 Okay, Tim, we're good to go. Hi. Oh, hi. Hi. Welcome to the today's open research webinar by Elife. So we're aiming to give early career researchers and online platform to continue to share the research during the covered era when when people can't go to meetings. So I'm Tim Behrens and I'm deputy editor Elife. I'm a neuroscientist. And so today we're hearing from Erie Maus, from Chapman Uni in the US, from Stefania Mattione from UC Louvain, Belgium. I'm from Ananta Sundaraman from Bristol in the UK. So after each 10 minute talk, we're going to have five minutes questions for the speaker, and then we'll move on to the next talk. So to ask a question, you can type it into the Zoom chat directly, or you can put it into the Google document that's linked here. So the link is also in the chat. And so we're joined today by Anya and Naomi from Elife who are working in the background to support. And so they're going to help line up your questions. So if you're able to, I'll invite you by name to ask your question and the team will unmute you so you can do that. If you have any other questions out loud and include your name where possible. The open document is also a place for you to contribute shared public notes. So we welcome that and to list yourself as a contributor in the list above the speakers for today's webinar. So thanks very much. Finally, I just like to let you know that we're recording the webinar and also live streaming it to YouTube. We ask you that you please be respectful, honest, inclusive, accommodating, appreciative and open to learning from everybody else. Don't attack, demean, disrupt, harass or threaten others or encourage such behavior. If you feel uncomfortable or unwelcome on any of these webinars, please contact Elife by email to events at Elife Sciences. This inbox is being watched by Miranda Nye at Elife. The Elife organizers reserve the right to ask anyone to leave or deny access to any subsequent webinar or Zoom request if they're being unhelpful. So if you need any help at this point please send Miranda Nye or Anya a chat message directly using Zoom. Okay, great. First we're going to hear about Uri and he's going to take over the screen he's talking about neuroscience of volition for decisions that matter thanks very much indeed. Thank you very much Tim. So, yes. My name is Uri Maoz from Chapman University as you were told. And I'll be talking about the neuroscience of volition for decisions that matter. Now, if I asked any one of you to raise one of your hands, you of course would be able to do so. And the order of events as you probably experience it is that you consciously decide which hand to raise, and then that that causes you to raise your hand. So, if you look at the intuitive order of events, there's the conscious decision, then sometime later there's information in the brain about the action, and then there's the the action itself. However, an experiment in the 1980s by Benjamin Libet and colleagues cast some doubt on that order of events. In that experiment or at least one version of that experiment. They asked the participants to raise their hand whenever they had the urge to do so. And at the same time to look at a clock that had a rotating dot. And then just say where the dot was when they first had the urge to move or when they decided to move so the dots started rotating. And then at some point they had the urge to move and maybe they said that the dot was at five when they had the urge to move. Now at the same time, EG was recorded from the participants scalp, looking at electric activity in the brain, and something called the readiness potential was calculated from that EG, and that readiness potential is typically associated with voluntary movement. So, this is here the readiness potential this is where movement occurs, and maybe somewhere around here, something like one second before the movement. We would say from the readiness potential that we see some something there's some information in the brain about the upcoming action. So, where did people report that they that they had the urge to move or decided was around here so about 200 milliseconds before movement onset. So the strange thing is, what is going on in this time range here, because on the one hand, we already have, as we said before information in the brain about the action, but the subject has not yet reported that they have decided to move. So, is it the case that the decision was made unconsciously. So, these limit results caused quite a stir. They have been cited a few thousand times already and they were mentioned widely in the late press. And I should also note that they were not a fluke. So this holds for, if you ask people to raise one of their hands or if you ask them to raise either one of their hands. So when to move or which hand to move. These precursors of movement were found actually much earlier using fMRI even a few seconds before movement onset. They were found in single neuron recordings in humans as well. We found some neural precursors of action, even online in real time so on the fly as things happen. So, some people have concluded from those limit results that the decisions and limit experiments were made unconsciously. From that, they concluded that perhaps generally decisions are made unconsciously. And from that, it was suggested that, you know, maybe this is a challenge for some of the social pillars like our notion of free will and maybe even moral responsibility if we are doing everything unconsciously whereas whereas our free will and how are we morally responsible. So the thing I wanted to tell you about today is is something that we've looked into regarding this arrow here. The thing that the trouble does was that the limit experiments as I told you about them were about raising your right hand or your left hand or raising your or flicking your wrist or something like that whenever you had the urge to do so. But we were wondering whether these results generalized to deliberate decisions so deliberate decisions could be things that you deliberate about which which route to take to work in the morning. Or more important life decisions which job to take life partner to choose to choose and so on. And we were wondering whether these arbitrary decisions and these more deliberate decisions are really the same. And this is work that I've done with collaborator Liad Mudrik and other collaborators. And what we've asked people to do is make donations to different nonprofit organizations and PO's. And those decisions that they made about these donations actually determined a $1,000 donation so this was, I mean this was real it was not it was not fake. So on a trial subjects might see to nonprofits maybe the hunger project and cancer research Institute, and some of those trials those decisions were hard and other trials those decisions were easy for those subjects. And they had to decide which one to donate $1,000 to, and these were the deliberate decisions. We also made arbitrary decisions. Interestingly, and importantly for us, the visual input for the subjects were the same as was the motor output. However, the decision, the meaning of the decision was quite different because this time we told them, no matter what you choose, both NPOs will get $500. So you can your choice doesn't matter. Whatever button you press is completely arbitrary. So if we look at the neural recordings from that we, we can see if you look at the arbitrary decision we can see a downward trend that's reminiscent of these readiness potential. But we cannot see the same for deliberate decisions we see it for arbitrary we do not see it for deliberate for and the statistics seems to confirm what you see here. So if you look on average over all the subjects you can see a nice readiness potential for arbitrary decisions, but just a slow downward trend that's that's not a readiness potential for for deliberate decisions and again the statistics confirms this. We ran various control analyses to make sure that these results are not the outcome of some trivial, some trivial tweak or something like that. And they were not. And interestingly, we also looked at one more thing called the lateralized readiness potential. That's a potential that's more related to motor events. And there you see that arbitrary and deliberate decisions, there is no difference. So this really chimes in with the idea that the motor output here is the same between arbitrary and deliberate decisions. And further on, we, we worked with some modeling, because recently there was a claim that this readiness potential is is maybe is maybe actually artifactual. So we took that model that was that was done by Aaron sugar and others in 2012. And we extended it to our case and we, I mean, it fit our data quite well and and and from that, we that model really predicted this slow downward trend in in deliberate decisions, but to really see a readiness potential in arbitrary decisions. So we took that model that was that was done by Aaron sugar and others in 2012, and we extended it to our case and we, I mean, it fit our data quite well and and and from that, that model really predicted this slow downward trend in deliberate decisions, but to really see a readiness potential in arbitrary decisions. So, to conclude from this, we would say that the neural signals that are associated with arbitrary decisions do not generalize to deliberate decisions at least in our experiment. And if you think that the limit results pose a threat to our notion of volition. This threat doesn't seem at least again from our results to generalize to deliberate decisions. Now, I would arguably say that we care more about these deliberate decisions, the, the person you're going to live with things that you think about, especially for a long time then, you know, which hand you're going to raise. Certainly if you want to talk about more responsibility, nobody's going to take you to court for raising your right hand and not your left for no reason, and for no purpose. So, I think our results suggest that the the onus is now on those who think that the limit results generalized to deliberate, generalized to deliberate decisions to show that they do because our results suggest they do not. And this is just the the paper that that was discussed here. If you want to look more into this, of course, this was just a brief overview. If you're interested into the neuroscience of volition. Let me just say one more thing that there is now a large international project that is happening. Looking into how the brain enables conscious causal control of action. We've got people from all over the world working together neuroscientists and philosophers you can see more information here. And I have if I may I'll just say that like everybody, we are seeking postdocs PhD students and RAs and you could contact me and either one of these addresses for more information. Thank you very much. Hey, thanks very much indeed very very interesting indeed gets right to the heart of who we are and what we are. So, as I say, questions, you can post them in the chat or you can post them on the document. And so I guess I'll start with asking already a couple while we're asked what were any from the audience. So, firstly, did they choose the same things for the two types of decisions seems like they might have done to me or. You mean, you mean the same nonprofits. Yeah, exactly. So, you can imagine non deliberate decisions that are nevertheless favorable, because you have simple provolovian effects towards things that you like etc etc so I'm interested to know whether they whether they whether those are what were driving the decisions. Right so for. We can actually for if you look at the consistency. I guess this is what you're asking about right the consistency of the decision so you can see that there is nice consistency for deliberate decisions but there was so I'll explain what I mean by consistency. I didn't go into it in the experiment but there were two parts of the experiment in the first part subjects just rated on a scale of one to seven. So, how much they would want to donate $1000 to it, and the consistency here is, if, if something that was let's say a three came up with something that was a seven you'd expect them to choose the seven over the three so how consistent were that. So you can see that for deliberate decisions, the subjects were quite consistent for arbitrary decisions. They were very close to chance, however, if you look at this cross here they were slightly slightly off chance so they were good just a little bit yeah so. So if something came up that you really don't like versus something you really like even though it's arbitrary doesn't matter which button you press, it's going to go 5050 to both of them you still had a slight slight preference for the thing that you, you preferred. So, I'm just while waiting. It seems to me that is it the case that the deliberate decision is less tied to an action somehow. So, in fact, maybe you've got that here as well so is it the decision that is it the case that the deliberate decisions are taking them longer or. Or somehow there's more delay between the decision and the, and the action on the deliberate decision than there is, or, or so for example, and so if that's the case then your ERP can be smoothed over a longer period of time and won't show that. So, let's say one trial is really rapid and the next trial takes a lot longer than once you start averaging the trials together. It's going to be a very different thing from things that are very precisely timed is it not. We've that's one of the controls that we did we we looked into I mean one concern that we had ourselves was was this due to the reaction time differences. So we did various things to look into that we we took only the longer arbitrary trials and only the shorter deliberate trials and we you still see an effect and we also looked at whether there is a trend between the difference in the RP between deliberate and arbitrary, and we could not find an effect and we did this both within subjects and between subjects and there were no effects in both situations so it's it's it seems that this does not explain our results but yes that was something that that came up for us also is something one of the first things that we checked because if then it's less interesting. It's not driving our results. And yeah cool and interesting and so zero here is the time and they actually make the action right. Yes that's when they press a button. Yes. So you would, in some sense, in some senses you would really expect that to be a bill of a brain activity in the deliberate decision before they make the action. It were. So did you measure the time at which they report to have made the decision to see what I mean. That's, I mean, so we didn't and we didn't do that for two reasons. One, I don't think that the right now there is a good measure of when people make up their mind limits clock is a very flawed idea. There are some other things we're actually trying to develop some of those things in my lab but. And that's one reason the other one was that we really wanted to try and keep this more like the real world. So it wasn't. It wasn't. I mean, asking people when did you make up your mind feels very unnatural to us so we did not ask them when they made up their mind but in other experiments, when we try to compare between arbitrary and deliberate decisions about when people make up their mind we found some differences in the, I mean, this was not like the one second difference that we see here this could not have explained these results. That's really interesting. So we're going to move on now but thanks very much for a great talk. So, we're moving on to Stefania now who's going to tell us about categorical representation from sound and site in the ventral temporal cortex of sighted and blind. Thanks very much. Okay, hope you can see my screen. Yeah. Hello, everybody. My name is the stephania maternity and I am from the UC live and postdoc at the UC live in Belgium. And today I will speak about the categorical representation in the occipital cortex of sighted and blind. The world around us is full of rich and dynamic visual stimuli and to avoid an overload of information our brain tends to group them into categories. The main one are the animate and animate and then also it creates some other clusters within these two. The main region known to be involved in the processing of these visual categorization is the ventral occipital temporal cortex, which is a region in which we find clusters of boxes that show a selective preference for one of these categories compared to the others. What drives the categorical organization of the OTC. In other words, what makes a part of this region to prefer stimuli such as faces and another part for example to prefer stimuli such as houses. There are actually two main points of view. The first one is that the low and middle level visual properties of the stimuli are actually driving this categorical organization. The eccentricity or the rectilinearity properties or the real world size so being small or big. On the opposite side, we have the idea that some more high level functional properties of these stimuli not purely visual might drive this categorical organization, something like the animacy of the stimuli or their categorical function. This is not easy actually to test this second hypothesis because it is hard to isolate the high level function from the low and middle level proper visual properties of these stimuli because normally all the stimuli share these low and middle visual properties within each category. A good way to test this hypothesis could be to present the categories not in the visual, but in the auditory format, and to present them in a group of people that are born blind. Another hypothesis was that if the categorical organization of your PC is not driven purely by visual properties, but also by high level functional properties of this stimuli. We should find the similar categorical organization also in the VOTC when the stimuli are presented as sound, and even in people that never use vision functionally. In this hypothesis, we included three groups in our study, a group of sighted people that perform the visual version of the experiment, 17 early blind participants that perform the auditory version of the experiment, and then additional group of sighted that perform also the auditory version of the experiment. And we presented our participants different stimuli from the main animating animate categories, including humans and animals, manipulable object and big object categories, each of these four category was including two subcategories as you can see in this slide. And we presented the stimuli while the participant was lying in the fMRI, either in the visual or in the auditory format, according to the version of the experiment. And then we looked at the reaction of the brain to this stimuli in the three groups. With a special focus, of course, in the VOTC. So the first question we had was to look at the general topography of this region, would the topographical map of VOTC be similar across the three groups. So just this, we basically select each voxel each small part of VOTC, and we assigned to it to one of these colors according to which categories this voxel was preferring more than the others. So when we look at the map that we get in the sighted for the visual stimuli, we see the classical map, the well known map with the medial region in blue preferring big objects, and then a bit more laterally in pink, the region preferring face. And then in yellow, the region preferring animals and finally most like laterally in green, the region preferring manipulable object. So what is interesting to see is that we also found the topographical map, when the sound, the categories were presented as a sound in sighted. And even though these clusters are not exactly in the same position as the one that we see in the visual map, we see that there is still a significant, highly significant similarity between these two maps. This similarity was even more striking in the case of early blindness. So this analysis is actually telling us that there is a shared similarity, at least in the topography of VOTC. But we wanted to go more in-depth to understand if, for example, VOTC really contained information about our categories. Would VOTC being able to discriminate each of our subcategories to the others, for example, to address this question we use the more advanced multivariate fMRI technique, in which we basically compare the brain activity for each of our pair of categories. And we repeated that for each possible pair that with the eight categories means 28 piracy decoding. And in here, higher is the gray column, higher is the ability of VOTC in the subject to discriminate the two categories. And then we repeated this analysis also in a controlled region, the early visual cortex, which is a region known to process low level visual properties. And we'll be the control region in most of the analysis we did. So first of all, what we did, we averaged these 28 values to look at the general ability of the VOTC in our groups to discriminate these categories. And what we found is that in the visual experiment, as expected, both early visual cortex and VOTC show a high accuracy in the discrimination of the categories. But we were more interested in the auditory experiments and here are the results. What we found is that, first of all, in the early visual cortex, only the blind can decode significantly the different categories significantly better than chance and significantly better than the sighted. But the interesting part for us was to look at VOTC. And what we see is that this region can decode the different categories both in blind and in sighted significantly better than chance. Even though we see that the accuracy in the blind is significantly higher compared to the sighted, suggesting that there is an extension of this ability in the case of early blindness. So this analysis is actually telling us that, yes, VOTC contains information about different categories, not only when they are presented visually in sighted, but also when they are presented in the auditory format both in blind and in sighted. Does this mean that the functional profile of this region is similar across the groups? First of all, what do I mean with functional profile? If we go back to the 28 paywise decoding, what we see in each group is that there are some categories that are more easily decoded compared to others. And we can consider this as a sort of a functional profile of VOTC in each group. And that we can also represent in the format of this similarity matrix in which we see basically in yellow the categories that are easy to decode than that they are different. And in blue categories that have a low decoding accuracy so that have a similar representation. If I quantify how similar or different is this functional profile across the groups, I can correlate these matrices. And if I plot this correlation, we see two important points here. The first one is that the correlation is significant in the three cases, meaning that there is a shared similarity in the functional profile of VOTC across the three groups. And here we see that the similarity between the sighted in vision and the early blind in the auditory experiment is significantly higher. Suggesting again that the functional profile of VOTC between the auditory and the visual stimuli is more similar between sighted and blind that even between the two sighted group. So to summarize what I show today is that the organization that the categorical organization that we find in the VOTC of sighted people for visual stimuli is at least partially preserved also when the stimuli are presented in the auditory format. And it is even extended in case of early visual deprivation. I showed you that the topographical map of VOTC is shared similarity across the three groups and that the functional profile of this region is also similar across the three groups. So unfortunately, in 10 minutes, I cannot, I couldn't explain all the analysis we did, but we also looked more in detail in our work at which mechanism could explain that. And also at the differences between blind and sighted. So I invite you to go and check our just published paper in Eli. And with this, I want to thank you for your attention and my supervisor my group and Eli for this great opportunity that gave us to share our work. Thank you. Thanks very much. Sorry, I just had some trouble on meeting myself. Thanks very much. Fab talk. Yeah, brilliant. So, again, if anyone has questions they can play post them in this chat down here or in the Google Doc, which is cited in the in the chat. Brilliant. Okay, so maybe as again I'm just going to start the questions and it's fascinating data. So what do you know how there is we know there's some organization by digital properties as well right so we know that from Doris sales work, looking at the basic shape properties across and for temple court is in monkeys and we know because just if you look at convolutional neural network representations they look similar to the representations in it or in natural LLC regions in in humans right. So, um, given that you know, so I guess there are several things one one might be okay earlier on in the infotainment cortex. There's visual information and then later on you get towards the more semantical meaning based information or it could be that evolutionary time scales it's been driven by visual information but then it can be when someone's in the map, it still maintains the same innate representation. Do any of those ideas. Trying with you or how do you see it. Yes, I think that is kind of a mix of the two. So what I believe is that the early visual cortex is a for sure a visual region and then going more like anteriorly we can, we start to find more semantic information as you say, but still we see also like my data showing that also in the site did we already find some kind of information also for non visual stimuli in these VOTC regions or more anterior region. And what I believe is that in the case of early visual deprivation so when we have a specific experience of visual deprivation. This ability of VOTC, this innate organization of VOTC just passed through another census and it becomes this auditory representation become even more extended. But yes, I believe that here what we are seeing is a bit a mix between the innate functional organization of VOTC that is definitely there for processing categories, and then also some change that experience applied to this to the nature of this region like starts to be more processing more non visual stimuli when vision is not there. And so here's a I wonder if you could try. So thinking about a different way so if you if you if you can, if you looked at the parts of the representation that are different between the blind and the sighted people. Do you start seeing any anything that looks like an organization by the sound or by the by the frequency spectrums or of the words that are used to describe it. There are two things I can say there we checked like specifically we test some representational model based more on categories or on low level properties of the sounds for example in the blind. And what we found is that in the blind we see that the functional profile of your to see it's not explained by free by the pitch of the sounds, for example, but it's more explained by a more high level functional categorical organization. I also looked at something in, I don't know in here you can see like a clustering hierarchical clustering analysis. What I find very interesting is the first line in which I'm basically I'm asking the algorithm to cluster the data in the two groups in each two groups of categories in each of my group. And do you see that in the sighted for visual stimuli that is the left corner, we have really the animating animate distinction. That is red is basically or the manipulable and big object and in green we have animal and human. While in the blind what we find is that the representation the organization is mostly human sounds versus all the rest. Oh right cool so yeah right there so there is some property yeah. Yeah. And so this is very interesting because I don't have this data here but if we look at the temporal regions so they are really like region known to process sounds. We find there, also in sighted that the sounds are more like clustered as human versus all the rest. It seems that this representation in the blind is closer to the one we have in the temporal region. Yeah, right. So some some kind of mixture of a of a learned or two representation imposing on an innate visual representation or something like that. Yes. Very interesting very cool talk. Thanks very much indeed so I guess we have to move on now. And so we're moving on to an anthem. Who's going to be talking about the role of Roger in Fibronectin. Hang on. Great. I'm going to let an answer give me her whole title because this is a lot further away from my research area. Thank you Tim. I would like to start off by thanking Eli furthest forum to discuss my research. So I'm an answer from Professor Harry's Harry Miller's group in University of Bristol and today I would like to talk about this particular protein Roger and its role in the process of formation of new blood vessels something we call angiogenesis. Angiogenesis is actually actually a very complicated multi step process, where an endothelial cells the cells that we see in green here are supposed to undergo a dramatic change in shape and polarity, whereby they acquire a lot of new properties that help them to respond to the call of tissues to make new blood vessels which can come out in the form of group factors. So this change is also accompanied by the ability now to remodel extracellular matrix which is lacking in the quiescent blood vessels. So this is something that we are really interested in. And we are interested in a group of proteins that enable cells to generally change shape and polarity, which are the road GTP is family of proteins. So I would like to grow from here to one of our most intriguing observations, which is on this family of proteins road of road GTP is of which the most well studied member is this protein CDC 42. It is well expressed in all tissues of our body. However, it has undergone duplication in the vertebrate lineage to give rise to Roger, which is much less studied and is only present mostly only present in the endothelial cell so it's a lineage specific protein. As we can imagine, being duplicated, these two proteins share quite a lot of feminists similarity, and we would expect them to function also in a similar way. However, we found that they are actually working in diametrically opposite directions when it comes to the ability of cells to remodel the extracellular matrix and make new fibres something we call fibrelogenesis. So how did we come to this observation? We had initially started off the project to understand Roger better, which is a much less studied protein. And we just looked at how Roger and CDC 42 can be distributed in a primary endothelial cell. And what strikes us immediately is that Roger is present in these vesicles, these large vesicles as well as small ones all over the cell body which is a distribution quite dramatically different from CDC 42. In order to utilize this distribution change to uncover new roles of Roger, we took a rather challenging approach to purify these vesicles from primary cells containing Roger and see what sort of proteins are enriched in these vesicles. So this is an approach that we call vesicle proteomics where we can separate out the plasma membrane pool of Roger from the vesicular pools by using a density gradient ultra centrifugation and follow that up with pulling out vesicles containing Roger which was tagged with GFP using a particle sorter and then take it through mass spectrometry to understand what sort of proteins are enriched in Roger vesicles. So this approach gave us a lot of integrins which are proteins that bind the cell to the matrix. And one of our top most hits is this protein alpha five beta one integrin. And this integrin binds to fibronectin as it's ligand and is important for the process of antigenesis. We were very happy to see that actually the co localization at the vesicular pool of Roger and alpha five is very good. In fact, 75% of the vesicles co localized suggesting that our approach to purify Roger positive vesicles have worked. Now having these two proteins together in the same vesicle may mean Roger can regulate the trafficking of alpha five beta one. Or they might be just passively co trafficking. In order to distinguish this we actually modulated the levels of Roger and saw whether it affects the levels of alpha five beta one or its distribution in the cell. Unfortunately for us with multiple approaches, Roger did not alter the levels or distribution in multiple different approaches of alpha five beta one. This suggested that maybe Roger is just passively co trafficking with alpha five beta one. However, literature suggests that actually alpha five beta one comes in two different flavors. You could have this bent integrins which are not bound to ligands or you could have this stretched out integrins with their tails apart which are actually bound to ligand. Now we know from literature that the active integrins are a very small pool of the total, but they are the ones that actually spend a lot of time in vesicles because they have to get rid of their ligand at the low pH of vesicles. And then they can be reused. So basically, we then asked, okay, is there a change in the active integrin alone when we modulate Roger? Here indeed, we could see that when we deplete cells using an SIRNA with Roger, we can see there is an increase in the level of active alpha five beta one alone. And we have seen that there is no change in total. So this really means that there's a confirmation specific change with when we alter Roger. We have also taken multiple approaches to prove that Roger is indeed confirmation specific. And I then had the opportunity to do some cool trafficking essays with Professor Jim Norman at the Beesons who has pioneered this essay wherein we can actually look at internalization and chase the fate of integrins by biotinulating them at the plasma membrane and watching and taking an approach wherein we can see how they are internalized into the cell at different proteins. When we deplete Roger, that's the green line here. We can see that the internalization layer rate alone is actually doubled and this is true only for the active conformer and not again, it does not affect the total alpha five beta one. So basically this these experiments prove that the internalization the point at which the receptor enters a cell is modulated when Roger is not there and Roger seems to be putting a break to this process. Now why would this be important for an endothelial cell? We know from literature that this process of trafficking is really important to build the fibrils. So that is coupled to the ability of cells to actually direct fibronectin to places where these long fibrils would be waved. So from this data our expectation is that when we put a break on this process which Roger does, there would be much less fibrillogenesis. Indeed that is really true. When we have depletion of Roger, there is the ability of cells to lay out these fibrils is markedly improved and the converse is true when we express Roger in its active state. Now this was a point where we were again surprised because as I mentioned Roger arose from the ancestral CDC 42 gene. And here we know from literature and as well as our work that CDC 42 is a positive regulator of fibrillogenesis. If you deplete CDC 42 you get much less fibrillogenesis. This brought us to the question how do two proteins which are very similar to each other regulate the process of fibrillogenesis in demetrically opposite directions. To understand this we first thought that Roger might have unique binding partners in the endothelium that might help us explain why this is true. So we undertook a study of binding partners in a quantitative way where we looked at Roger and CDC 42 binding partners in endothelial cells using TMT mass spectrometry. Again we were very disappointed to find no unique partners of Roger. Actually Roger bound to pretty much everybody that CDC 42 bound to. In fact we could find CDC 42 binding partners that were just unique to CDC 42 but not Roger. It's more recent protein. So the protein we decided to follow up was this protein Patree which bound really nicely to Roger but even better to CDC 42. To cut the long story short Patree actually behaves exactly like CDC 42. It is a promoter of fibrillogenesis. So Patree alone enables cells to form fibrils as also quantified here in a different way. But if you put Roger on top of Patree it changes direction in the sense Patree alone drives fibrillogenesis. Roger plus Patree is an inhibitor of fibrillogenesis although there is a slight improvement from Roger alone. So what all of these data together suggests to us is that Patree might be quite limiting in the cell. If Patree is available to CDC 42 there is more fibrillogenesis and when Patree is taken away from CDC 42 by Roger the process is inhibited. So basically two proteins which are very similar to each other can also drive processes in opposing directions through the same shared effectors if they can compete for the same effectors. Of course all of this data is actually from in vitro growth of primary endothelial cells. We wanted to know whether this is true in vivo as well. So we collaborated with Professor Akiyoshi Meira in Japan and they have done this beautiful experiment with Roger knockout mice where we are looking at pictures of retinal angiogenesis. And we can clearly see here that the fibrillator deposition at the angiogenic front here is much more when we do not have Roger. This suggests that indeed Roger is a negative regulator of fibrillogenesis. So what does this all mean to angiogenesis per se? We think that Roger activity has to be actively inhibited when the cells have to form these fibrils when they are sprouting and this is exactly what one of the growth factors does. And indeed it has to be reactivated when we have to again stabilize the vessel and form lumen. So this is a brief overview of my work in Professor Harry's group and it will appear online in a couple of days. So kindly do check out our paper to know more about this. I would like to stop here and I would like to thank the left firstly for this opportunity and I would like to thank my mentors. So Harry Miller and Professor Jim Norman and our funding bodies and collaborators. Thank you so much. I can take questions now. Hey, thanks very much indeed and so I can see that that I'm not brave enough to ask any of my own questions to you because it is a long way from anything I know about so I'm going to ask a couple of questions that you've kindly prepared for me. So first, what's the role of other PAC family members PAC one and PAC two in this process. Thanks Tim for asking the question so basically pack three the pack three is a part of a large family of proteins pack one to six. So basically we found that pack one which is also highly expressed in the three cells also might play a role as it's also competed for but to not to the same extent as pack three so pack one is also important, but back to does not seem to. Even though it's expressed in the three cells back to is not important. And so, Shivani has a question for you, Shivani, can you unmute. So is the phenotype of Roger knock out similar to CDC 42 in mice. Well that's a really nice question. So, actually CDC 42 being the ancestral protein it's way more important than angiogenesis and Roger so if we knock out CDC 42 the mice have a very early embryo embryonic death because they cannot make blood vessels right from the beginning. Whereas for Roger knock out mice the effect is much more delayed so they have a delayed angiogenesis phenotype in the adults, and this actually said this difference actually suggests that they have unique functions. So, which we have not understood as yet, but indeed as an ancestral protein CDC 42 has a much higher effect on angiogenesis than Roger and they don't seem to be compensating for each other. Thank you. Cool. So, another prepped question from an answer. So can it happen that Roger and CDC 42 are active simultaneously and if so, what happens to five, five brilliant dentists is actually before you answer that we have a new question, but maybe you should answer this instead because it's not from you. So it's from Vinod. Vinod, can you unmute yourself. Yes, I'm unmuted. I hope you can hear me. It's a very nice talk and then actually a collaborative initiative. I was wondering if the Roger mutant mice are actually viable because I thought we're showing the retina in one of the slides. Yes, if they're viable, are there other factors that can compensate during development. Yeah. So like I was mentioning in the previous answer, basically Roger knockout mice do not have any phenotype until they reach adulthood. So there is a slight delay in angiogenesis. So they are completely viable. They are completely fertile. The mice are absolutely fine. So we believe that the CDC 42 knockout mice are actually embryonic lethal because of angiogenesis and that does not seem to be compensatable by Roger. So the delay in angiogenesis that Roger cause is of course in CDC 42 wild type background so there seems to be unique functions that these two proteins do not compensate for. Great. Do you want to go back and ask about whether Roger and CDC 42 are active simultaneously and if what happened if so what happens. That I should actually say that that's a pretty lame question from my own side, but yeah so what we know from very recent studies actually our collaborators are also doing work on how Roger gets activated and not much is known in the field as such. So basically, what we think is that there is this a temporal sequence so we need Roger to be really active in the quiescent blood vessel and that's important because if you have fibro, neck and deposition in a quiescent vessel that's very a heurogenic. So you don't want fibro, neck and fibro genesis to happen in a quiescent vessel and then you want it to happen suddenly when there is angiogenesis and then you have to stabilize the vessel again when the vessel formation is complete. So we believe that this there is a switch between the road GPS is at all of these points and they should not be simultaneously active basically. So this is what the most of the growth factors do Vegev activates CDC 42 inhibits Roger and there are now more reports from our collaborators that there are more stimuli that kind of switch between the two road GPS is cool brilliant. So actually I think we had one unanswered question for Stefania maybe I grabbed all the questions without noticing one name is that right. This came in in the dark afterwards, but if I mean all those talks were fantastic and if we like we can open up for the speakers now. Yeah, good plan. So, what was the question for Stefania and is the participants still online. Yeah, it's from Vinod Vinod you can, you can speak again if you like. Oh yes. Do you hear me now. Yeah. This question for Stefania. So this was on if people can actually identify localizations or sounds of animals that they haven't actually encountered before. For example, I was thinking more about kangaroo sounds, which, you know, for people in Australia maybe that's very familiar but for someone who's outside Australia they might not have heard. And do they categorize in similar regions. Do you mean the if the animals are categorized in similar regions for the sounds in blind and the visual stimulation in sighted. Yes, exactly. I mean, for me personally, because kangaroo sound sounded more like a human was doing the noise. So I thought maybe that's human. So that's why I started thinking about this. That's a nice question. Actually, I don't know because I didn't have any kangaroo sound in my study. And yeah, I cannot like respond to your exact question what I can tell you is that the animal category in general is the one that seems to act more differently in blind compared to the sighted. I have no idea but this is pure speculation is that actually we use vision to discover to learn animals and also like a kangaroo I never saw a kangaroo in my real life but I saw on the television, for example, like videos of kangaroos moving and I can recognize it and as an animal. What blind people normally do they either heard the sound, or they learn the shape of these animals using small miniatures and then touch. So static miniatures of course, so either they put them closer to to object because they discovered them explore them more as object, or they put them or similar to the environment because for example they hear the sound. And they probably integrate it with the wind sound or with the other like nature sound, but specifically about the kangaroo that has a sound more similar maybe also monkeys can have sometimes a sound more similar I don't know. I pitch and I don't know maybe yeah this is something that should be investigated I think in future. So, so I think we're going to wrap up now for the main event I think there might be some discussion afterwards but so quickly as I just say thanks to worries to find you an answer for sharing their work. And to everyone who's still still around for listening we had, we've lost a few participants we have more earlier. So, our online research talk continues on Thursday chair by we go and we'll have from her and Karolina and Phillip. And so if you want to get updates on this series if you just follow elive community on Twitter, then all the information's there. Yep, and great and so we're going to keep this zoom open for a little longer if there's any further discussion so people got more questions or if they just want to listen to any more general chats between any of the speakers. So, that's great thanks very much. Okay, Naomi, how do you want to say this point. I just wanted to say that with this is our last week of the schedule talks so far. But we would like to continue this, not quite the same frequency, but we'd like to continue to offer slots for people so anyone who'd like to speak and hasn't had a chance yet. If you want to go to the web page, which is here this light green web page. There is a link in the first paragraph to register your interest and that is what we can then email out to you all to say when we've organized how we'll invite the next speakers. We can, we can say to everyone on that email list. So please do register your interest if you want to speak in in upcoming talk slots. And other than that, I mean I think that's just Anya has posted lots of things in the chat people to follow different news. If you have any questions for the speakers there's not very many people left online at this point so I don't want to take people's time if there's no more questions but perhaps if the speakers have any questions for each other. If anyone has any questions we can just move to that now and otherwise we can end whenever we'd like to end. I'd also like to say thank you. Thank you Tim for chairing. Thank you. It's very, very smoothly organized as always. Thank you so much. Thank you. Thank you. Excellent. So do people want to hang around I can continue asking in name questions if people would like me to or we can all go about our day's work. You're welcome to ask if you want to. So how do you think, if there's no build up to a to a voluntary decision, what's happening in the brain that makes the decision then. So I don't think the decision is made in the kidneys, but I mean, I think there is a buildup in the brain that leads to decision. If you if you run some kind of classifier on the brain you can see that you can you can predict better and when you're closer to the decision. However, if you look at the at some of the studies that have suggested that you can tell that what the decision is going to be a second in advance or maybe even you know there's some studies that claim maybe even 10 seconds in advance you can tell with better than chance that there's basically there's some information even 10 seconds in advance whether you're going to raise your left or your right hand. It's a little weird so if, if, if the idea is that the brain that the brain doesn't know but that that there is clear information in the brain that the decision has been determined 10 seconds in advance so what's the brain doing for 10 seconds just waiting for, for what exactly right. So, maybe, I think the people that have said that have maybe suggested something else which is something about the state of the brain has a probabilistic influence on what the actions are. Right. I mean, and so. There are different interpretations. And of course, I mean, even those studies that suggested. You could tell sort of, I guess you could tell six months in advance, whether I'm going to choose to eat something hot and hot and wholesome or something light and summary. You know, yeah. Yeah, no, I get I agree it's not all about prediction, of course. But there is, for instance, some, some information when we try to really look at predicting. More carefully, it might be that you can really that the prediction ability really rises just close to movement onset and not really well in advance to have movement onset as was thought. So that's, I mean, one option is that the other one is that, even if it's not specifically the readiness potential that is predicted upcoming movement, it could be that there are other things in the brain, you know, that it's not I mean that that specific signal is maybe you know, it's been known since at least 1965 and it's been it's it's kind of like where people are well versed in it but it doesn't mean that that's, that's necessarily where we need to look it could be that there are other things in the brain that are more predictive of upcoming actions and the So if you look at like, I think it's dominated this but so if you look at the more than the decision making literature and you'll say okay maybe one decision is about an action itself but this decision isn't about the action the decision is about the the good or the charity and so maybe it's those regions that some different region like the maybe all front record is rather than the action system is making decisions about those in those different frames of reference and maybe And so you're just not detecting them with your with your particular sense on the EG is that right? Is that what you're saying? Yeah so yeah one thought that we have is it could be some ventromedial area or something like that that is involved and we're actually now trying to do an fMRI version of this similar experiment in fMRI to try and really see if there's something that's as you said it's not it's just not accessible with the EG and we could find the information there again I really think we're going to find the information somewhere in the brain it's not just coming out of the blue but it seems at least from our results it seems to be not the readiness potential that's the important thing here and I think again I think more than just saying that this is readiness potential or not I think it's just this idea that arbitrary and deliberate decisions are not the same and the fact that you can predict I mean if you think about it if you can predict if I'm going to raise my left hand or my right hand all you need to do is is basically I mean there needs to be some kind of symmetry broken in the brain so if you're the fact that you can do some kind of symmetry breaking that maybe maybe even something like the readiness potential is just a matter of symmetry breaking you need more than that to predict a deliberate decision right then there's values that go into that and you know the vibration back and forth and so on so we were just trying to highlight and hopefully we did it in a good manner to highlight that there are these differences in between arbitrary and deliberate decisions and the fact that you can maybe predicted an arbitrary decision doesn't immediately mean that all decisions are like that and hopefully if I convinced anybody of anything maybe I've convinced you that so I've done that I've done my job So Safanya how do we so it seems like as you move away from anterior to LOC or something like that we're going to need a different way of thinking about the organization of the anterior parts of the infertile cortex How are we going to get that? What are we going to as you move towards the hippocampus we're going to need something that knows an awful lot about what those objects are and how they contribute to our ongoing behavior if we're going to understand those representations So it feels like the moment the best we've got for understanding representations is just what they look like you're showing that there's more than that other people show there's more than that as well but where do we go next? So I think what would be interesting is to try to put it in a more general view so for example to look at the mechanism explain this and why the brain does need a region that is more posterior like VOTC so that is still involved in visual processing and why this region would need to process also auditory information for example what is the function of this and yeah our idea is that yeah the brain is all we know that the brain is all connected and that there is a sort of intrinsic connection that probably might push these representations So we looked at the connectivity of the brain of VOTC to the rest of the brain and what we found is that yeah there is a sort of feedback probably information this we don't know because we didn't check the direction but that might push might create some networks for faces for human voices all connected together and making these regions speaking to each other and then creating sort of brain networks for different categories for example such as human or like environmental sounds and visions visual stimuli and so on so I think that's the way we should look now to understand better how this process is more integrated in the rest of the brain Cool okay I'm done with questions now so if you guys don't have questions for each other then I think we should wrap up Thanks very much again Thank you Thanks everyone