 Welcome to the second part. This is the panel. My name is Nankong. I'm Professor of Interim Head of Biomunica Engineering and We have the privilege to have Dr. Huda Zakhbe to be here And she's going to participate in the panel together with a few Purdue faculty and let me introduce the moderator is Professor Aaron Bellman Thank you very much. All right. Well, let's have fun with this panel Let me call the panelists up. The first is already introduced our invited speaker and guest Dr. Huda Zakhbe, investigator of the Howard Hughes Medical Institute at Baylor College of Medicine as well as founding director of the Jan and Dan Duncan Neurology Research Institute NRI at Texas Also my former postdoctoral advisor. So Huda welcome and please take whichever seat you feel most comfortable in Our additional panelists are Dr. Tamara Kinzer-Ersim Who is Associate Dean of Graduate Educational Professional Education for the College of Engineering and the Martha Gross Associate Professor of the Weldon School of Biomedical Engineering. Welcome Tamara Second Purdue faculty panelist is Dr. Maria Macon Hello Maria, I think this is the first time that we've met Who is Assistant Professor in Biomedical Engineering Next we'll have Dr. Chris Rochet Professor of medicinal chemistry molecular pharmacology As well as director of the Purdue Institute of Integrative Neuroscience here on campus and finally the the fourth Purdue faculty panelist is Dr. Rihishi The Mary Holman George Endowed Professor of Applied Neuroscience Professor of Biomedical Engineering Director of the Center for Prolacist Research and the Department of Basic Medical Sciences in the College of Vet in the Vet Med College Welcome to our panelists, and I guess I have a seat as well I don't need anyone keep this here as a backup Mike. You have a mic? Very good All right, so We do have some starting questions So the the title sort of the focus of today's conversation is the future of Neurotechnology in the discovery and treatment of neurodevelopmental as well as neurodegenerative Conditions, and I think we'll go around the room the panelists just to say a few things Just maybe we'll move down the line from left to right from the audience perspective We'll some go some the opposite way. So who do that'll leave you in the middle most of these So the first is on what in your view is the next big leap in neuro technology? And how do you see it interfacing with translational neuroscience Chris? All right so the next big lead I usually think of it in terms of How we'll be able to tackle problems at multiple scales and also breadth across the brain so thinking about different scales we can consider the Microscale, so thinking about particular cell types and synapses, but then more of a macro scale looking at full brain function and connections amongst brain regions and Neurocircuitry connected across regions as opposed to just simple Neurocircuit so more like networks and so I think a key question is To what extent how deeply do we need to understand and also Across how many different neuro circuits do we need to understand in order to really get the minimal understanding needed to then have a Strategy in place for a translation. I think that we're able to consider Really beyond a frontier with respect to how much data can be Analyzed or collected to be able to analyze just because of the advances in artificial intelligence and Big data approaches, so I think we can really push ourselves, but at the same time I think it's important to to consider What is most practical? You know do we need to analyze everything or for certain diseases doesn't make sense to just pinpoint specific brain regions? and so we could think about Neuro engineering innovations as being spurred by these types of problems, so we have Innovations on campus where we're able to look at functions of neurons at the level of individual spines now So really of getting to a much finer more granular view of how Neurons function in the context of circuits, but then at the same time we have sensors physiological sensors as well as brain sensors collecting information simultaneously with behavior and so this gives an idea of the types of technologies that allow for a greater breadth of recording I Think just so Think just so we're not sequential if anyone has any particular comments on that and then we'll jump back Just I'm going to open it up to the panel any additional comments on that or we can keep keep moving forward All right, we're not jumping in so well as it happens. I think I agree So I think you know the greatest thing that we see In neuro technology for treatment of diseases is this idea that we're increasingly getting The technology that will let us record from more and more parts of the brain and record chronically from individual patients So that's the first step the second step is that there has been improvements in real-time analysis, you know So treatments need to be closed loops So for example in Parkinson's used to be the case that you do deep brain stimulation all the time constantly and now We're moving towards an approach where we do stimulation when stimulation is needed and then the third Improvement is again what Dr. Rocher said, you know, we are developing we're working with people who can develop algorithms that can really pinpoint How can you modulate what you're doing at a single lecture at a single part of the brain and then affect what happens throughout the entire Network so I think all three of those things together will be extremely powerful for treatments for neurodegenerative generative diseases Thank you So I agree with what has been said and perhaps if I want to add a little bit in the realm of hopeful and Where we could go maybe in 10 years or so I think Mapping and knowing all the nodes in different circuits and all the way to bypass a Deficiency in a particular circuit so we need a lot of basic science is really important. I simply view Fixing the brain for a variety of diseases. This won't work for rats because you have to fix the whole brain and rat But for many diseases, it's usually one part of the brain That's really the lesion for many of the terrible catastrophic epilepsy it's usually one part of the brain and Similarly in strokes and other things if we can understand ways to bypass the area of deficiency and restore Activity just as Google Maps will find you a different route when there is traffic and congestion in one area I think the more we map them and the more we know these critical nodes That cover redundancy in the system if you will and are opportunity to stimulate That's really an exciting area also And I think while Electric would be one way to do it and you can do it fine-tune as you said in closed loop and so on I wouldn't put it Beyond the possible that as we know how drivers for very specific cell types We can deliver perhaps a virus in the vein so it'll go only to that cell type with single cell sequencing We're now can do that define One cell type that maybe 50 cells But they could be critical for a function if we can deliver that and then develop a small inert Molecule that can activate these cells that would be sort of in the long realm Of perhaps activating circuits in a very simple way if possible Thank you I want to just do a quick summary to keep the audience on this We have a diverse set of knowledge based in the audience and then we will move on what I'm hearing as the Moderator is the advances relate to being able to Detect neural activity being able to evoke neural activity being able to understand the Relationship between those when you do those and then lastly to be able to map to find level how those how those occur To be able to bypass or move around differences utilizing that technology So this is I think a brief summary of where we're at. I'm seeing nods and so well we will move on I'm not sure how much I can add but Because I think these are all excellent points if I was going to build off of your summary Erin I think the some of what Dr. Zagobi had mentioned towards the Antifa talk about developing technologies to non-invasively stimulate the brain and To get to those very targeted areas if we understand how we can target particular circuits in order to evoke a response to ameliorate Symptoms that is kind of this next maybe five to ten year kind of horizon and then going beyond that Also to build off of what you were saying in terms of targeted therapeutics. That would be the dream right completely again non-invasive targeted therapies to Manipulate particular brain circuits or a subset of cells and that I think we're looking more like 20 to 30 year time horizon But we should all be you know kind of as engineers thinking about what is it that we can do now to set us up to move forward on those goals So in summary just kind of electrical stimulation non-invasive targeted and then moving towards therapeutic intervention future Yeah, thank you I agree with with the exit discussion I think it's great and the other things may be slightly different from what it's discussing is that perhaps Interesting to find out that the trigger point of these phenotype How is that being tricked? Is that the genetic alone or maybe the some environmental factors or maybe the experience that you add say Trauma maybe even mild trauma How is that and also even the social structure determines also the environment to the air you breathing water you drinking So this whole thing maybe if we can figure out the way to delay the onset Say indefinitely and that's a cure right so if you can reduce so that's something. I think perhaps also could be Could be important then then looking to this perhaps that to reduce the incidence then we'll actually have a have a Founding pack. Thank you. Excellent. Yes, please. I think you raise a very important point I want to highlight for the audience You know a small percentage of the disorders we see today are genetically determined and it Another percentage is purely environmentally determined toxic stress in a child can cause a variety of Neuropsychiatric problems and some is a combination where your mild genetic background might make you vulnerable or resilient and but if you're vulnerable and you experience toxic stress or some form of Terrible event or an infection that may compromise brains function You're gonna be you know, you're gonna deteriorate. So your point is very well taken in that Finding way to modulate brain circuits that have been modified by environment and I can tell you Environment does really modulate brain circuits in the sense that people have shown Enrichment enhanced synapses Enrichment change network activity Whereas deprivation and constant stress can really change So that's purely environmental on a healthy wild type mouse So I just wanted to amplify your point that this is an opportunity for us to understand these and do intervention If we only prevented psychiatric disorders that we're seeing today. We can have a huge impact on addiction We can have a huge impact on a disease. That's pretty much affecting one in two people All right, I think that's a great point I would just add in terms of collecting data We can think of some strengths that exist in engineering here on campus with respect to wearable sensors So we're thinking very much about engineering in the brain but if they're wearable sensors that allow for a better understanding of environmental exposures in time over time how they fluctuate and similarly also with dietary Intake and variations in terms of diet and start to piece all these data together to better understand diseases first and Then later on perhaps have monitoring systems for individuals to get a sense if they have a genetic Vulnerability to in which way will their disease manifest itself in consideration of these environmental exposures Thank you, and maybe When you collect all these data, it's very important that that you need to analyze the data I and that can make sense So the data science is extremely important even the AI system What how does that make sense to analyze these to have a quick reaction or even sometimes automated reaction? Using I assistant to actually change the environment Automatically or maybe for advice for doctor, but quickly as quick as possible to correct that. Thank you So one of the potential challenges in In the future in the processes and approaches that we're discussing here May relate to at what Granularity do we need to understand the circuits? How does for example presumably at some level even the very same behaviors might be circuited slightly differently between two Individuals even potentially identical twins would have differences in their circuits that relate to perhaps the environmental How they learned it or just stochastic elements in how our circuits learn could the panelists discuss in their view where we need to be focusing in terms of the level of granularity and the and Their view of this challenge that I've outlined. Maybe we'll start on this side That's a great question because throughout the history of discovery of science and medicine I think that these level very important Of course, we have a holistic treatment then training things like that But on the other hand, there's a lot of discovery that some of these disease can be traced down even just a single channel a single protein mutations I think these are In all levels important and the more specific we can get the more target treatment We can have like dr. Zogabies Research is sometimes these involve a certain task. There's only small group of the wrong So that could be maybe a target that which means it's you have a great level of a granularity So it's a need to look through these specific cells. Yeah, thank you Yeah, so my answer is going to be Biased by the fact that I focus on molecular level events But really I think Dr. Zobig's talk today the way you laid it out in terms of here's what we know from the human behavior mouse behavior Molecular biology behavior focusing on all levels of granularity is actually going to be important because We see the manifestation of these diseases at all levels. There's a mutation in a single protein or in a particular loci and Then that manifests itself all the way up through the way the circuits are wired it the way the behavior is then expressed So it's important that we have research and we have funded research and focus on research at all levels And technologies we see really major breakthroughs in In new knowledge in generating new knowledge when we have new technologies that we can come on board So optogenetics The ability to map circuits non-invasively The ability to sequence single cells So I think there's there's work to be done and exciting developments at all levels Thank you. I'll pass on to who to what on a comment going in between this discussion of all levels though I would say who does I think some of the work that you presented today part of the reason why I Wasn't present I'm going to speak for others a few others By the rigor of your findings was because you applied those findings to cohorts of animals cohorts of cells and got Responses and predictions and understanding of mechanism that spanned multiple individuals So this leaves off one of those levels of granularity. I maybe speak to this as opposed like you mentioned actually seeing a rescue in one mouse is Impressive, but seeing it in multiple is more impressive. But what about the differences in responses between? Individual mice, so I'm just gonna use that as a passing it on to you He passes the tough questions I Totally agree with you and to amplify your point Aaron There are many studies now done in mice where you take Healthy wild-type mice and you expose them to stress and half of the animals will Decompensate will behave poorly and the half will be great You don't even have to stress them you put them together in a cage and eventually you'll find one group resilient to depression Induced behavior and one is not and so on the many studies on that So it tells you that the context of the environment and who becomes a winner and not and so on and That's modify brain circuits. So there's a lot of variability to your point I Think with the question you posed initially the first thing that jumped to mind is that It used to be when I started neurology. They said you learn neuroanatomy stroke by stroke. It's true You at that time We didn't have a lot of imaging and we didn't understand how to study brain cell functions based on the stroke Where it was and what did the person's deficit was you start learning in brain and aty We are so fortunate now. We are in an era where humans are having recordings in a lab During surgery so many patients who are either undergoing epileptic surgery or who want to undergo brain Debrain stimulation. There are multiple recording that are happening in these people and during those recording people are People stimulate different areas To really see how is the patient gonna respond and where does language disappear? Where does it come back and so on so forth? That's one. We've got technology in imaging. It's not great It's still population level, but I think I imagine it might improve with time So finding ways to really combine what we can learn from human patients one by one But gather all these data across the whole neurosurgery suites of the world if we can put it all together We stand to learn a lot and and I think that This to me together With behaviors and certain conditions and animal model studies were really enriched our Understanding to a level where we can intervene in a meaningful way Yeah, thank you Yeah, so I think I think I agree with what everybody has said and I think one of the things that that we're learning and I think Dr. It'd be pointed out very nicely in your talk is that you know, we're thinking about different scales The question isn't you know What is the right scale because it looks more and more like all of these scales interact with one another You know So changing something at the molecular level will change things at the physiological circuit level will change things at the level of behavior But also the other way around it also goes top down so I think it's important that we have people tackling as many scales as possible and Then that will also help with these individual variations between people because maybe You know intervention at a different scale work for one person versus a second patient Yeah, I don't have much to add that was very well covered. I think That there could be if we think about the level of granularity needed as just sort of an overarching question It may depend on a particular type of dysfunction or particular disorder So if it's more focal then perhaps there's and and and also more common amongst different individuals with the disorder if that exists then There may be less of a need to understand a network dysfunction in a in a broader sense and in those instances Focusing on a more specific region and then as you mentioned Digging more deeply towards some of the protein perturbations or genetic perturbations where we can start to think about Translational strategies that might make sense whereas for the Rett syndrome case that you presented and clearly it's much more widespread and so then I think we have to balance granularity with also Extending the network over which we're collecting information and then ultimately Hopefully being able to find ways to still pinpoint parts of circuits for intervention strategies as you mentioned Okay, thank you. Um, does anyone have the time because I don't have it. I know we undercheat All right, so I think we have about 17 minutes or so As I understand I'm getting nods good So I do want to I want to be able to ensure we have some audience interaction I've got a couple more panel questions that we can go through as well, but um, it is anyone from the audience wanting to Input oh great. We have a hand over here. Hi, so Unfortunately, I was not able to be here for the lecture, which is very disappointing So I apologize if I'm repeating anything But I wanted to get you guys's opinion on you know because you asked about what the next big step was and for me I'm wondering, you know and a lot of fields the next big step is not necessarily information, but it's tooling and for Obviously, that's part of your guys's focus And one thing that's been looked at recently is the fact that a lot of computer neural networks are very good at emulating physical ones and especially nowadays I mean we've even seen You know the electromagnetic waves coming off of them and they're very similar to actual physical neural networks And so I'm wondering what you guys think about the potential advancements in that field and how that relates to Neuroscience in particular and the fact that even though it may not be a direct analog to the physical system Whether or not that can be useful in deriving You know similarities and doing further research and how You know we can use comparisons between the two fields in order to influence future discoveries Thank you. I'll let Instead of going in order. I'll just let the panels jump in on the answers I'm going to come in from a kind of a computational modeling perspective. So Um, I love this question Um, and I think it is on top of mind for a lot of us um the early kind of the early Hmm I'm not going to reuse the right word here. The early bias against artificial neural networks and what they were able to produce was the fact that they they Did not give us the ability to probe the underlying mechanisms of what was going on in the physical system So in this case the brain so um artificial neural networks are you know weighted sets of of linear equations and how do you take Those weights how do you take those linear equations? And and then probe them to understand the physical system that you're replicating and I still think that's Where the some of the frontiers in in AI so these mechanistic kind of AI? Investigations and and kind of the future of that to really again, it's comes back to to mechanism. How can you develop a simulation develop something? and then Use it and probe it to understand the physical system. Um, and that's one of the next frontiers AI from an information analysis standpoint, however, um, I think that's where we've seen a lot of really good Outcomes um in terms of of predictive Predictive AI. So, um, yeah, it's it's really exciting I think there's a lot that we can still push on And I'd actually actually like to to hear what dr So Gobi has to think about the way they may be thinking about analyzing the vast amounts of data That are coming out of laboratories these days with methods such as machine learning and AI I'll answer the easy question first I I think you're absolutely right. We're really now Focusing on revisiting some of our data and really use machine learning to look at integrated also data sets not only You know RNA of Cell cell behavior physiology. There's so much to learn that we will miss Coupled within humans when you add human genetic data, there's a lot to be learned there But to come back to your question, you're absolutely right I actually jumped a little bit over that because I felt we're already there They're already tools that are helping somebody With a paralyzed arm to use computer assisted functionality even now I understand that That when you speak even if you you can't speak Somehow your larynx is actually tremoring where the computer can read that And mimic what you would want to say So there's a lot of things that are going to come on Board that will be great assist In people who're suffering from neurological damage. I sort of leaped over that Jumped a big jump. That's why she gave me 30 years. I'm hoping 10 to 15 But she's stuck mine 30 years away because if we can now do exactly what the computer is doing If we had a better understanding of the brain And really we may not have to activate a bunch of regions of there And we may have just two types of neural cell type and The promoters are coming for all the cell types if we could do it through an IV and a small molecule So these are tools back to your point absolutely tools are the ones that make us make the big jumps 30 years to the approved clinical trial Yeah, I think it's a really interesting question and again in the context of dr. Zolby's talk and and the answer To a question that was asked regarding Where do we stimulate in the brain? I'm wondering I guess I'll turn it a little bit back to you dr. Zolby To what extent are you thinking about closed loop systems so that then they're continuing learning With the stimuli so that then the patterns of stimulation are changed with respect to as you suggested Different intensities and also different pinpoint sites evolving with time Absolutely great comment and question For red you have to use the closed loop system Because as you've seen when we've done it in the animal models, you've just done it for Two weeks and it lasted for a long time. So that tells you you don't have to have it on all the time Also, we've learned from so many study In the brain, there's the excitatory neurons 80 or so of the cells on the inhibitory neurons But those have their own closed loop system. They're talking to each other. You have too much excitation Right, eventually it's going to activate the inhibitory neuron to come back and subdue it You have too much inhibition, you know, you have to take it down So having understanding those as much as we can in which You know in every node which inhibitory neurons and which excitatory neurons are the key drivers Is going to be really the solution. So for red if we were to do dbs And I have to share this story. It's an anecdote. It's not a study But one of the ceo's of the hospital the biggest adult hospital in houston houston method is The ceo of the hospital has a child with red syndrome. She's public about it. I could say that And her daughter got more severe with dystonia as time progressed And she said look you've shown that that's brain response to the brain stimulation She can't move. I mean she's stuck and we need something and dystonia is when you have Two types the flexors and the extensors contracting at the same time. So it's very painful So she goes I need deep brain stimulation We know that deep brain stimulation can affect dystonia and healthy people. We need to do it So it's an n of one and neurosurgeons are very good. They don't need FDA approval It seems like they don't and she had it and she tells me now she can take steps and she's you know moving So the point is These things really do help the challenge for rat. You've got the whole brain What are we going to stimulate and how that's why my dream equipment is something per do is going to belt Or we can stimulate per task region and figure it out and I will work with you But I think we need something that hits the whole brain. We can't just we can't Put five transducers or six transducers in a patient I would like to add one last point. So I think you know, you're covering a lot of ground that we were going to cover Next anyways, um, so I think one of the The last areas that AI will really shine is now we've seen they've made a lot of progress in imitating a lot of the stages of neural Processing, you know, so they can imitate, you know from the very beginning to like higher level Action and so you can even find like oh this layer and this neural network Relates to this layer in the human network and I should say some of that work is being done failure Right and so I think what will be really important for also is again, we can't have too many transducers So we can use them as a sort of mechanism for simulation Right and so we can say okay What should we do and how can we predict that these circuits will react and then we can minimize the amount of testing that actually has to happen in the humans Okay, we do have a another question from the audience. Yeah. Hi. Thank you so much every panelist for sharing your insights My name is Gregori. I'm from PGD student from biomedical engineering and You touch that We don't know how the mapping of the brain is right? We we don't know where which neurons connect with each other and seems like one of the Things that will definitely help us to understand why let's say identical twins have different reactions to something why physiology may be different And maybe I'm wrong but in my opinion the solution could be to get the Start massive taking the genomes from people and analyzing the genome Trying to map the brains and so on like getting even bigger amounts of data so we can Analyze it and figure out actually what's going on in the brain and my question and I ask you to dream a little bit and tell me How do you think that can happen? How can we map the brain? So what kind of tools? Can be invented in like 10 20 30 years I'll start and I I want you to be a little bit more optimistic than you are In that we are mapping the brain. I think the brain initiative is doing a great job now What resolution we are imagine? The brain is You know the united states And there's so many highways that you could see on certain maps I think we know the major highways And for some neighborhood we actually know the small streets It depends some region of the brain are studied so extensively Even at the level of a lot of local network, but that's a small part And over studied and there's so many parts. We know nothing about so you're right But I think Just like the genome project, you know, we didn't think we can get the whole genome And now we know there's more to the genome than what came out in 2000 The tiller means to still hear us teaching us more and it continues And the same we have to do with the brain So I totally agree with you and I think we will arrive. I really think with the efforts ongoing Tools being developed We will probably get quite a bit known about the brain So that's one area. I don't I cannot predict how long it will take. I mean even for the genome even To this day, we're still learning about it. You know, we're still we're not done Um, I like your idea to really integrate and I'm glad you're the first person who brought the word genome here Because I think The more we're not going to be able to map the brain with the same intensity in people with different genetic background That's not going to be possible But From genomic studies, we can learn about these mutations That in the severe form of the mutation are going to cause severe phenotypes and then go backwards We can then every gene that has shown us in severe form because disease It's emerging in much milder form. It causes a much milder effect Very minimal maybe one symptom maybe another and as we keep moving down To really the minimal level that makes someone vulnerable Having that we can intervene early on with behavioral therapies and other things to prevent these people from having full blown psychiatric disorders So I I think It's a combination. I cannot lay it all out today I can dream but I cannot lay it all out today My dream is to get as much of the gene of the brain mapped and understand all the neighborhood But also understand all the genes within that neighborhood That can affect its function And if I known that and I genotype people I can begin to really Identify those vulnerable and prevent them from ever experiencing the disease start in second grade You know Thank you. Um, yeah, I think it's a fascinating question, but uh, um, I really like what Dr. Rock me says Today's lecture shows that they're not just a structure, but also functionalized actually it's it's linked So I think maybe my my dream. Okay I think it's probably in the near future is happening It's like a map map the brain is not necessarily just a structured genome But also functional for example, there's a google map You know how to get to dc somewhere But also, uh, there's another kind of map is how the people here in the people in Los Angeles They linked because they have they like the same sound Uh, some yeah a popular sound whatever some so this kind of a map could be also very important That's beyond transcend the physical barrier So this kind of things is the circuit to the level even with the same circuit level that you can actually have different kind of Structure, so I think yeah, that's just my Yeah, I I think we really mentioned the point that I was going to bring up that I think we're at the stage now where we're not just mapping circuits without looking at function at the same time because it's just so uh Routine now or becoming routine at least to do the same Do do both of those as to take care of both those aspects simultaneously So there's an exciting program here at Purdue being developed where the super resolution microscopy is being layered On trying to map circuits as they are affected by different types of behavioral experiences So all these elements are taking place together And I think people have that mindset of integrating right from the start because we also then have The ai and data science capabilities to manage the data So it all comes hand in hand at the level of individual proteins and changes in your activity across circuits and networks of circuits All right. Thank you. Um, we're going to move on to one of the other panel question. Wait. Sorry. Was there another? Oh, yes. Go ahead, please Um, what's one dream tool you wish you had right now to work on so for me I do injections and I hate that when I do injections. I damage the brain So I wish I could Inject in the cerebral spinal fluid and have it go just to the spot. I want So what's one dream tool that like is reachable in the next like Five years that you all wish you had right now That's a great question. Thank you. I think all the panelists might have a dream. Let's hear about your dreams Um, we will I think start on this side now Yeah, I think uh, well, you you touched on it really with, um vector delivery, you know, I think But we are at a point where there are viruses that, uh Reach the brain and they can be administered in the periphery so I think, uh A dream tool just experimentally and then potentially therapeutically would be to Have such a virus that could target a particular cell type and potentially a disease cell And to have a virus as well as I mentioned that could be delivered through the periphery with with those other targeting Capabilities, so I think we could be optimistic about that because of the large amount of effort that's taking place as well as the Progress the really revolutionary progress in that area That's occurred over the past five ten years in academic labs, but also a very large industry effort as well Yeah, for for me my dream. I feel like it's slightly modest because I'm I'm happy with a lot of the tools that we have But one thing I really wish we had was that they would just last longer So what I would really like to do is just really long-term versions of all these technologies that we have So we can track progression really well We can maybe go to newborns and then track all the way through through death That's what I wish Give some more specific examples within that you said our tools give some a specific example Sure, so um right now, you know my lab and a lot of other labs use two photon imaging to um Look at neurons within the brain But you can only really track the same neurons for a limited amount of time because every time you image those neurons You're actually damaging them a little bit because you're pumping lasers into them You know in the accumulated heat damage or you wipe out the fluorescent signal. So that's that's one change We also have um electrophysiology. So people are moving towards You know it used to be that all of your recordings were under anesthesia So if you wanted electrical signals from the brain, they were largely done under anesthesia Now we have things where we're recording from awake Animals that are behaving And now now actually the next step is you're recording from animals that are freely moving and doing all sorts of really interesting natural behaviors But your recordings don't last that long, you know, you're limited on the weeks to months time scale So what I would really like is a truly long term solution to that exact same technique So I have a dream tool. I wish I would have tomorrow um And I think it'll have a huge impact And we touched on that it is if I can deliver a virus in the vein And it gets to where I wanted to get in the brain And in a homogeneous level, I'll give you an example right now We do that in mice it gets everywhere in the brain In humans, we haven't gotten everywhere and even where it gets to a structure in some neurons It's really high in others really low and we haven't really hit the glial cells as good as we have hit the brain cells so Plus some diseases you don't have to hit every neuron. So to me having a library Of viruses with under different promoters that can hit different cell types It has to be a library literally a library of a 2000 if you will such viruses to have at many cell types Individuals possible and to hit those In a homogeneous level would be a wonderful tool and the reason I want the 2000 or 3000 or 50 Whatever it is at once Because i'm tired of the cost of gene therapy Of what's emerging. This is not sustainable. This is not going to be equitable This is going to be horrible for some people who will be treated and some will not it's going to break the system It's not sustainable But if we can somehow Do it where it's plug and play so you don't have to do FDA the whole package for every new vector if you've done them show the safety for all of them Once and for good, then it becomes really you have gene x you want to deliver it just to the Superior Guys my nucleus you just got it there and the person can sleep again, you know, whatever it is So i'm just thinking this would be a dream tool and it'll really accelerate therapies Thank you Um, all right. I'm going to come in with something completely different. So being a computationalist Um, I'm going to kind of harken back to our multi-scale question um, so we currently have um really good computational algorithms To model brain circuitry at different levels. So from from brain circuits all the way down to the molecular level To the sub molecular level An atomistic type of simulations, but they don't talk to each other So we're not able to to kind of connect these information this information these different Software and algorithmic approaches because we use we need to use different types of math We need to use different software tools to model the brain at all of these different levels And the different levels don't talk to each other. It's very difficult to get them to talk into each other So if I had a tool tomorrow um, it would be a computational tool That we would be able to simulate these different levels and these different layers Of the brain and then be able to probe One level and figure out what happens three levels up Just to complete a circle here. Okay. So I think um, uh, my dream is that probably is as Using engineering principle how to monitor the system and maybe biochemically but also electrophysiologically So so you can know first of all diagnosis second perhaps The monitoring the treatment evaluation and also select patient. So I think that particularly non-invasive kind of Sensor I think it will be crucial because it could be biochemical based could be functional based that really can push the whole Feel moving forward. Thank you All right, I do think we have we're we're getting lower on time I think we might have time for maybe one more audience question And I think we'll do a wrap-up like an overall wrap-up that sort of summarizes things. So is there anyone With a burning question right there. Thank you blue You have a mic coming so, uh, I think It's kind of like a follow-up question from his so I want to know if there are any particular markers on the brain cells as in Whenever there is a change like in red syndrome whenever there is an onset of disease Is are the cells changing are are there any receptors or any biomarkers which the cells express So that we can differentiate the normal cells and the deceased cells. So instead of having viral vectors I mean we can maybe have nanobots based made from dna or rna and Transport the molecules to the specific cells By using the receptors it'll avoid a lot of Vile screening and vectors screening, but I just want to know if that is possible at all So the questions is there a surface marker on the red neuron that's altered that you could somehow use that as a way to transport If I was to reduce the whole pathogenesis to about a certain Key number of genes that are very sensitive to either the loss of the gain of the protein There are about a hundred that no matter who does the experiment Which animal model which mutation it's almost consistent and of those hundred they there's some nuclear proteins Some of them are receptors Some of them are secreted factors. So it's a variety And I don't know that there's any one pathway you're going to fix red syndrome with You're either going to fix red syndrome at a circuit level Or you're going to fix it by providing the protein back at exactly the right level. I I cannot imagine any one manipulation Cutting it unfortunately just a Some more so when we are talking about the multi-scale modeling Did you think about an agent-based model where we have A model for intracellular Stuff and then there's a model for like intercellular stuff and then Maybe so if you could repeat the question once more, I think there's some background noise unless yeah Go ahead. So I I just want to know how How do you imagine a multi-scale model work? Like do you have do you imagine it having an intracellular stuff happening and then that intracellular stuff Of multi cells with Yeah, I love the way I love the way you're thinking so the question is you you mentioned agent-based models Of these different layers and that's one way to think about it But there are our mathematical limitations in the way those computations are done that they take a very long amount of time Yes, so that might not be the most efficient mathematical Representation so if we're looking at you know subcellular You may want to use one types of of math that you can simulate very quickly And then that would inform kind of the slower Computationally slower agent-based models at a different level or vice versa depending on the time scale and the question that you can ask So having that flexibility to have these different types of algorithmic Simulations talking to each other That's the that's the current challenge All right, so I'm now going to ask thank you. I'm now going to ask we're going to wrap up here relatively soon So I'm now going to ask each of the panelists. Um, we will not have you be in there I'm going to have you say the last word Dr. Zogby, um, but we'll go around Um Instead of a take-home message What is it that should be on everyone's mind as they walk away? From this about what the future holds in the context of what we're talking about so chris. I'll begin with you Sure. Yeah, I think uh We we can sort of recap some of the key concepts that were discussed here and so I usually think about Again, the idea of differences of scale and so understanding at the level of not only how circuits are perturbed But across how many networks in the brain and then actually we haven't spoken much about it But extending into the enteric nervous system thinking about the gut brain axis So that sort of breadth and then also digging deeply in terms of synaptic function Perhaps the localizations of particular protein. So we have the ability now to span all of those realms And and then I think a second point is to consider how the the truths of the relationships that we establish in those Areas, how do they vary in different contexts for different individuals with different genomic backgrounds? and different environmental exposures those are the challenges but with Remarkable advances in AI. We really have a way to integrate all this information and better understand individualized trajectories of disease Thank you Maria That was an excellent summary. Um, I'm not sure I have much to add that But I think Yeah, it's it's useful to keep in mind as you go away that you know, you can approach a single problem from so many dimensions And that many things might work Depending on the individual so leave it at that Um, I being the associate dean for graduate and professional programs. I'm going to speak to the to the trainees here um And if there's you know, some takeaways, um that I would that I would leave with you is that That there is work to be done at all scales There is work to be done With different types of approaches, whether those be computational approaches like we just talked about experimental approaches technology development And that you know that each of you have have something to add that you go deep into your studies and and understand How what you are doing can make an impact and Let that drive you Thank you really great. I'm going to speak probably from slightly different point of view is that As a this mainly for the for the towards the training or my past Is that when you have experience when you first started versus somebody's already been there for a while Each has pros and cons Experience experience could also equal to bias or limit your own thinking So in a stage that you have I think it's fantastic because like dr. Zogat So to be said that sometimes you find things serendipity But those opportunities only opens to the people who is prepared So just remember that you might stomp into anything that you but the more question that you have the more prepared That you will be so that's my thank you I think you hit on a very good point I was gonna say What I want you to leave this room thinking You are all going to be part of a very exciting revolution It's a revolution that's going to help us understand the brain at a scale. We never understood it And it's a revolution that for once it's going to help us to bring different disciplines Whether it is genetic biochemical metabolomic lipidomic Circuit behavior at a level where computation across multiple scales is going to help you without even Sitting on a bench or building something Understand biology of disease better. So those are the two things that are right there in front of you And you're going to be part of that What you just said and I want to amplify If I knew it's going to take me 16 years To find the red syndrome gene I wouldn't have even left clinical medicine not that I would have even done research I would have just taken the clinic be a pediatric neurologist being naive and daring and capitalizing Of the nidus of all the technology around you Is a very great asset I was very very naive I saw one gene mapped there was one gene mapped when I entered buddhats lab And that was the huntington disease gene mapped by polymorphism And I was like convinced if they could map the huntington gene I can map and clone the red syndrome gene My point is I was starting in the genetic revolution My career started with the genetic revolution as your career is Starting right now with the neuro biology revolution brain mapping revolution and computational revolution So keep that drive that naivety that daring take a risk. This is the time to do that What a wonderful message to end on please join me in thanking the panel