 Well, good morning. Good afternoon. Good evening everyone wherever you are around the world. Thank you for being here Welcome to the symposium on aging gyroscience and longevity So as you know Aging it's a greatest risk factor for most of the major human diseases in developed countries and the research that you guys are doing has just been amazingly like transforming the state of what we know about aging and longevity interventions And so I'm oh, I'm Jessica Tyler a senior editor at elive I am a chromatin person, but I also do some work on aging and I Felt that elive would be a wonderful venue for the aging fields where everyone In biology could gain access to the great work that's going on in the aging field and elive Would be a rapid fair and transparent review Journal for the aging field. So I really wanted to bring Elive to the attention of the aging field. So that was the reason behind doing this special issue and we brought on matt cablin who you all know is A leader in the aging field not only that he's been an amazing partner in this Special issue effort that's been going on for the past year nearly We also brought on eight amazing guest editors from around the world I'd like to thank those wei wei dang from baila usin su from columbia university Dario ricciardo valenzano from the max plank institute of biology of aging in germany Jingdong Jackie Han from peaking university in china veronica galvan at the uti health san antonio Pankaj kapa he from the book institute jan grubber from yale and sarah hag from the carol institute And together we have handled 159 submissions for the special issue on aging 21 have been accepted 29 are still under consideration and what this symposium is going to do Is really to showcase the work From selected accepted papers that will be coming out in the special issue And with that i'm going to hand over the first session to matt cablin Great, okay. Thanks. Thanks jess. Um, and uh, and I also just want to uh Uh thank jess for organizing this special issue and and For taking the lead in uh In editing this so when I started at elife jess Jess has been a senior editor at elife For a while and I came in as a new kid on the block and now I have moved up also To senior editor at elife and I and I just want to um emphasize just a couple of things that that jess Said which is that I think this special issue has been a great way to Bring elife to the attention of the the aging field, but I think the reverse is also important in that um, you know, one of the goals of having this special special issue is to um To to bring the aging field into greater awareness in the broader scientific community And I think elife is the perfect forum for doing that and I this special issue has exceeded my wildest expectations in terms of The number of really high quality papers that we've gotten across the whole diverse range of the the biology of aging and I hope that's something that you will appreciate from the talks today is Is indeed how how diverse they are in terms of Aspects of aging that are being studied the model organisms that are being employed Um, I think this is such an exciting time in the field. And so I think this this has been uh, just a really um outstanding uh experience uh doing this special issue with with jess and all the staff at eddie life and so Uh, I think there's no reason not to to kick it off and get started um, our first speaker uh of the day is Peter Fedachev who is the co-founder and ceo of jero Peter, uh, is now based out of singapore and he's going to talk to us about uh, his paper on germline burden of rare damaging variants negatively affecting human health span and lifespan And it looks like he's already got his slides shared. So take it away. Peter Well, thank you very much for for the introduction and actually for the journal issue. So it was a pleasure to participate and To contribute Well, I will tell you about the work with such a long name Which which is in fact very simple inside Uh, the work was actually the idea of the work was actually suggested by Vadim Gadoshov Can I can can I know that everything is fine? Yeah, cool. It looks good. Cool. Thank you So the the idea was suggested by Vadim. Uh, we uh, we have been working on genetics of human longevity For a while with a number of people and then uh, suddenly Vadim approached us and uh, and asked uh about The possible effect of very rare mutations on human longevity. How much of human longevity genes we actually see When we investigate common gene variants Well, we know although with some of us doubt after some recent calico work that quite a lot of variability of human longevity can be genetically Predefined Nevertheless, uh genome white association studies of human longevity did not return many gene variants associated with longevity They well return quite a lot, but the effect size is quite small So I would say that the amount of dark matter and the human longevity genome still still watch So, uh, maybe we were thinking rare protein variants which are not which are not obvious from standard genotyping procedures I implicated in aging as well. So if if yes, what is the effect? And uh, fortunately over the last years Whole examine now whole genome sequencing sequencing are taking up We have now uk biobank Which had uh at the time of the work Published had 50 000 fully exomes available for consideration. They have just recently released another 150 000 Genomes whole exomes, which means that we're slowly entering a situation where we have now access to ultra rare mutations So what we have done together with vadim and his colleagues Is that we actually Started the effect of the mutation frequency Well, the phenotype to aging so we looked at specific gene variants, which are Protein truncating variants those which are kind of focused associated with the loss of function for the for the for the protein product Or the absence of the of this product and we've been to them according to their frequencies. Well, each of us is a working mutant We have I don't know, uh, maybe 60 or 80 60 Genes which have protein truncating variants and it is one copy of the gene But as you can imagine as we are getting into less frequent Variants we have less and less of them, but each of us has about six Ptv variants, which are very rare less than 100s of a percent so Most of them it is a number of data set where supplies donor acceptors and stop codon Gains, so what was the relation we asked ourselves? Between the number of such variants the burden of such variants and human health and longevity Well, and the answer and the answer was pretty straightforward So we decided that if those variants are all they sound that is bad protein truncating variants So hopefully they have the same sign which means that the more you have the less Years of life you may have and this is indeed the case as you can see the most rare Variants the number of them the total number of them in the genome was negatively associated from the cox proportional hazard model Point of view with the remaining lifespan in uk biobank. I will tell you about that But also with the health fund in uk biobank So here we use two different components of uk Biobank we used the The overall data set, which is the well we used 40 000s genetically british individuals to make sure that scoring some artifacts of ancestry and Among those about one thousand of them died within 11 years full of time But on top of that we also created other popular phenotypes used in Human genetic longevity research. We also used health fund, which is the age of the first chronic disease But also mother and father and this because we are sharing half of our genes from two power parts As you can see, there is a more or less consistent picture in all phenotypes we have seen the total number of presumably deleterious variants Where ultra rare deleterious variants were associated was longevity in the negative way If you look specifically at specific type of mutations, you would see that those top game Where let's say the most consistent features of the association between the ptv variants and human longevity Well, the the effect is not small by the way. So we looked at At the first and fifths Or not quantities, but I mean we looked at the people with the less number of ptv rare ptv variants and with the most number of the rare ptv variants And this is just the survival plot within the full uptime and as you can see the result is not small People who have more rare mutations do indeed leave less and die faster Well, I think the well, this is kind of expected although this is still far Far far beyond the expectation. I mean, we still do not know where the rest of a lots of percentage of lifespan variants Which is presumably genetically Determined resides. Nevertheless, it's it's the step in the right direction But the same result. I mean the effect size of the Of ultra rare mutations Can with respect to To longevity the effect on longevity can can help you answer another important question So we always think what is the role of of mutations spontaneous mutations on lifespan And here we have a glimpse of what it can how this can work Uh, we know well from the publications from literature that there is a certain spontaneous mutation rate in our body So in every cell in our body, there are spontaneous mutations and some of them are those ptv mutations So you can think for the genome of our size size How many of those ptv mutations are produced every year and then you can multiply it by effect and then Correct for the conference mortality model and you can see that the expected effect of spontaneous mutations on Each dependent mortality acceleration is very very small So which means that maybe at least when we talk about human longevity by the time Related to the average life expectancy the effect of those spontaneous mutations is not very large So I will conclude with that. I mean the work is not Very sophisticated, but I think it's still very interesting We do observe that Ultra limitations have more effect on phenotype than common mutations I like this picture because this is exactly the view people got after world war two when mathematicians tried to investigate How to redesign airplanes to make them more durable and survivable. So that's more or less what we are trying to do here in aging So the first observation was that the airplanes are coming back and have lots of holes So let's reinforce the places where there are many holes. So these are our common genetic variants which can change a lot Well, the correct answer was the other They are the other way around you'll have to reinforce those places of airframes where there are no holes because every hole there means that So that's exactly what we see here. We see common gene variants which can be changed that will these are not your health These are your ancestry hair color eye color and so But there are very you know very very lethal Variants which if change changed reduce large effects and that's where the whole exome and hopefully whole genome sequencing Will I hope provide us with lots of interesting variants of genes with larger effect on human lung? So I would conclude with that and I would like to say thanks to everyone present on the work people from my group Which is alexander zenin on andrey tarpoff and also to anastasia edak and oblasty wadim was The brain behind the the first version of this work Thank you Great. Thank you, peter for that very clear talk and I just want to remind all of the Participants to please put your questions in chat. We have A couple of questions. Uh, so the first one comes from alex chen Who's wondering he got a 30x nebula genomics genome scan. How can he find his ptvs from the genomics report? well We have andrey tarpoff in our group who has also received his Genome scan and he has just produced such list. So if you if you mail me here I will contact you with andrey and I think you will share the code Great, thank you Although Uh, you may be setting yourself up if you start offering to have people email you Uh, with their genomics reports for a lot of work, but uh, thanks That's great. Uh, I just promised to connect. Right, right. Uh, okay. We have another question. Uh, Uh, very nice work. Did you observe any sex specific variants? That has an individual burden on maternal or paternal side in your study? Uh, yes, uh, well surprisingly, we have seen uh, such dependent effect I mean one one of its it is Serious and mysterious because when we look at health come and lifespan We can see that certain specific types of rare ptv variants affect different sexes in different ways We have a table for that no explanations This is, you know, local I mean in quotes lower quality observational science So we don't know the reason the anecdotal part is I have on my on my slides You may mention that there is an association of ptv burden of an individual with mothers age at death But not fathers age at death. You may think whatever you want about that So it's also not known why but maybe it's because Fathers of that generation were involved may be in war. I don't know So it may happen that aging did not contribute much or as much Of an effect into the death tables Of the of one parent in comparison to the other part. So yes, we do observe some sex related effects They are reported but not properly explained Great, thanks. Uh, another question is Are the are the results stable with changes of definition of ptv for example the frequency of ptv Well, you can Can please repeat because I lost a bit Yeah, so I'm not I'm not sure I completely understand the question. But the question is Are the results stable with changes of definition of ptv for example the frequency of ptv Yeah, well, thank you for the question. And then once you have such uh, such a frequency dependence You have to be careful about that. So I'm happy to report that since three weeks we have to 200 000 Genomes and the result is there is the proper pv new improvement According to the number of samples. So I would say that this result has been cross-validated Okay, thanks for the question correctly Yeah, and then uh, this is something I was also sort of wondering about is do you think that your findings Can inform functional studies in for example model organisms. Are there specific pathways that are enriched? Well, I hope yes, we still need to prove that There is a follow-up version of this work where we do study specific pathways And we have generated a bunch of hypotheses and then we're asking our friends Nemotops to confirm them. So the work was in progress at this time Great. Okay. Thank you. Uh, I'm not seeing any other questions. Uh, so So, uh, let's just thank peter for the the great talk and uh, he did put his email in the In the chat. So if anybody has Additional questions, you can either put them in the chat and peter may be able to answer them there or you can email him directly There was an error in the email. So I have fixed that Okay, sorry one more time Thanks. Yeah, thank you All right. So we will now uh go on to the second uh speaker of this session So sarah hog as just mentioned, um, sarah was one of our, uh, fantastic special editors for for the special issue And uh, she also is an author for the special issue. She is at the department of medical epidemiology and biostatistics at the carolinska institute from sweden. So we're we went from singapore to sweden We're going jumping all over the globe in time zones. Uh And her talk title is measurements of biological age in swedish longitudinal study of aging Let's get started and well, thank you for inviting me to this Symposium today. Uh, so I'm not the first author of the paper. It's my brilliant student. Yeah, uh, who has done all the analysis of our paper And she couldn't unfortunately do the presentation today. Maybe she's listening. I don't know so Yes, so the talk today, I will present some of the background on what biological age is in in human cohort studies and some of the knowledge gaps Which led us to the aim of our study And briefly go into some of the methods and then present the results of the paper So biological age in human studies Uh, it can be measured in many many different ways, but it should be a measure that is correlated with chronological age And it should also be something that provides information on top of chronological age so that people at the same age can have different levels of biological age and that means that A person with a higher biological age would also have a higher risk for age related diseases and mortality later in life And there are many layers of aging metrics today So we can think about molecular and cellular mechanisms of aging and how you can quantify that And this is what we like to refer to as the hallmarks of aging and has been written And referred to intensively over the last years in the aging literature We also have a biological metrics of aging Where we look at more global measurements So many of the omics clocks, for example, and the most common one is of course the epionetic clock Very very often used now in studies But we also have other clinical metrics of aging where we can think about An individual or even whole organ or whole system level of aging So we try to quantify this in humans by looking at disability frailty different frailty measures Multimorbidity and functional measures such as cognitive and physical function So what we don't know is of course how all these different layers of biological Aging interact with each other and how we can use that information to learn more about the aging pathway So we have multiple different biological age metrics that has been proposed such as The omics clocks, for example, and the clinical different clinical biomarkers that can be combined into A summarized metric such as the clemerida ball algorithm and so on And some of these measurements have been very well studied lately and we understand how they work to predict mortality and health related outcomes But what has not really been known is How many of them can be combined together in the same population and how then the different biological age metrics correlate together And how the mortality association will then be predicted when we combine different metrics So this led us then to the aim of our study And we wanted to understand a bit more around this and estimate the correlations between different biological age metrics in the same population and individuals And also then to explore the associations with mortality. So both separate as we know them But putting all of the different biological age metrics jointly in the same model as well And the methods this study that we used to do This investigation was a longitudinal aging study It's called the swedish adoption twin study of aging Sapsa for short and it's part of it's a sub study of the swedish twin registry And in this study we have up to nine in-person testings between 1996 and 2014 where we collected biological data and on top of that a lot of questionnaire data and other registered data as well So at each of these IPTs A maximum of nine different biological age measurements that we used in this paper were collected So a maximum of 845 individuals had at least one of these measurements available in this study But many of the individuals had Many of the different measurements available and also with repeated measurements over the this time period And the different assessments that we used Were the loci telomere length using qPCR method And the epionetic clocks And we quantify this with a human methylation for 50k b-chip array And we had a multi-tissue orbit clock, which was the first one that was really initiated And we also used the whole blood hanum clock and the pheno age, which is a quantification of physiological age And also the fairly new grim age clock, which is more of a mortality predictor clock We also combined our own physiological age metrics by summarizing 10 different biomarkers from blood and physical examinations In using clemerodobal algorithms with some modifications and this we have put into Bioprotocols for the assessment of how we did this physiological age score We also did a cognition general cognition assessment using Different domains of the cognitions of verbal and spatial ability memory and perceptual speed Combine into this general cognition score We also did a functional aging index, which is something we come up with on our own within the cohort Where we combine also different functional measurements. So here we used vision hearing lung function grip strength and gate speed And we standardized them and combined them into a summarize score as well And finally we had assessed the frailt index, which is based on the rockboard Accumulation deficient model Where we included 42 different health items that were self-reported in the participants in the satsa cohort So all of these are together the the nine different biological age assessments that we used longitudinally in in this paper And we also had mortality associations From the swedish population register linkage up to the end of 2018 included We did correlations of the different biological age assessments and we did this in a longitudinal way to Maximize the power and the samples that we had in the study And by doing this we had a method where we could Adjust for the relatedness also in the correlation coefficients For the association with alka's mortality We performed cox regression models. So a time to event model where we adjusted for age sex education smoking and bmi In the models and we also included robust standard errors to adjust for the twin relatedness that we had in the cohort So two different types of cox regression models were Down so we had this one that we call the one biological age models where we Assess the association to mortality and for the risk For each of them assessment individually and then finally we did the combination model where we put all of them together in a joint model And also together with the covariates To see how they would behave in a joint model So moving on then to results And in the baseline measurements we had 845 participants and Around 60 percent were women and the mean age at baseline was 63 years. So this is a Aging cohort and we follow them from midlife and onwards So from what we can see here from the figure as well. These are the nine different markers and the longitudinal trajectories and telomere length and cognitive general cognitive function Are declining with age, which is of course what we expect the telomere is getting shorter each each time we each year we age And also cognitive function is a declining Event while all the other biological age assessments by construction then is something that is increasing with age And more follows this biological age metric or idea And what else we could see that for the functional measure. So the last row of the figure here We had cognitive function and this Functional aging index that we composed and the frail index All of them seem to have some sort of acceleration in the trajectory So from around the age of 70 years, there seems to be some sort of increase in this slope So it's not really a linear model anymore We could also note that there is a sex difference for most of the biological age metrics that we saw And this means that there is also which we can note here at this Interesting sex paradox because in the cellular more and molecular markers of biological age. We can see that For telomere length and for the different epionetic clocks that women have A lower biological age. So they have longer telomeres and shorter biological age assessed by the epionetic clocks Meaning that they have a more beneficial profile Whereas if we look at the functional measures So here in particular the functional aging index and frailty index, we can see that women have a higher Functional and frailty index And are worse off in terms of biological assessments here So there is some sort of paradox going on here for them For what is noted in the cellular and molecular mechanism versus the functional markers where women have worse function and the worse health in the later span in life as well And then we looked at the correlations between all of these different biological age assessments And as we expect they all correlate very well with each other and the red Corresponds then to high correlation coefficients And the blue is a negative correlation coefficient And the blue is then of course For the telomere length and the cognitive decline where we would expect inverse correlation coefficient But what is interesting to note then is that if you adjust for chronological age in all of these correlation coefficients There is not so strong correlation patterns Noted anymore So we can see on on the top we can see that there is still a functional cluster. So we have a Fairly high around 0.3 correlation coefficients still left in the correlations between the cognitive function the functional aging index and frailty index But for the others, it's not very It's not so apparent correlation structures anymore. There is some sort of cluster also for the echinetic clocks where there is At least for the Horvett and Hannum clock. We can see some sort of Fairly strong correlations still left, but this means that a lot of the correlations were really Driven by the chronological age correlation itself and when we regress that out Um, they really represent different parts of the aging Pathways, so that is an Interesting notion that I think that more and more papers come to the same conclusion now And finally then we looked at the mortality associations in and first we had this one Biological age model where we looked at them individually And then we had the combined model. So in the first Forest plot here to the left, we can see that almost all of the biological age measurements except for telomere length in in this particular cohort Shows an association with mortality so we can predict the hazard from using these measurements But interestingly then when we combine all the Biological age measurements together and put them in the same model Much of the estimates are attenuated or the effects are attenuated in the model So what we can see is that from the functional area the functional Markers, it's only the freight index that is still predictive of mortality And from the aging clocks, we can see that it's the grim age and the Horvett clock that are still Predictive of mortality meaning that they are significantly predictive On top of each other. So they provide some sort of predictive value on top of each other in this model Whereas the the effects of the other markers are really attenuated and not really contributing To have an important value in the combined model So I think this is really something that was not or has not been shown in in any other cohorts before And it's also of course Restricted to what type of data you have access to and what type of analysis you can do So I think this was the first time we showed with so many different markers how it behaves together in predicting mortality So to conclude them, we looked at the longitudinal growth of different functional Biological age index and could see that they accelerated around the age of 70 years Sex differences were also apparent in in the way that we Perhaps expect them to be but also pointing out this Paradox with molecular and cellular and functional aging And correlations between the different biological age metrics were attenuated after adjusting for chronological age And most of them except for telomere lengths were individually predictive of mortality But when putting them in this joint model, we could see that only some of them were still important to have in the model So maybe also resembling the correlation clusters that we could see that there is enough to include one functional measure Whereas two of the 18 clocks were were still predictive of mortality So yes with that, thank you for your attention And I would really like to thank the funders and my brilliant student Chiya who did all of this work And also my colleague Nancy Petersen who is the PI of SAFSA and responsible for the nice study collection Yes, I may take some questions if there are Great. Thank you, Sarah right on time. Uh, looks like we have a few questions popping up. So the first one is Molecular based methods telomere lengths epigenetics, etc are tissue specific And mostly use data from one tissue blood generally Well clinically based methods are not necessarily tissue specific How much do you think tissue effects? influence Methods to predict biological age And do you think it's fair to compare methods for molecular data versus clinical data? Yeah, I think it's it's a good question. And of course, it's tissue specific Specific estimates when you look at cellular and molecular markers and we know that But for telomere lengths, for example, we know also that Um Measurements in blood is fairly well correlated to other tissues. There are of course exceptions and to that, but it's somehow Perhaps the circulation is more capturing the bigger picture of the individual as well But for other tissues, of course, it will be helpful if the more you have the better you can you can understand the mechanisms, I think and For the clinical perspective. Yeah, it's Again, you're catching more of them of the bigger picture them. So if you look at frailty, for example, it's You get a picture of the whole body and the health of that person which is also perhaps more titling to clinical aspects and Clinical outcomes. So I think that is also very useful But of course you don't go into the molecular details then so in my perspective I think we need both or everything Yeah, I I agree with that. I mean, I think, you know, and Sarah, maybe you can comment on this I mean, I think that you know the the the blood-based measures potentially tell you something different, right, but but We also need things that are useful and and obviously we can't take biopsies from Thousands of people of every tissue. So I think I think you need both, right? You need the clinical assays where you can actually Look at function in different tissues and organs, but we also need to be pragmatic And work with the the kinds of biological samples that that we're likely to be able to use Going forward that that we can get access to from a large number of people Yeah, yeah, I agree Okay, so there's a few other questions Hi, very good talk. Did you try and fit in age By predictor model in the Cox models for telomere length. We know that especially in old people This is not a good predictor of mortality, but less than 65. It is okay And related to correlations between these measures break down with age Um, that's a good question. I think I think we looked into that age interaction terms in the models And I think it's in the paper. I would have to refer to that. I don't know the exact details of that But of course it's important and we've seen from other studies that In terms of the predictive value a lot of the hazards really go down. They decrease with age So if you are able to look at Risks at midlife, it's probably much much more useful than in late later life in many of these measurements. Yes And then there's a question about just just out of curiosity. Have you thought about adding to the analysis? The dna m age based on Horvath's clock The newest one which is based on skin and blood tissues So I think it's asking about newer newer clocks that have come out since you did this analysis Yeah, yeah, I we have that already created And I don't know we did not include it in this analysis, but I I have a separate analysis on that as well I still think that the green mage Is perhaps the the best one in terms of predicting clinical usefulness For for mortality because it's created in that way and we could also see that it was the highest hazard ratio For that clock. So I think that is still better than the skin and muscle clock Okay, thanks Was there any other remarkable accelerating jump over 80? Besides 70 that was dessert discernible No, I don't think so. I I think what we saw is really this Change in acceleration of slope around the age 70 And I think in terms of the cognition It's been known for long that's around retirement age Then this is the time when you see a big change in in function and cognitive function And now we could see that it's it's similar also for other function Functional metrics and freight index as well. So I think this is this is where the change is happening Okay, one last question How do you explain the different epigenetic measures show an independent association with mortality While only grim age is created using mortality as an outcome Yeah, yeah, that's an interesting question. And I mean it's an ongoing discussion of Um How the different aging clocks or the epionetic clocks are created where the first ones were really Trying to predict chronological age, which is In one way not a good Way of looking at biological age because we don't want to predict chronological age We want to predict biological age So I think the grim age clocks and other clocks have really been like what we call the second generation clocks where we try to Adapt the the way that the clocks are created so that we can look at biological age itself because by Really by intuition, we don't we don't want to predict chronological age. That's what they do in forensics So I think this is why there is a big difference between the first orbit clock and then for example the grim age clock They all work somehow to predict health outcomes, but in different ways Um, so yeah, I think we're still far from understanding exactly the mechanism behind it And I still think we have a lot to do in this area Yeah, that thanks sarah. That's a great finishing thought I think for for this talk and clocks obviously are you know a major Area of interest and research in the field right now And I think there are lots of questions about chronological versus biological age and mechanism and what these things are telling us and Just as a teaser, we will have morgan levine who is also going to talk about her work with some some clocks Coming up in a session later today. And so I will come back to this topic and I think the same Conceptual questions keep coming up over and over and over again. So lots to be done in this area So thanks for the fantastic talk. Okay up next. I think dario is here now um, so dario valenzano is our Uh, third speaker of this session Yep, I see him. Hey dario. Good to see you Uh, so dario is at the max plonk institute for biology of aging in germany Uh, and he is going to talk about interest species differences in population size shape life history and genome evolution All right Can you see my presentation? We can see all of your slides What about now? Yep, looks great All right, let's see please interview Like this Yes, all right. Perfect. So it's great to be here. Um, it's great to discuss about this work Um, I want to thank you matt and jess for Putting this imposing together I think this is a great initiative So the work I will present you today Uh, is the work that has been largely conducted by Former PhD students in the lab the first author david villansom Here's david and uh So I graduated last year. So what I want to stress about this work and I'm very excited about Is that this work started, uh, from a discussion in the lab And then moved on to the field in zimbabwe So we went and collected the samples that we use for, uh, you know, the whole the whole study Really in the in the in the in the puddles in zimbabwe And I will talk a little bit about that and then david carry on the whole study from the field work to the molecular analysis in the lab to the genome sequencing to the analysis of the data the bioinformatic pipeline to the population genetics and genomics Done to uh to tease out the genetic variants associated with lifespan differences Among populations that I will tell about so it's really like a fantastic story started from From field work all the way to to data analysis So data harvesting etc etc So the the main question that we're trying to tackle With this work is what is the evolutionary origin Of the genetic variants that limit lifespan In other words, we are interested in understanding how within a species different population Get to live shorter And what genetic variants are responsible for shortening of lifespan within a species so what is the evolutionary basis for the emergence of Deleterious gene variants in a population so rather than thinking about what makes you live longer As a species or as a subpopulation within a species We are thinking about the the other way the other direction. So what contracts lifespan? And so in other words, we want to really dig into the evolutionary mechanisms That are in action when Lifespan is contracted aging maybe is accelerated And this type of question Can be asked When you have at hand a system that has a natural variation in lifespan. So we're not addressing here what variants can be manipulated in the lab Individual variants, but we're talking about genome-wide variants in natural populations Okay, so we have a quite handy framework to address this question and so there are two General ways to think about how mutations that cause aging come about in evolution And I will briefly discuss them So in one such scenario, you have new germline variants like this little C here That leads to increased fitness Maybe because this variant when acquired leads to higher resilience And eventually to higher reproductive success to the individuals carrying it This same variant, however, may have adverse effects late in life that have no impact on fitness or that cumulatively have a limited impact on fitness, but still these variants are favored by selection So we are in other words talking about variants that are positively selected When a variant is positive selected, it's Kinetic the the the frequency over generation of these variants will follow a predictable Rule, so you will have an increasing frequency until it reaches fixation Which means most of the individual if not all individual will carry that variant because it's advantageous So this is the textbook case of hard sweep positive hard sweep positive selection Now another scenario It's quite different Consisting in having a variant Emerging in the population that has no effect on fitness no whatsoever or very limited effect on fitness this variants At the same time may be associated with age related diseases and decrease resilience robustness late in life Now the distribution and the frequency of these are such variants in a population It's very different from the from the previous one you expect completely different landscape of these mutations in the two scenarios So we can study genomes of long live and short live populations And species to mine for variants that have evolved in this way or in this way And these two Ways of looking at evolution and the evolution of these variants have you know have very famous names in the Evolutionary theory of aging one it's known as antagonistic play outropism and it's underlies really positive selection adaptive evolution This is really what george williams is talking about when he thinks about antagonistic play outropism Is that mutation accumulation me the war is a scenario that population geneticists would Would make it more similar to what happens in a scenario of nearly neutral evolution which is what otta tomoko otta and Has has been has been proposing as a as a potential mechanism to explain gene variants Distribution across populations So we want to know and we want to use natural populations to study This type of problem and to see how aging related variants distributing populations So we look at the at the world out there We look at species in their natural environment and our taxon of choice is that of killi fishes african killi fish in particular These are a wonderful system in my opinion because they are naturally represent a natural experiment in diversification Of life history traits evolution so multiple times independently in the phylogeny of this taxon You have evolution of short lifespan and longer lifespan So we can really study the evolutionary basis the evolutionary signature associated with different life history traits So past work from our group actually has shown that different species of killi fish Have particular genomic features that evolve hand in hand when long lifespan or short lifespan involved But what I want to talk about today instead is what happens within a species. So when you take one species Uh, and so you will look at the micro evolutionary scale. So within a species different populations How can we explain differences in gene variants leading to longer short lifespan? Among different populations and this is exactly the case that notobranches the genus notobranches It's several species within the genus notobranches offer as as a model so just a few uh, if you introduce a slide about annual killi fish the short-lived killi fish So they have a very peculiar unique life cycle. They, um, complete the life cycle in When water is available in Savannah africa and so they hatch when the rainfalls come they reach actual maturation in a few weeks They spawn they lay their eggs They will keep on reproducing for several months But their their eggs and their embryos in their eggs will arrest their development for several months up to several years Until the next rainy season comes so there is no overlap of generation within one season Most of the time of the year in the dry areas embryos were survived in the dry mud Okay, so like I said before we have long-lived and short-lived killi fish species Sorry populations within the same species So the the the short-lived populations come from extremely arid areas And so this is southeast africa zimbabwe and Mozambique This is the area where we conducted our fieldwork and we collected our samples Like I said most of the year Short-lived killi fish live in an environment which is like this and you really have to dig down in the mud to collect The the eggs so you can actually extract them from the mud and this is a pond You know how the pond looks like most of the year So we know that different killi fish populations are distributed in a gradient of Precipitation very few precipitation very arid environment here in red to wet wetter and wetter area as you move towards for example The coast of the Indian Ocean Here the precipitation lasts longer rainy seasons are longer and killi fish also live longer in those environments Just to show you how ecology explains these features in killi fish So in the wet areas you tend to have large ponds like these even small lakes with a lot of fish and there can be a lot of exchange of Of populations between the different the different localities because these areas are often under monsoon You know this is a monsoon area and they get flooded often time So there is a lot of gene flow between Ponds in the wet areas the dry areas instead above here are you know represented by smaller You know localities smaller populations smaller ponds that last for less time every year They have a very small population So in each pond you have continuous bottleneck and repeated bottlenecks And so the gene flow also among these populations is very limited So this we predicted already back in 2015 that this actually would lead to severe genetic drift and potentially effects, you know This might have affected the distribution of gene variants in these populations So this is the case for the turquoise killi fish where you have populations living in dry environment Which have this type of scenario in population in the wet environment Which have this kind of scenario from these populations you have shortly populations in these localities You have longer population now. You may think that short lifespan may be adaptive that Individuals living in the dry areas may actually be adapted to their environment However, what it's very important to point out is that the time to sexual reproduction in both dry and wet areas is the same So killi fish don't display within these species the turquoise killi fish Differences in timing of sexual maturation in other words both population from dry and wet environments rich sexual maturation at the same time And they are, you know Reproductive lifespan is shorter in the dry populations than in the wet populations So we tested actually whether the bottleneck is real and whether the population size indeed is smaller in the population That lives shorter. So this is actually we see so we collected samples fish from four different localities here represented tree But actually there were two localities in the dry area here So this is a dry area and these are like more and more wet area And by pooling and sequencing different individuals from these localities We could actually assess the From genetic polymorphism the effective population size, which is a is a measure of Actually the actively breeding individuals and is a is a measure of genetic diversity We can become smaller and smaller as we go to the dry area This is a psmc plus plot that represents actually population size. Each of these lines is a population is Represents population size over time. So on the right is the past and on the left is the present So as you move for example a long time towards the present on the orange line Population size undergoes a bottleneck a severe shrinkage in population size for these other lines instead You have increased population size in recent times and actually the orange line here represent actually the The the the the the short lived populations that come from these arid areas in Zimbabwe All the other populations actually have undergone recently an expansion of population size So indeed from population genetics, we can observe directly that there's been severe bottlenecks affecting the dry populations Now, uh, is it true though that smaller effective population size affects the Amount of of of gene variants positively selected and increases genomic load in other words The accumulation of deleterious gene variants to answer this question We can actually assess we can weigh The number of adaptive variation and the frequency of adaptive variation in a given genome By using the so-called mcdonald kreitman alpha test I'm not going to go into details because you know for the sake of time But what is mcdonald kreitman alpha? This is a asymptotic mcdonald kreitman alpha tells you It actually compares genetic variation within and between So within populations and between species so ancestral versus actually polymorphic variants And it compares this variation in neutral and non-neutral variants. You know the words. I mean acid leading, you know changing variants versus neutral synonymous Leading variants and what you can well all you need to know about this these slides here is that The red line represents the shortleaf population which are the bottleneck the smaller populations here and the fact that the y you know the y the y Access here is Values are lower for the for the red population indicates that Shortleaf population have weaker have a smaller portion of the genome under positive selection at all Frequency bands and in particular as you move to the low the rival your frequencies You have more and more slightly You know deleterious gene variants So these negative values of mcdonald kreitman alpha indicate an accumulation of deleterious gene variants specifically in the Small populations So this is done actually using two outgroup species not a branches racobai and not a branches ortonautos And the result actually is robust to this test Not only we can assess the genome y that is lower portion of the genome under positive selection In the in the shortleaf population, but we can also assess based on the coding variants The impact on you know the phenotypic impact of these variants and we can actually find that in the Shortleaf population gmp the red ones. You have a larger portion of early stops So stop gain So you have a lot of pseudogenization that is happening in the smaller population compared to the larger population In other words, yes, it does seem like does seem like, you know Smaller population size leads to an accumulation Of slightly deleterious gene variants that we believe cumulatively may lead to shortening of lifespan in these species So to conclude David also ran together with ray the second author on this paper. They ran the Test which is called direction of selection and is a distribution of variants that are on the left negative are actually Driven by relaxation of purifying selection or deleterious on the right instead are adaptive and on the middle at zero point Are neutral and so what you can study is actually what is the nature of these genetic variants that are other deleterious or adaptive so If you are on a positive on the green side, these are adaptive variants If you are on the left side, these are deleterious gene variants Well, all you need to know is that For particular for the for the shortly populations We have a large number of genes involved in the wind pathway any Alzheimer disease You're the generative diseases an immune function as well as cancer that have accumulated Over time And they've fixed that means that they have high frequency of deleterious gene variants For terms associated with degeneration So now why is this helpful because this type of population genetic approach Leads us to list of gene variants that can be tested actually in the laboratory. So now we have our tables We can go back to those and we can test that the phenotypic impact of each of these variants on a specific pathway So to conclude what we have found is a connection between demography population genetics and genome evolution And I didn't have the chance to To to to talk about it But this paper also provides a new genome assembly for the turquoise killifish the most recent Reference genome for these species. We see that population size matters in the accumulation In the efficacy in which selection can remove deleterious gene variants in particular small populations Population with smaller effective population size lead to a higher Number a higher load of deleterious gene variants And this would believe leads to a shortening of lifespan specifically in this population So this is the conclusive slide And this is the acknowledgement I would like to thank the whole lab in particular the people who did the work are David here and ray And also our collaborators in the Czech Republic Martin Reichardt, and this is our funding and if we have time I'd be happy to answer questions Great. Thanks. Thanks, Dario every time I hear you talk. I realize that I I've picked the wrong model organisms to work in Because I never get to go to Zimbabwe. Oh, you can come anytime mass I'm going to take you up on that Okay, uh, my dogs actually in Zimbabwe, you know, the Swedish dogs. They're wonderful Um, so if you have questions, please post them in the chat I'll I'll start and I don't know that I've really Formed this well even in my own head, but you know I guess the question in my mind is when you have these deleterious variants that that pop up in these short-lived populations It's easy for me to think about how you would get, you know, sort of randomly accumulating mutations Processes that that will limit lifespan. I think what's interesting is at least it looks like what you see is Is mutations accumulating in Pathways and mechanisms that seem fundamentally linked to the biology of aging like these not all of them But many of them you can they tie right into the hallmarks of aging and I wonder if you have thoughts on Why would that be the case that these these deleterious mutations that pop up in these populations? You could make Aging process right rather than just random stuff Well, so think about this way if you if you're deleterious gene variants affect early life processes like embryonic survival or time of sexual maturation even you know or sex organ maturation Even in the small populations where selection is lousy Even then the effects so natural selection will wipe those out Uh, so in other words what I'm trying to say is that the gene the variations in aging Pathways may be Dispensable. So these are not highly deleterious gene variants. These are slightly deleterious gene variants So these are actually bearable and so what we believe also from our previous work in comparative genomics of aging across different species rather than different populations is that If anything positive selection in kilifish in annual kilifish in the short-lived is compensatory In other words, you produce a lot of damage these accumulate a lot of tiny little, you know dense In the in the canvas of this of this aging pathway and then the the residual genetic variation Will be used by natural selection to compensate for those for those losses basically So we think that there is a lot of redundancy in the aging pathway and those are slightly deleterious gene variants These are not highly deleterious gene variants, but they're just a lot. So it's a polygenic trait It's not simple genetic architecture with one main locus. It's actually For probably the one of the most polygenic traits lifespan in aging that you can imagine Is you know, even more polygenic than height or probably weight Right, right. Yeah, that makes sense. Thank you. So we have one question the population genetics of variants are Consistent with the demographics, but there is no direct evidence to connect The accumulation of deleterious variants to the shorter lifespan, right? So it's the I guess the question is Do you have direct evidence connecting the accumulation of the deleterious variants to the great question? It's a it's a it's a crucial point. So what we try I agree. I agree. So we you know, uh, it's very hard to prove that So so far we think that this is the most parsimonious explanation that we have we don't you know, um, Let's put it this way, right? So, uh, the alternative hypothesis that I presented initially is that positively selected variants In other words, a few very Highly frequent variants are those that lead short lifespan, you know, the drive short lifespan in the shortly populations But we don't see evidence for that But we don't see evidence of a few Gen variants that are unique to the shortly populations So we see that genome white that is like a high humongous gene, you know, genomic load in the in the shortly populations Uh, another thing is that we We see a association. So there is also like a timing of expression. So we see that Genes that are expressed later on in life are those that are under more relaxation of purifying selection in the shortly population We have done these analyses on the basing of transcriptum. But, you know, if you were to test each of those variants probably None of them would alone lead to shortening of lifespan. This is a polygenic trait. Like I said before so Great. Thank you. So, uh, we are right on time. So we will move on to the the next speaker. Thanks again, Dario That was fantastic So we are at the last speaker for this session John on uh, so this is really sort of a treat for me to get to introduce john John Did his phd thesis in in my lab and now he has gone on to To to to run his own laboratory in the school of dentistry at the university of washington and john is sort of a rare breed in the field of aging research in that he is a licensed dentist who practices clinically And also has a phd working on the biology of aging and I think it's it's it's kind of neat that john is really taking the geroscience approach in the sense that Geroscience is is the idea or the the area of research to try to understand the mechanisms that link biological aging to disease And john is one of the one of the few who is doing this in the context of aging of the oral cavity So so it's really really great And fun for me to get to introduce john's talk on the application of geroscience to extend oral health span And your slides look great. So take it away great. Thank you matt and uh, thank you jess And thank you. Eli for the opportunity to speak today. Um, so as matt mentioned, my name is jonathan on I'm currently at the university of washington school of dentistry So i'm a dentist scientist that studies the biology of aging in the context of oral health Let me just make sure there we go So age is the single greatest risk factor for many known diseases and decline Including Alzheimer's disease cancer as well as heart disease however in contrast to All these major age related diseases that the field of aging has focused on The impact of biological aging and the oral health that relationship is often neglected And as a clinician I commonly see our LOD patients do coming in with various oral conditions Those could be related to dental cavities periodontal disease low saliva or zirastomia Candidiasis as well as oral cancer And in fact understanding the biological mechanism of aging in the oral cavity is critical to not only reduce the impact of age related decline in the mouth But really the optimal functionality for any system requires overall organismal health and this includes oral health And so for our study to investigate this line of inquiry We first looked at an oral disease that commonly affects older adults And we first looked at a disease called periodontal disease, which is a chronic oral inflammatory disease Adults over the age of 65 or roughly about 70 of those adults have some form of periodontal disease And in fact the definition of periodontal disease is the inflammation that occurs around the supporting tissues of the teeth That leads to bone loss with the variable microbial pattern And in fact one of the clinical hallmarks of this disease is the bone loss that happens around the teeth And so as with many research inquiries, we first wanted to come up with a model to evaluate this disease and its process during age Prior studies have utilized artificial disease models in young animals or young rodents like this here CT imaging that we produced in our lab And to induce the inflammation or induce the bone loss to mimic an aging phenotype However, a limitation in many of these studies is that we lose the actual contribution of the age local as well as the stomach environment And so for our study, we did not want to induce any artificial disease But we just wanted to evaluate a normative age mice And in fact, we just we just discovered that just age alone causes not only periodontal bone loss Like you see here in an old animal But the fact that both the gum tissue the gingiva or the bone There's an increase in inflammatory cytokines or inflammation And so this is a protein dot based assay where we're looking at a full change of the various cytokines That increased with age relative to the young here is set to one So as we uncovered that age causes common oral disease or periodontal disease We wanted to evaluate whether targeting the biological aging process could delay this disease And we decided to investigate the impact of rapamycin Rapamycin is one of the most robust Most studied as well as a reproducible intervention for increasing lifespan as well as delaying age-related phenotypes Rapamycin was first isolated from soil samples on Easter Island It has a potent immunomodulatory response As well as genetic studies in yeast first identified that target, which is Tor And so M Tor or mechanistic or mammalian Tor is a nutrient growth factor response of kinase And is structurally functionally conserved from yeast all the way to mammals And on the left you're seeing a lifespan data produced by rich millers group as part of the international Interventions testing program or ITP by the NIA Where they showed that with increasing dose of rapamycin in the food We're able to get an increase in lifespan extension in both females and as well as males And like I said, M Tor is a nutrient growth factor response of kinase Because our kind of questioning was one of the first studies to look at the relationship between not only just oral health and age But also the impact of rapamycin To increase the robustness of our study. We had two different animal cohorts going across the United States So one quarter we based here in Seattle And then another quarter we had collaborators at the Jackson laboratory in Bar Harbor, Maine Where we aged those animals to those ages and then we either gave it a control diet or rapamycin diet for eight weeks And then they were harvested with the Jackson core. We were also able to take an initial CT imaging before the rapamycin was administered and then a final CT of the same animal after the rapamycin After the rapamycin treatment A lot of our data is in the elike manuscript So I just wanted to highlight a few of them that kind of kind of ties up all our story The first is that what we found was that rapamycin was able to attenuate an age related Gindival inflammation and so on the left we're showing you NF Kappa beta p 65 and ikb alpha which are the hubs That causes the production of inflammatory cytokines What we're seeing just in the gum tissue There's an increase in age and then rapamycin was able to decrease That increased expression and in fact various inflammatory cytokines that increased with age was attenuated by rapamycin treatment If we look at periodontal bone we used wrinkle which is as a marker for osteoclasts That we see that wrinkle expression increases with age But then rapamycin was able to attenuate it But just like similarly in the gingival tissue even in the periodontal bone What we find is that animals treated with rapamycin have this attenuation of inflammatory cytokines and chemokines If we look at the clinical phenotype What we find is that rapamycin seems to rejuvenate the age related loss of periodontal bone So this here is showing you cores from here in seattle where we can do a cross sectional. So these were taken Micro CT imaging of those animals you can see in the maxilla or the top teeth or the mandible on the lower teeth Just age alone leads to periodontal bone loss But rapamycin treatment for eight weeks was able to attenuate that And in fact if we look at our Jackson laboratory cohorts of the same animal as before and after On the left, this is just one image and we have a few more in the manuscript You can see on the left the periodontal bone loss both in the mandible as well as the maxilla So the maxilla or the top teeth We've highlighted some areas of bone loss in the white arrows And on the right you're seeing in the orange arrows after the treatment So again, this is the same animal where we took the image before and after You can see areas where there seems to be bone deposition around the teeth As well as both in the lower teeth as well as the top teeth And finally what we wanted to also look at was because this disease has a variable microbial pattern We wanted to look at the oral microbiome and so this we're looking at alpha diversity So you're looking at how many different species are within the environment We find that the old oral microbiome is significantly different from the young But in fact if we look at the actual microbial composition And we find that the young and the rapamycin which is in the turquoise and the blue are overlapped While the old oral microbiome is completely segregated Showing that there seems to be this reverberative shift of the aged oral microbiome to be more similar to the young oral microbiome And in fact if we go back to the very definition of paranormal disease Which is inflammation of the periodontal structures which includes the gum tissue and the bone With bone loss as well as variable microbial pattern What we find is that rapamycin is able to target all three clinical features of this disease And so what we're able to include in our paper was that in short-term rapamycin treatment was able to Impact these three clinical features of periodontal disease And really supporting the geroscience hypothesis that any interventions that target the biological aging process Will simultaneously delay multiple age related disease as well as functional decline I also wanted to kind of share some additional data that we have kind of in supporting for evidence Is that one of the questions we commonly get is that after the eight week treatments where we see this rejuvenation of oral health Do those results persist? So all this beneficial of the periodontal bone the microbiome changes and the inflammation um Those those effects continue on and in fact we had a separate cohort at the same time where the first animal cohort what we did was we treated them with the Eight weeks of a control diet and then they were transferred into the eight week rapamycin treatment diet The second cohort we had we started them on the eight week treatment of rapamycin But then they came off the rapamycin. They were given just the control diet for eight weeks And in fact if we go and look at the wrinkle expression So um, you were seeing the young old and the rapamycin and then the first column is the control Diet and then the rapamycin and then the rapamycin diet and then the control we find that wrinkle expression Maintains its decreased expression levels even after the rapamycin was taken off And in fact if you look at ikb alpha similar to our Our findings previously is that rapamycin not only attenuates that But it persists even after the rapamycin diet is removed and really the If we look at the micro CT imaging of these animals So these are the animals that were given rapamycin for eight weeks and they were off the diet We find that initially at baseline before the treatments given we see the periodontal bone loss as indicated by the orange arrows And after eight weeks rapamycin we see again very similar to our prior data that there seems to be this regeneration of that bone around the teeth And when these animals are taken off their rapamycin diet, and then they're on the control diet We find that those results are persistent and so we're we're still Completing some other analysis on this and hopefully we get this prepared during this time So with that I just want to thank the university of washington, especially matt cabelline as well as peter revivich who really allowed me to get started especially matt for allow me to kind of Get started on this kind of field as well as helping me As I gained my independence as well as our collaborators at jackson laboratory And then obviously elie for giving me this opportunity to speak today. And these are my support for this during the project so Thank you. I'll go in and take any questions Okay, thanks. Thanks john. We've got a few questions. So So one question was whether and I think this probably applies generally whether it's possible to get the the powerpoint slides after after the talks so All of the talks are recorded and I believe that the link to the recordings will be sent to to all of the participants So you can certainly you'll have an opportunity to go back and review the presentations then and then It'll be up to each individual speaker if they want to to actually share the powerpoint slides if if people want to request that directly and it sounds like uh Anya can can help connect Participants with the with the speakers if that's something you want to you want to seek So one question is How different were the oral microbiota between facilities between jackson and udub and was that a confounding variable So, yeah, I think it might I think it leads to kind of the batch effect that was actually included within it So there was not surprisingly there was not much of a difference in the oral microbiome because Um, it could be related to the resolution But if we just look at kind of the final levels, there was not a difference between the jackson laboratory as well as the university of washington Um, especially when you look at those treated with rapamycin. Um, they were very close Okay, thanks john Uh, great data nice talk really impressive information Two questions. The first one is why did you choose six-month-old mice as the young control group? uh, and then the second question is Uh, osteopontin deposition did not seem to be increased after rapamycin treatment. Can you elaborate on that? Yeah, so for the first question of six months, um, that was just based upon A linear trend, you know, the young animals there was a range. Um, we always say between, you know, six to eight months The other reason why The young animals for us because we had the adult cohorts We knew that there was going to be and uh, if there was an age-related trend That the younger animals even they were at six months would show a difference at least with the adult animals and also because Our primary or preliminary data that we utilized was not only from the ITP studies But also from peter ravinevich's court a peter ravinevich's court and they actually use a similar month of Age mice and that's why we picked the six-month-old younger animals To the second question of osteopon. Yeah So one reason is could just be is that the way we collected the mandible bone So osteopon is uh, like you mentioned One of the osteoblast regulators and you know in the mandible bone There's both trapecular bone and corticobone And so for our study, we just we just took the whole entire mandible bone to do it And so maybe the subtle differences between osteopon. It was not detected Um, uh, we could if we do if we go back and look at our histology slide Who would actually do it for vexular which is portable? We may be able to see osteopon But because that was kind of out of scope of what we wanted to look at we did not look If you're interested we could always send you the slides Because we have those if you want to look at osteopon, but that would be something of interest So Thanks, john and this wasn't asked but sort of related i'm gonna i'm gonna push you to To give people some some clues here So I think one of the questions is you know when we see these effects of like bone regrowth or You know regeneration not just from rapamycin, but from other aging interventions as well You know, there's always the question about is that sort of you know high quality regeneration And so do you know anything about so you showed that the bone Regrowth seems to persist. Do you know anything about the quality of that bone? Is it brittle? Is it likely to break? I think this is you know, clearly an important question as we think about Translation of some of these interventions Right. Uh, yeah, that's a great question. Um, so we actually um recently Set up a collaboration with the mechanical engineering department mechanical engineering department here at the University of Washington where Normally what we could do is look at these biological structures Using what's called a nano or nano indenter so basically normally For crowns, for example, I'm going to a little bit dentistry is we could look at the kind of the strength of that material Utilizing this device called nano indenter where we could look at kind of the modulus and The hardness of that structure and in fact, we actually did that with some of our animals And what we find is that much like a lot of this data is also in femurs, but we do see it with age that there was Um More modulus as well as less hardness and in fact those animals tree will wrap amycin their bone quality was actually better than the old animals And so we haven't really published that data yet. Um, I know you kind of pushed me on that But um the so we're actually finding that Utilizing this kind of mechanical engineering technique that the bone is actually better Even though there is more now the concern is obviously, you know One of the questions you're leading to is there's a lead to osteoporosis But what we're thinking is actually it actually is making the bone a little bit stronger around the area of that Great, thank you. So one comment Very interesting data. It would be good to test the effect of wrap amycin On cancer development in the long term Probably the mouse model is not good enough for that. So you don't you can respond to that if you want to I don't know if there's anything on oral cancers that you would think about looking at. I know that's an age related yeah So, um, I know that they've done it in tongue where Again, it's some kind of artificial way to do it, but they irradiate the animals And they induce this kind of cancer in the tongue and in fact that paper is actually published years ago But they actually showed that pre treatment with wrap amycin is able to kind of help Reduce the impact of kind of tongue cancer But beyond that I haven't seen any literature on it But yeah, that would be definitely something of interest to in a mice model. Obviously it's an it's an induced model So, um, I'll have to think about the best way to approach that And then the last question is Uh, what was the dose of wrap amycin and I think they're asking based on body weight I don't know if you have that off the top of your head that maybe just Comment on the dosing. Yeah. So, um, our dosing we just did a 42 parts a million in the food um, and so, um, you know, we could go back to look at how much each Mouth scar based on the wrap amycin blood levels as well But and we picked that 42 parts a million dose because that was the highest dose that the ITP study went to show The lifespan extension and and because this was the first time we were completing it We wanted to see that if there was any phenotype that with the highest dose wrap amycin there there It was highly likely to see a certain phenotype. So that's why we're going with that Our protocol should be published on the e protocols. Um, and if you don't you could always Look at our paper or contact me directly and I can give you the protocol On how we created that the child for it Great. Thanks john. Yeah, and I'll just mention I think Like john said, this was the dose that that has been used by the interventions testing program And I think they did some calculations in that that paper where they first used the 42 parts per million To give an estimate of body weight, but of course, you know, this is variable depending on how much each individual animal eats And so I think it's not typically not typically quantified that way in mouse experiments Um, okay, so we are right on time. Thanks john. That was great