the history of the mumps vaccine, note that the Smithsonian has preserved some of this history. And I thought I would pass it on. And she links to that article. It's the development of the vaccine. It's the Maurice Hillerman and his daughter story. um this um uh deal about periodic uh mumps outbreaks is in my mind i don't i don't recall the details but in my mind this is it's a little different than a measles situation this it's like uh immunity may wane or something like that it's not as airtight a vaccine as measles put it that way. And so if you, in particular, put a bunch of guys together in a football fraternity in college, so there's a lot of closely associated individuals, you tend to get these outbreaks.

1:26:34One of my ex-students used to work on this at CDC, and it was a real puzzle for them. So it happens down then. Yeah. So we don't know anything about the vaccine status of the players, do we? I would have to look at these links. I don't know. Yeah. I mean, if the school requires the vaccination, so they confirmed 14 cases of mumps on the team, which, you know, I don't know, it's probably a few dozen players on the team altogether. That's a significant portion of the football team that seems to have mumps, right? It makes sense because they're in the locker room, right? Yeah. And some of this is coming back to me.

1:27:22I'd have to look it up. But I think this can happen in a vaccinated population. Yeah, I know that mumps immunity wanes. Yeah, this sounds like a waning immunity problem. I mean, if I remember at some point early in the development of this vaccine or when it was first released, they started seeing mumps cases, so they put a second dose in. But they also did that with MMR, right? Because it seemed – well, maybe that's because of the mumps component also. I don't remember. I think part of the deal with MMR, sometimes the second dose is to make sure that everybody gets at least one. Yeah. Right. Okay.

1:28:04All right. Patience writes, hello to a long-time listener, first-time question asker here. My question is this. Which viruses are the most beneficial in history since they can't all be bad? Thank you for all the work you do in increasing accessibility to amazing biological science. Horsepox.

1:28:28That'd be on the list. Very good. Okay, Alan's reference there is that horsepox presumably was the virus that Jenner used to vaccinate people against smallpox. So horsepox indirectly eradicated smallpox. And what I love about that idea is humans probably saved horses from extinction. Right? I mean, there's some data in the archaeological and geological record that equine species were in decline at the time of humans. And it doesn't seem to have been entirely our fault, unlike a lot of species that went into decline when we were around. There was a population of horses in North America that vanished.

1:29:17And they didn't come back until the Europeans brought them from the old world, the domesticated ones. and there's a theory that horses might have just vanished if humans hadn't discovered hey you can use these to pull things and ride them and we're going to breed them and then the horses returned the favor with a disease that they'd carried probably for a long time that's cool which now they no longer carry as far as I know that eradicated itself could we say that aquatic bacteriophage in general are beneficial because they play a big role in the carbon cycle right yeah that yeah that's really good see i was thinking more along the lines of whatever virus was responsible for syncytion or the viruses that were responsible for um the vj recombination rag transposons or even i remember in a couple weeks ago, we talked about a phage virus that had competition things that we thought was maybe the roots of all immunity.

1:30:30I would say some of those would be on my list as well. Because without syncytion, we don't have placental mammals, right? Right. And without RAG1 and RAG2, we don't have adaptive immunity. Yeah.

1:30:44So, without placentas, you think humans would not have appeared? I don't know. Or any of our ancestors. We wouldn't be like we are now. Yeah. I don't think we could have as large a brain without the amount of gestation that we get. Right. Maybe the cephalopods would have taken over instead. On the cynical side, I'm not convinced that anything that promotes human population is beneficial to the planet. It's true. Okay.

1:31:25Who read the last one? Did I? Yes, you did. Yes, I did. Okay. Back to Brianne. Diane writes, Dear team, my name is Diane Ducre. Until COVID, I was primarily known as a French bestselling writer. My books have been translated in around 20 countries, sold more than 1 million copies. Long COVID profoundly disrupted my life and unexpectedly turned me into a scientific researcher. I began investigating some of the biological mechanisms that might underlie neurological long COVID. That work has now resulted in a first peer-reviewed scientific publication, recently published in Viruses. In our respective cohort, more than 75 % of patients with neurological or neuropsychiatric long COVID tested positive for HHV-7 by PCR, and gives a link.

1:32:20It is, of course, a small cohort and a study with limitations, but I believe the signal deserves further investigation, particularly as HHV-7 has received far less attention than EBV in discussions of herpes virus reactivation and persistence after COVID. I also thought the story behind the paper might be of interest to you. One of the striking consequences of long COVID has been the emergence of patient-led research on an unusual scale. The disease affected large numbers of active, often highly educated younger people, many of whom were unwilling simply to wait for science to catch up or for their illness to be forgotten.

1:32:58Patients have become data collectors, collaborators, funders, organizers, and in some cases, researchers themselves. Best regards, Diane. Well, so she's listed on this paper as an independent researcher, which was the same in the HLA paper we just did. And I've never seen that before. Now I see it twice. How? How about that. And I just, while you were reading that, I looked up Diane Ducre and you have, looks like you've written some pretty cool books. I'm going to have to look into the translations of those. Fiction and nonfiction, including some interesting histories of, there's one of the wives of famous dictators and another, yeah.

1:33:41So it's cool stuff. Cool. It's very interesting.

1:33:48Alan, can you take the next one? Yes. Let me get off Wikipedia and join you again. Anthony sends a link. Insights into longevity and virus-driven adaptation from myotis bat genomes. That's the one we did. We did that paper a couple of weeks ago. Yeah. The positive correlation with body size of lifespan, otherwise seen across mammals and other metazones. Anthony says, at a glance, this statement appears obvious, so it's repeated on and on. Under scrutiny, the correlation between size and lifespan all too often is either meaningless or not true. As an elephant takes 16 years to mature, of course they live longer than that.

1:34:23Lab mice are very old at two years. The similarly sized deer mouse lives for five. Canaries generally live longer than 10 years. The larger budgerigar rarely lives that long. Large parrots easily go past 50 years. Poultry and waterfowl, not much past 10. And for other metazoans, sea anemones live decades. And gives an example. There's speculation that with reproduction by division, for sea anemones, death might be irrelevant. Thank you. Anthony's always got good tidbits. Always got interesting insights. Good tidbits, yeah. Yes. So positive correlation with body size of lifespan. He's saying basically it's obvious.

1:35:09Yeah, he's saying, nah. It's either obvious or it doesn't apply. Right. Intuitively, you look at it, you say, oh, yeah, an elephant lives longer than a mouse. But then you say, yeah, but why does an animal the size of a mouse live twice as long and another animal, a bigger animal, doesn't live as long? Okay. Rich, can you take the next one? John writes, TWIV team, I thought you did a great job of explaining the complexity of the nature paper, the anthropogenic fingerprint on emerging infectious diseases in the last episode. It reminded me of one of my favorite books, which deals with this subject in a more historical and cultural point of view, Paul Farmer's Fevers, Feuds, and Diamonds on the Ebola Outbreak in 2014.

1:35:56It is a truly amazing book and explains how the human institutional interaction with infectious disease outbreaks is as important as the science. all the same as that TB book that I referred to where he calls TB basically not an infectious disease, a human disease that humans make it. At any rate, my lab is currently working to map the current DRC nightmare, which will probably top the cases of the 2014 outbreak. Here's our latest map, and he gives a link. you ever want to add someone to your team who deals with all of this from a more geospatial and geographic perspective i would love to participate i have recently retired from johns hopkins and the library of congress where i was a specialist in computational geography and gis and advised congressional members and staff of infectious disease spread it uh you might be interested in what I do at Biomap and gives a link to that.

1:37:04All the best, John, who's a lecturer in applied mathematics and geographic information service at Johns Hopkins. Very cool. Thank you, John. So if you hit that first link, you get a movie showing the spread of the outbreak across the country, right? And then you get the epi curve. It's very cool. You see how it's spreading. First complete map of the old world fruit bats. That is very cool. Neat. Good stuff. I'll take the last the next two. So Lisa writes, I wanted to share a fun detail. I watched the new Theranos documentary. You could see everything in the theater. And Billy Elizabeth Holmes' husband is listening to your virology lectures in the beginning of the movie.

1:37:52I recognized your voice immediately. And Brendan writes the same thing. I'm 99 % sure you could hear Billy Evans listening to the podcast and you can see everything. Probably not the listenership you're going for, but funny nonetheless. Thanks for a great pod. I have listened since COVID and have learned a lot. James is a presidential professor in the Department of Government at Dartmouth. There you go. We have a wide reach. Did anybody see this movie? Not yet. I haven't either. I don't think I will because I don't really want to see movies about dysfunctional people. Yeah, I usually don't get to those.

1:38:30But it's interesting that he's watching my virology lectures. Apparently, he's watching at 2x speed. Oh. Okay, let's do some pics of the week. Brianne, what do you have for us? So I have a space story that I came across earlier this week. Um, that's, uh, it's called astronomers say they've discovered the first green star. Um, and so there is an astrophysicist who actually, um, took a look at, um, a number of different stars. And instead of looking at sort of the general, uh, sort of spectroscopy across a bunch of different wavelengths, they looked specifically at what all of these stars would look like to humanize instead of, you know, to computationalize.

1:39:24And they found that white dwarfs, unlike other types of stars, white dwarfs can have a lot of different colors. And they were able to actually describe that, you know, normal stars are sort of white, orange, blue, but white dwarfs can be all sorts of colors. So they've got, blue ones and yellow ones and orange ones and pink ones and green ones and all sorts of things like that. And so they actually did some looking within all of the white dwarfs within 65 light years of Earth and kind of figured out what color they would actually be to human eyes. And then they looked at what elements were involved and things like that.

1:40:10Unfortunately, we can't actually see any white dwarfs from Earth ourselves. And I'm a little sad by that because I was imagining that if we were on a different planet in a different area of the universe, perhaps that means that when we looked in the night sky, we would see like a multicolored display of all of these different colors. So how do they see these if you can't see them from Earth? They are using telescope and spectrum data. But you can't see them with human eyes. Naked eyes. Okay. Okay. You can see them in the right telescope. Okay. So what is a white dwarf? So apparently it blows anything that goes by it.

1:40:48It rips up. Yeah. So basically once the sun – we talked about before the sun and the sun sort of using up its hydrogen and making helium. Yeah.

1:41:03Eventually you run low on hydrogen. And as a result, you're not making very much energy. And suddenly the mass of the star collapses on itself because it doesn't have energy sort of pushing out anymore. And so there's a whole set of evolution of stars based on this collapse of when they're doing fusion reactions and when they actually run out of energy and collapse and how much energy is left over and how big they blow up. And so white dwarves are one of the stages of all this. The white dwarves, I think, are kind of an endgame stage for main sequence stars. But you don't have to worry too much about the sun doing this because before it does that, I think it will expand to a red giant, which will likely encompass Earth's orbit.

1:41:56At which point everything, including the planet, will just be completely gone. Which means we never get to see it be potentially green. We never get to see a green star. No. By the way, this image that's an artist's impression of what this should look like was on the same day that this was published, the APOD picture. Yes. Yeah. So the sun is going to do this in like 5 billion years, right? Yeah. Quite a while. Yeah. So do you think humanity will last that long? No. No? Why not? What do you think will happen? Oh, well. I guess there's going to be other stuff going on. in our planetary environment that's going to take care of us first.

1:42:40Oh, we're not going to take care of ourselves, you think? Well, that's entirely possible. That's entirely possible. We are the biggest danger to us. I mean, look how much life has changed on Earth in the last 4 billion years. Yes. Extrapolating a similar rate of change for 5 billion more gives us... Look at what we have done in like 200. It just boggles the mind. Well, I mean, in a million years, we'll probably be very different from what we are now, right? If we're around, yeah. I mean, it's so weird to think of the whole solar system being gone, right? But it's almost hard to wrap your head around.

1:43:25Anyway, Rich, what do you have for us? I have an article from a Johns Hopkins magazine from the Johns Hopkins School of Medicine called Inside the Johns Hopkins Psilocybin Playlist. Where do you get this, man? Where are you looking at Johns Hopkins Medicine magazine? I'm glad you asked. There is a choral group in town that I really like. They're called Austin Cantorum. Okay. And they do all sorts of just wonderfully playful and serious shows. They actually rehearse in the same space that my group now rehearses in. And in fact, the reason that we're rehearsing there, we had to move recently, is because of them.

1:44:18Longer story in the background. But I become a fan of these guys. And they recently put on a show that was titled the Johns Hopkins psilocybin playlist. And what this is, and I went to the show, okay? So what the Johns Hopkins psilocybin playlist is, is there is an area of research in many places, including University of Texas at Austin, where researchers are looking into the possible therapeutic benefit of hallucinogenic drugs like psilocybin. And this has been going on for decades. And over time, seated to some extent at Johns Hopkins, they have developed music that research subjects listen to during their experience that's tailored to different stages of the experience and et cetera.

1:45:18and this has become the Johns Hopkins psilocybin playlist. How did they pick? It's pretty subjective. If you look at this article, they'll give you down at the bottom. It's staged, okay? When you're a research subject, you're in a comfy room lying on a bed with an eye shade and headphones, and your experience can last up to like seven hours. And they phase it. There's the sort of when you're going up, there's the sort of relaxation phase. There's a peak phase and a coming down phase. And, you know, they over time, probably partly by trial and error, picking stuff and getting patients' reactions to it, developing this playlist.

1:46:19So obviously the concert I attended wasn't seven hours long, but they took samples of this. And these are, I presume, though I don't know, that a lot of the pieces on the playlist are orchestral or instrumental. Okay, this was all vocal renditions of this. This is an a cappella group. But actually, they preceded this show with a lecture by a UT researcher on the history of psychedelics and a history of research into their potential therapeutic value that was very entertaining. Don't they use low doses so you don't actually hallucinate? Not necessarily. Not necessarily? No. Okay. I think they can use different sorts of doses, but they'll dose you right up.

1:47:14Okay? You'll go on a significant drip. Okay.

1:47:20I've never been a patient, put it that way.

1:47:27So you're told. I thought this was extraordinarily amusing, and the concert was great. This is cool. And they link to a Spotify playlist, which I was just scrolling through. It is overwhelmingly instrumental orchestral music. There are a couple of other. It wraps up at the end with some vocals and the final song on it is What a Wonderful World by Louis Armstrong. Which actually we sing regularly at our shows because we just did one of these recently because that song is just – that song was written in the 60s as sort of a response to all of the turmoil and stuff that says, wait a minute, okay?

1:48:14There are good things here, okay? Just chill out and have a look. It's a great song. that's neat Alan what do you have for us I have a book pick so this is I think Rich in particular will probably like this I've already ordered it on the library this is great this is Eric Dolan is the author the book is The Wreck of the Mentor and no this is not a book about getting your thesis advisor drunk the mentor was a whale ship out of New Bedford, Massachusetts. In 1832, they were in the Pacific, which was a pretty standard place for a New Bedford whale ship to be then. Charts at that time sucked, and they'd been under an overcast, so they couldn't get celestial sights, got lost in storms, and hit an uncharted reef.

1:49:14So normally that would be the end of a short story that nobody would ever hear about, because the crew would all be gone. But they survived. Eleven crew members survived, got onto this little atoll that they'd hit and salvaged some bits of the place. And very soon they saw several canoe loads of indigenous Micronesians approaching them. And what followed was a couple of years of, actually several years of a collision of worlds. So it turns out they had grounded in what we now call Palau, which at that time had been seen and visited by some Europeans. And there were even a couple of so-called beachcombers who were living there who abandoned their ships and decided to live with the natives.

1:50:11but they get enmeshed in political situations that are far beyond their comprehension and much more intricate than Americans might have expected from these so-called primitive people. Things don't all go well and they don't all quite make it but enough survivors get out to have left a very interesting story of this whole interaction. And what I really liked about the book is that even though this is a really hard thing to do, Dolan really tries to incorporate the Micronesian perspective. They didn't have writing, so their oral traditions are what you'd have to rely on, which aren't going to provide the same level of detail as a captain's logbook.

1:50:57but there are enough clues and there's enough information that he's able to piece together what was likely going on and who was in charge of which village and how the conflicts were developing at the time the ship came. And it's just, it's a really cool snapshot of a time and place. But they landed on this Palau where there were people living, right? Yes. And Palau is a, it's a series of a whole bunch of islands. It's an archipelago of a whole bunch of islands. And they, being Micronesian, similar culture to the Polynesians, they're seafaring people. They didn't have writing or the wheel, but they had the most advanced navigation the world had seen up until that time, really.

1:51:45So they would conduct business over hundreds of miles of open ocean, but they hadn't had interactions with the West very much and they had their own villages that, you know, sometimes unified, sometimes not and sometimes went to war with each other and of course these guys landing who brought a bunch of weapons and gunpowder and such from their ship and steel tools became a source of conflict in their own. So how did they get off the island in the end? Eventually they were rescued by another, ship, actually a couple of ships, picked up one person and then saw the natives on the beach were having some conflict over whether to bring anybody else out and the captain of the ship, who only had about a dozen guys on board, said, we're just going to move along now.

1:52:38And so the crew members who were still on the island stuck there watched as one of their guys got away and their ship sailed away and they were stuck there for quite a bit longer. That sucks. Yeah. This sounds great. They were eventually, a handful of survivors were eventually picked up and got home. They must have been bummed out being stuck on that island. There was a lot to be bummed out about. And then they want to kill them. So they were initially welcomed once some initial discussions happened. Like, oh, wow, you got all this stuff. And they said, okay, let's see if we can make a deal. We give them some tools and weapons and they help us build or repair a ship and we get off this place.

1:53:23And things were going okay, but then there was this conflict going on between a couple of tribes and they eventually ended up being taken to another tribe which actually enslaved them. And they lived as enslaved individuals on a rather destitute island where even the people who were native to there didn't have enough to eat And, yeah, bad situation. Wow. Really uplifting story. But it was a fascinating, I mean, yeah, a horrible thing to happen, and it makes a great story. Okay. All right. My mind is a little more lighthearted. Mine is a very short YouTube video about the Apollo 11 code. and this guy goes through it and there's less computing power than on your thermostat basically and this guy took the code which is open source and he ran it and on an emulator and it's just full of interesting things like like um how little ram the ram was was fixed you Because the RAM were cores with wires going around them.

1:54:44If the wire went through it, it was, I don't know, zero. If it went around it, it was – so you had to fix everything in memory and it was sewn by people. And then the lady who wrote the code, I mean, there's just so little of it. But the code comments have been put – so this whole thing is on GitHub. Yeah, right. And you can see the source code, which has comments in it. Yes. Which are the hilarious part. One of them is like, we know there's no second chance. But yeah, it's pretty funny. This guy goes through it and explains how simple it was, yet it worked, right? Everybody got there and back, which is amazing.

1:55:30So it just goes to show, and I think the lady who wrote this just died not too long ago. Yeah, September 30th, 2026. age 90 years old Margaret Hamilton so she was about 30 in 1966 I think she started writing us in like 67 or something like that man she did the whole thing unbelievable well not unbelievable very cool it's cool and it worked it just goes to show that you can do things with simple code, right? It doesn't have to be millions of lines. Anyway, that's TWIV 1369. You can find show notes at microbe.tv slash TWIV. You can send us questions or comments or picks of the week to TWIV at microbe.tv.

1:56:24And if you enjoy these programs, we'd love your support. microbe.tv slash contribute. Rich Conda is an emeritus professor, University of Florida at Gainesville, currently in Austin, Texas. Thank you, Rich. Sure enough. Always a good time. Dixon had a good time, too. He's wide awake now. Breanne Barker is at Drew University Bioprof Barker on Blue Sky. Thank you, Breanne. Thanks. I learned a lot. Alan Dove is at alandove.com, turbidplac.com. Thank you, Alan. Thank you. It's always a pleasure. I'm Vincent Racaniello. You can find me at microbe.tv. I'd like to thank the American Society for Virology and the American Society for Microbiology for their support of TWIV, Ronald Jenkins for the music, and Jolene Ramsey for the timestamps.

1:57:12You've been listening to This Week in Virology. Thanks for joining us. We'll be back next week. Another TWIV is viral.