Webinar

<Katy>All right, you’re in. It’s really nice to see you here.

<Jan>So it’s a pleasure, Katy. Thanks for the invite. I was here for some of the earlier presentations, and it’s a stellar degree of advances that we’re observing across spatial biology and mapping, and I’d like to congratulate everybody in the world that is actually out there doing all those studies trying to make sense of the organism that we consist of. As we heard from Aviv, you know, we got like 37 trillion cells around a little bit less than 10 trillion of those are actually nucleated most of the others are actually red blood cells, but there’s a lot of different cells and there is a lot of sophistication in how they’re integrated and how they’re built. And I think you can all imagine that we need to know how the tissues are orchestrating it’s built because that’s where function comes from. The cells are doing their part, and then the tissues are doing theirs, and the organs will be able to perform. Of course, when the structure begins to fail we may have disease we may have emergence of things that we really don’t want, but you can also think about what you’re doing is not just actually an opportunity to study why we become sick, it’s also an opportunity for those of us that want to make the cells to actually build the right thing.

So we need some references, we need some understanding of what do you need to put together to make, even say, a functional replacement liver, how do you make a functional replacement kidney, and how do we restore, for instance, neural function the location the precise organization and the exact cell type is absolutely critical in regenerative medicine and cell-based therapies to succeed. You also realize that there’s something that somebody referred to now as FDA 2.0, meaning that you know we can do functional studies on human tissues but not primary tissues because it can be very difficult to get. It’s actually organ on chips, it is organoids, it is built materials that are coming from pluripotent stem cell-derived cells that are assembled into structures that then resemble what you are finding in the normal organism. Again, your work on reference building is fantastically important for the success of that because how can we use that in order to do new drug discovery unless it is actually a faithful replication of that biology that exists while we’re sitting here.

<Katy> And that’s how we connected, and that’s why we came down to Bloomington; that’s how we got to talk more, and it’s great that you’re in the Midwest, so it’s actually just a drive away. So, Jan, I will give you the floor, but it’s really cool to have companies come in here because ultimately, we researchers don’t necessarily want to do the 24/7 service, right? We want to do research, and teaching, and some of the service also. But I think we need that partnership between industry and academia and also policymaking ultimately to get this all done together. And so, it’s really wonderful to see you here, and Jan, please feel free to overview your company.

<Jan> Yeah, I’ll share my screen and give a little inspection here on sort of what it is that we’re doing at Trailhead Biosystems, which I’m very very proud of I was with Academia for most of my career, but I left with full force back in 2020 and now I’m with Trailhead Biosystems. I hope this will share now, let’s see. Yeah, and that share is fantastic.

<Katy> I also see your meeting reminder.

<Jan> Yeah, that’s off now. So, this is about the multi-scale human from anatomy all the way down to the cellular level and into bio markers. And what I’m talking about is sort of going the other way, making the cells from scratch. How can we manufacture human cells, and when we do that, what is the importance of being able to map the material that we’re making? And the company is focused on delivering high-quality induced pluripotent stem cell-derived human cells that can then be used as tools, say for drug discovery, organ on chip, function studies, and so but certainly also as material that you can actually use in cell-based therapy. And my background is, I’m from Denmark, I have been with Nova Nordisk the pharmaceutical company for my beginning career, and then I came to the US in 2001. I became a faculty at University of Colorado, and then I moved to the Cleveland Clinic where I set up my second lab, and that was to head up their diabetes basic research. And in that tenure, I then developed the technology that I am going to talk about today and that made me the founder of Trailhead Biosystems. And we are building a company that has a sort of another way to make cells than the other companies that you may notice out there if you want to buy some.

So, let me try to see if I can succeed at explaining a little bit about the difference. Again, I am with the company, so I have to make the disclaimer that I have a financial interest in Trailhead. I’m also on the board. So now you know.

The company is located in Cleveland because I was with Cleveland Clinic, and it seemed the right thing to stay here. We own a facility that we’ve equipped with laboratories, and bioinformatics, and some cell production, and what we need to operate.

The motto is “Biology. Controlled,” which is a very, very, very big statement, and I would like to see if I can at least succeed partially today to tell you why it is that we chose that. This is all about induced pluripotent stem cells, and honestly, induced pluripotent stem cells today is like a common fair. Everybody knows about it. We can generate an induced pluripotent culture from cells; we can get it from anybody, from an adult cell that’s reprogrammed into a state that is more or less identical, at least very, very close, to that of the Inner Cell mass in the embryo. When they’re in that state, they’re immortal because they have maintained the telomere length, and they can be cultured to very large amounts. This can now be done at a reasonable cost, and it has opened up the opportunities to actually consider “what can we do with these cells?” so the iPSC technology has simply given a promise. It says now you can make the cells that you need. And that promise is a little bit left unfulfilled at this moment, and that is where this presentation comes in. It really is the future Holy Grail of regenerative medicine, but only if you can re-specialize them – and that is not so easy as just reprogramming them back into that state, that is the embryonic one. To take them forward is actually a very very complicated set of events that usually happens only in one place, well and that is of course the embryo, that is where we came from, that is how we were built and we went through the progenitor stages, and there were multiple signaling pathways and many genes that play roles at that time that we actually in most situations don’t even need right now. So, as we consider the problem of this specialization, we’re entering into the development of biology, and that’s usually a descriptive field that talks about how cells are actually specialized while they’re in a three-dimensional structure here, the embryo, where they can communicate with each other.

So, when we get at it to the lab bench and we try to do this type of research, we use a common scientific principle. So, a common scientific principle is, you know, I can come up with a hypothesis: “I think that this is what drives this process,” and you make an experiment, and you test. That’s rather one-dimensional. Secondly you usually do it with your hand and you are sort of limited by both the amount of hypothesis you can bring to bear on that and then you’re up against the system itself, it is never just one thing – there’s more than one signal always operating, because that’s how we’re built. If you had a class in developmental biology, you would have been told in the first class that you’re built on regulative development, that means that the cells are communicating with each other. So, nothing is given beforehand of what they’re going to turn into; it’s something they figure out between them.

So right now, we have this situation that a lot of people can actually use iPSC-derived cells, but they have very few options and in some situations they don’t even have an access to get the cell that they need – for instance the cell in the brain that’s involved in some rare disease, etc. If you can find the cell you may not be very happy with what you get. The purity can be problematic, the function can be problematic, and you may not even get that much. And what you can get you have to pay a lot for. And most of this is because how these cells have been produced have more or less been in bio-safety cabinets with very low scale and on-demand production. So, of course, when we have a big problem than, most people tend to think that “you know we got to solve it,” and “we’re going to put a lot of effort to try to solve it,” and this is exactly what’s happening.

If you go to the reporter and just look at what the government here in the US is spending on stem cell research, pluripotent stem cell research, and directed differentiation – which is studies going into try to find a way to make a cell, it is absolutely staggering amounts that is being spent into this. So, it doesn’t look like we’re actually getting to sort of a closing end of trying to make cells. And maybe it’ll just take it a long time there’s almost like the Human Genome Project where we felt, you know, when we learn how to sequence with Sanger sequencing that as long as we just get at it for 50 years, we’re going to make it. And maybe this is the case, but I’m trying to make a counterargument today that I can demonstrate to you that it doesn’t have to be like that.

So, what we’re doing at Trailhead is something entirely different and we have been now spending actually a decade on this here in Cleveland, and we’ve come so far that I have something I think fantastic to tell you all today. When we do the science that we do we put the hypothesis aside. We make experimental designs that are so vast that we can test the combinations of factors with a very very high effectiveness. You cannot do that with a hypothesis driven based method. So, we replace a manual discovery method with robotics and computerized designs. That allows us to identify mathematically what are the actual critical process parameters in the behavior that controls the behavior of a system like a differentiating cell. And when we have these critical process parameters as mathematically defined coefficients, then we can use them in bioreactors, and we get better success at actually getting some control in getting the cells to specialize even when they’re outside the body. And that allows us also to lower the cost, because we can usually replace a very expensive molecule with something less expensive because we can compare them and see “do they do exactly the same thing?” And the way to look at it, this is empiricism, this is not AI driven drug discovery where you take somebody else’s data and just see if you can find something that they couldn’t. We must create the information that we lack. So, that’s why I call it powered empiricism. If you think about a theoretical physicist or an experimental physicist, there are those that think about it and then there are those that do it, and those intern, they’re the practical or the physicist. We tend to think of ourselves that we’re the practical biologist, we do the experiments because we need to do that otherwise we don’t have the data. And it simply runs on liquid handling robotics that otherwise would be used for high throughput screenings, but here conducting its experiments in a very different way because we’re only looking for combinations. And that allows us to actually appear into biology in a new way. To do that we had to develop some software tools because the data formats are different, and we have a new set of data that we can then peruse to see what is it that actually controls these cells.

That allows us to build new protocols completely from scratch because we don’t have to read the latest paper just to see “can we improve it a little bit”. We actually erase the whole thing and start from the beginning. At a foundational level, what we’re doing is that we can use the design of experiments, which is broadly used in Industries around the world, but at a much higher dimensionality. Because this is biology, it demands more. If you don’t do the high dimensions, you’re never going to find the interactions. So, we can compress designs at a high dimensionality into a meaningful level, and that is what we’ve been doing and sort of what I think perfecting over now several years. So, the technology is now de-risked, and we can apply it. When we do these experiments, we extract something called effector response matrices. That means that an X number of individual factors as perturbagens together and/or individually will actually control a series of outputs – of course, we measure gene expression. And we get a relationship with the behavior of a gene according to all these regulatory inputs. Now that allows us to, for instance, do something as you see here; this is a dimensionality reduction plot, not of genes as they’re co-expressed because that’s easy to do; that’s what you see in all your UMAP plots with single cellular and data, for instance. This is actually a reduction plot of the regulatory similarity of a gene.

In other words, it behaves in the same way if you give it a set number of inputs. One factor may inhibit it, and together with another factor, something else might happen, and if two genes behave in the same way to that combination, then they end up in the same space. In other words, this is a recipe determining plot, and I think we may be the only ones that can actually do that right now. And what you’re looking at here is a fate split in a medial ganglionic eminence progenitor of the forebrain that splits into two distinct interneurons one being the Pavlov expressing cell and the other one is the somatostatinergic interneuron, and they can’t coexist – the system doesn’t allow you to get a mixture of the cells either or. And that has been known in developmental biology for many years that this is how it works. You have lineage bifurcation, and we are now able to control the walk through these.

So, we built the protocols up and again you know I’m very very very fond of the teams that we have, we have several and got some fantastic group leaders and they’ve learned to do this and then they have now taken on this task of building not just one protocol but many in their respective teams. So, we can handle more than one protocol per scientist in this company, and some of the protocols are much faster because they’re much more direct than what you see in literature. And we usually find that we never have the same starting conditions to get out of the gate. When I talk about the islet cell induction here, for instance, this is a protocol that does not include activin in the first stage and I’ll ask everybody that listens to me that might be doing this to think about isn’t that what you do because everybody that I know in in the Sciences use a paradigmatic way of using activin for that induction.

We know it’s not working for this protocol. The same thing for the neural cells is common paradigmatic wisdom amongst those in the field that you use what’s called a dual SMAD induction. You inhibit the BMP and the GDF beta receptor Pathways, and then you get onto the neuroectoderm. We know, and have for a long time known, that that is not very ideal. You cannot specify the neural Fields very well when you do that. Yes, you will get neurectoderm, but you get trouble later because you need more specification. So, we build these protocols up, and we have to spend the time doing each stage, so it takes a bit of time to go from beginning to the end. But we must do it because otherwise we don’t really stand a chance to make the cell that’s fully specialized, and that’s the one that has the utility.

So, if we think about what you do on the mapping side, you can imagine how grateful we now are on all the work done by the single cell work, both in the human cell Atlas the human reference Atlas and people that are not there but are publishing it’s fantastic to be able to do that. And we’re very very interested in this data because it gives us a combination of genes that tell us that we’ve made that cell and, for instance, I want to bring out one paper here that of course is a breakthrough paper this was from Chengxiang Qiu it was published first in Nature Genetics and then a new follow-up paper in nature has come out this year that work I think is fantastically important. Yes, it’s done on mice, but this is a time-resolved sequencing of the early embryo, and there are some beautiful studies in those two papers that I will recommend to everybody that is within the area of developmental biology to take a deep look at.

They made this what is truly what’s looking like a metro map. I mean you start at the main Central Station in your city that you live in, and you jump on a train, and you stay on some stations until you get to the destination. And then they pick on some transcription factors because transcription factors are, of course, a class of factors that should control cell fate, and it sort of looks like there’s sort of train stations along the way. I mean, we all know that that’s a rather simplified version of looking at it, but it nonetheless gives it a very, very interesting and relatable expression and the sort of metaphorical perspective on what’s actually going on. So, as I both don’t like this figure, I love the figure too because it helps me to explain what’s actually going on. And what we did was that we saw that and thought, “Oh, what do we have right if these are the factors that were put on through this exhaust single cell space in the early embryo, what is the regulatory information we have access to just to give us an idea where the company is right now?”

So, I asked the team through our databases and actually say, “Well, which of these genes do we actually know something about?” so in figure A, every blue dot is a gene that we have measured and exhaustively understood through combinatorial gene control, what actually might control it. And you can see that that’s probably around half the amount of dots on that, of course we couldn’t expect that we would have touched every single Gene, but because the teams that we have are doing a very exhaustive job, they’re actually covering quite a good amount of the relevant fate controlling genes in essentially all lineages in the organism. And that’s because we have to measure the lineages that we may not even be interested in to make that cell for that we desire, and if we take a look at one of these genes, for instance, the HOXA9 Gene, just pick one. A HOXA9 is a factor that is a very very key factor, not just because it’s a HOX Gene in the parahox cluster, but it’s very very important for the hematopoietic stem cells, and we have focused very highly and intensely on the hematopoietic stem cells that we are now making at scale. And that belongs in our mesoderm team headed by Angelica Ueltschy, and she’s been over the time that she’s been studying this, done up 38 different high-dimensional modeling experiments, so she’s tested a very, very large number of perturbagens on HOXA9. Altogether, she has identified 443 significant coefficients that can control HOXA9. That’s just an enormous regulatory richness. And I had to redact the names of the inputs there, but we just picked three positive, three negatives, and then drawn triple interaction in purple and one dual interaction in blue, and the regulatory richness is actually higher in the blue and the purple regions because it is built on combinatorial signaling inputs.

So, if we look at the combinations of our various teams here, the mesoderm, ectoderm, and endoderm team, they have tested around 800 genes that were custom-picked and have then found that we have control models of the very, very vast majority of these. In other words, it gives us an ability to focus in on a gene and ask what is it that we need to do to get it or repress it. And I can’t really tell you how much total regulatory information that the company is in possession of, but it is very, very significant. So, this results in that when we get to the back end of these protocols, we actually now can make some cells. And I call these protocols second-generation protocols because they were built again. And they’re built in a method with a method that wasn’t hypothesis driven. So, there’s much less sort of investigator bias behind it.

We are, of course, of course, it needs to validate that the cells that we’re making are exactly what they need, so now we get into the slow mode. We have to do the same thing as everybody else we have to do, you know Immunocytochemistry is shown here for tyrosine hydroxylase, which is a key gene for the dopaminergic neuron. We can do molecular and sequencing, flow analysis, etc. So, it takes us some time.

But we have a large number of cell programs in the company right now, and it ranges from the ectoderm team that focuses on the brain and the glial cells. It is the blood in particular because there’s a rich number of cell linages in blood that after you’ve made the hematopoietic stem cells, you can jump into all of these. There is a strong need, I think, for replacement products for what is called apheresis-derived blood products because that’s how you usually get these cells today. They come from MACS bead-isolated cultures or purifications from donor blood.

So, this is what we had and what I’ve been using in my presentations here for a while now, and I liked to do this so it sort of conceptually explained what the heck it is that we’re doing. We’re able to systematically control the lineage selection. So, if you think about the human lineage tree, then that is usually always represented as a tree diagram. You know where you start with a fertilized egg, and then you bifurcate it again and again and again. And the most, the depth, the best lineage tree that we have for any organism is C. Elegans, so for those of you who want to see a complete and detailed linen diagram of an organism, go look at C. Elegans. But imagine that you have one also for yourself; it just hasn’t really been made yet, and it’s been very, very, very difficult to do this because that’s what we just heard Aviv say: It’s very difficult to get the right data because we’re talking about human embryos. And to time resolve the sequencing of the human embryo is not just something that’s easy to do. So, we are not actually equipped with a good lineage map of the human organism, but it does exist. We need to know it, so we must try to build it, which is what we’re doing.

What you’re seeing here is a conceptual where you put the starting point the pluripotent stem cell, the fertilized egg, and its first descendant divisions in the middle, and then time goes towards the periphery. The embryo grows, and the cells specialize, and these areas of red, green, and blue lineages are essentially just showing you that you got some choices. And that’s what happens in the embryo, and when you then end at the periphery, well, that is where the adult cells are resting, and then they’re in homeostasis in the adult organism. So, it is circularizing; the lineage tree is useful.

So, what we did was we actually went and we’re now building up a human lineage map, and we must make it in the modern day and age, and that is that it must be built as a knowledge map. Again, we circularize it, and we make the center stage the starting point of our protocols. That allows us now to have some opportunities to do things with this map, and I’m very happy to discuss this with Katy. So, Katy and I are ideating around how to expand the utility of such a map and embed and implement as much HRA and HDA data into this to make it with the highest utility. We needed to navigate our studies and our work, but I think it’s also extremely valuable to have a time-resolved relationship map of the cells, not just the relationship that you see in space. We must relate the cells to each other in time as well. And Aviv said that the fourth-generation atlas of the human cell Atlas is actually focusing on time, which I find it very very very important. But we can’t wait, so it has to be built, and we’re progressing with this. And it also allows us to look at what it is that we’re actually doing, because aren’t we generating multiple lineages, in other words how much have we covered, and what do we need to cover, and how much have we already done say aren’t we halfway through most of the lineages already. So, we’ve got very little work to do to complete the other ones, and that’s exactly the case. So, I’m particularly fascinated about the connectivity’s between the map and then the ability to actually sail along the lineages with the control inputs.

So, we can, for instance, very easily articulate, you know, here is what we’ve made when we’ve made a hepatocyte. You know it started blah blah blah it moves out in the endoderm and then it specializes, and it turns into a parasite, and you can run these trace elements in it, and this allows us then to show our progress.

Now, I want to shift a little gear and then talk about the actual results. So, of course, we hear this is a company. We must become a successful organization and therefore we are of course interested in making material available to you. And the way we do that is that we want to offer these cells. We want to make these cells. We want to sell these cells, we want to see them successful, and we want to see them successful both in therapy we want to see them successful in assays. But we need help. We need customers, and we need partners. We need to make sure that we can get the necessary criticism that the cells are, you know, good or if they need something – we need the feedback.

So, there are many ways where I want to have you guys think about whether or not something that we can do can help you in your cartography efforts, in your assembly efforts, and functional comparative efforts. And so, when we look at the dopaminergic neurons here I don’t want to go into details with that but we’re very excited that we actually we’re able to create conditions so that we can control the sub linage of the dopaminergic neurons, because it happens to be five different dopaminergic neurons of the midbrain and there’s only one of them that’s actually lost in the Parkinson’s brain. And we don’t find evidence that the existing protocols, even those that are clinically translated right now, has succeeded in making the only cell type that’s lost. So, in other words, what is now being transplanted into the brains of PD patients is actually a mixture of many different dopaminergic neurons, many of which are not actually lost or should have any functional replacement value.

So, if we go and compare it to another product that we could acquire in the market because there are some that offer dopaminergic neurons, we get it in, and we see, well, how does that look. And here is an example of an ELISA data for dopamine, which is sort of the litmus test of the cells. And we compare it here to another product, and it turned out that we really couldn’t even see the dopamine in the competitor’s product, but if you really blow up the axis, it’s there. But, there is a vast difference in the total amount of dopamine produced between these two cultures. If we, for instance, look and shift to the blood, here is a picture of the bioreactor harvested hematopoietic progenitor cells that we’re now very close, or expect to release as a product to market during 2025, and this is a rather quick differentiation process that go through what’s called the “hemogenic state of the endothelium” that then creates these emerging hematopoietic progenitor cells. We’re very interested in showing that we can actually get them to graft into the bone marrow because that opens up therapy windows for these cells, but we can also use them as a material to test and produce various descendant blood lineages that I’m not talking about today.

And, we can again compare to a product that is be purchased on the market induced hematopoietic progenitor cells and here we’re actually doing, I’m just collecting a lot of data in one slide – you have a picture between the two viability is fine for both products but for instance when we test induction of monocytes we have a much stronger induction of cd14 and 16 capabilities same thing for the neutrophil lineage is an increased purity and viability, a much better expression of cd11b, and less early cell death because that’s usually what happens when the neutrophils differentiate. For the red blood cells, this is a big problem, usually because you get this annoying thing called primitive erythropoiesis, which is the first thing that a hematopoietic stem cell can do in the embryo, and that’s really not a good defining characteristic for the definitive hematopoietic stem cells.

So, you want to reduce the primitive erythropoiesis, which is also what we find. And then we have an increased lymphoid output as well of this, and many other studies are ongoing to clarify, you know, exactly what these cells can be used for. So, we can descend into specific product lineages from them.

If I shift gear again, here is the hepatocyte program that was developed very quickly by us. So, it took less than a year to get to where we are right now. Again, it’s a novel differentiation process, it’s very low cost. The cells are capable of expressing the cytochrome p450 detox enzymes, and they also secrete albumin. They reduce their expression of the alpha fetal protein marker, which is actually quite important because, that cells, if you have that, the cells are actually not mature. And we’re now moving forward to doing large scale of these.

Looking at these is a quite interesting cell type actually, because they can do so much on their own. And it almost looks like it’s a histological section that I’m showing you there, but that is actually almost a sinusoidal function with the canaliculi that you see, and the cells will produce up take lipid and channel it into these canaliculi structures. Bear in mind that there is no sinusoidal endothelial cells in this, and they need to be added if you really want to build up a hepatic sinusoid, say for hepatic transplant studies. But they are very well progressing to both polyploidy back-end maturation, and that’s what we wanted to replace primary liver cells.

Islet cells are also interesting. It’s my legacy cell; it’s what I started the whole thing with. I became an investigator in pancreatic developmental biology in the ’90s and studied genes like neurogenin-3 and PDX1 before they were even named appropriately. Now, today, we’re able to make bioreactor-produced islet cells also in a very rapid and low-cost method. And this protocol is channeling through what’s called the dorsal pancreatic stage, indifference to the published protocols which are actually channeling through the ventral pancreas. And both protocols can make pancreatic islet cells, endocrine cells emerge, and we of course comparing this and we’re very interested in seeing this go into patients.

We have many cells, and I do not have time to talk about them. We have never failed at making a cell that we actually wanted to make, and we’re getting faster at making those. So, if somebody is out there in the world, listen to me now. If you need a cell, I really would like to talk to you because I think we can succeed at building it for you – also within a reasonably short time and reasonable cost. And many of the cells that we have, we chose because there is a disease waiting for them.

The Parkinson’s disease is important because that’s where you have the loss of these A9 dopaminergic neurons. Huntington’s is equally important because that’s where you have a specific loss of the DRD2-expressing medium spiny neurons. There were some very recent data published that showed that you could successfully treat epilepsy with implantation of interneurons into the hippocampal region, and that was a dramatic reduction of seizure activity in human subjects. So, interneurons are also very interesting from a cell replacement therapy standpoint.

Oligodendrocytes are interesting. They are important for the remyelination that would perhaps been, for instance be, a therapy opportunity for multiple sclerosis. We found conditions to build vascular leptomeningeal cells, which is essentially the parasite of the of the glial lineages. It looks very much like parasites, but they’re not. They’re coming from the from the neural side and they participate with the astrocytes and the blood-brain barrier. We expect to release these within three months, so if you guys are working on the blood-brain barrier, please reach out to us because we would like to see that these cells are studied further.

We’ve made endothelial cells, we make hemogenic endothelial cells, and we made hematopoietic stem cells, and progenitor cells, and from there we’re developing a lot of lineages into for instance macrophages, monocytes, dendritic cells, t- cells, Etc. And erythrocytes, here is a spin down of a bioreactor produced set of erythrocytes with that protocol. And then, of course, we have the liver that I just showed and the insulin producing cells.

So, our method to market is to make large amounts of cells that we can sell from, so you can benefit from batches. And the way it works is we do all the work in the development phase in adherent cultures because it’s where we can cut the cost into a meaningful level. We identify the critical parameters at this point and then we take it through a process of design transfer, or what was otherwise called MSAT – for manufacturing science and technology, and then we see can we actually take these cells forward and succeed at some level in making a well differentiated amount of cells at a much higher yield.

So, this takes time. Of course, this is production, this is towards manufacturing, and that’s what the company currently invests in so that we can actually deliver the batches, and the sizes, and the function that you need.

And so, the product begins to look like, what you otherwise would get is a cryovial, user instructions, and then thoren seed, and test and complain or send your results back with a thumbs up.

We can also do something else besides selling the cells directly to you. We are actually having some engagement where we can receive cells from others, iPCSs say made in a patient-specific manner, and then we would essentially contract manufacturer specialized derivative, and send it back to you. It’s your cell, we convert it, and you then do whatever you need to do with it after it was differentiated, and we simply become almost like a contract manufacturing organization in that.

So, for those protocols we have developed, that is an opportunity to engage, and I do think that there is a strong need for a service like that, because I think a lot of companies, and academics have actually generated a large amount of patient derived iPSCs that is now finding its way into the various banks with the hope that we can actually generate a disease-specific model. But then it becomes difficult to actually differentiate them, and then frustration hits, so in that situation, this is where I want you to think about what we do and whether we can actually arrange something.

So, the way to look at Trailhead, when you see the logo in the middle, it is actually the process. We can jump across lineages; those bases that you see in green are the first one represented the iPSC state, and the last one is a specialized cell, and our method simply jumps from base-to-base. That’s why we picked that logo; it’s also why we picked the name Trailhead Biosystems. It’s not Trail Therapeutics, because we’re not the therapy developer per se, we’re not. We would like to partner for that, but we are a biotech company, and we focus in on that technology where we are the protocol developer, and then we take what we know about that to build us into a manufacturing company. And the way we can engage with the outside world is to find out what sources of cells exist, and take it in create disease-specific descendant products, or normal ones maybe for therapy – hyperimmune strategy would be the best place to go because then we can create an allergenic, more grafting acceptable output, and then the use of the cells is as vast as your imagination. It can go into disease modeling, it could go into high throughput drug discovery, or toxicity testing. Or it could go in and you know build, for instance, the future cell-based therapy and cure type one diabetes, cure Parkinson’s disease, cure Huntington’s disease, do bone marrow failure cell therapy replacements with a clinical therapy that is actually safe – because currently bone marrow transplants are not the safest clinical practice.

So, as I’m sort of at the end, the take-home message here is that what I told you about HD-DoE is a technology that we use is mathematically driven; it saves you a lot of time and cost. The multi-var data analysis that is embedded in the method converts all these data into a mathematical set of modeling, where biology is now, it can be, queried in a different way than before. And that is mainly done to capture interactions, not how one factor can control one gene. When you gear it up sufficiently, then biology begins to yield.

I would love to connect with anyone, any of you, so just send a connect in the follow dropdown button on my LinkedIn page. And I will also alert you, too, that we’re actually building up our website, and you’ll find material on that that may be interesting to some of you, where we begin to talk about the cells that we make and sort of contextualize the work that we do.

That’s what I have today, is to Katy it’s just so fantastic to be here and talk to the general community and I hope it’s as it was such a pleasure to be there and visit the center and actually begin to realize how much of a resonance that we have as we get into this fantastical, amazing thing that human biology actually is.

<Katy>Thank you Jan, I think there are lots of things to discuss here and we have a little bit of time. I was super excited to see your interest in time-resolved cell lineages. There are number of teams within sea working on this so there is Nozomu Yachie who is on Osaka University, actually got funding to develop his DNA event record quarter cell from a Japanese funding agency, and then he can basically track every cell division and at the end he does this for mice – can’t do that for human, at the end you can sacrifice the mouse, and then every single cell would tell you the entire sequence up to the beginning point and you wouldn’t have the time resolution there, right. You just have the end product, but still, it’s amazing that you can do this now.

<Katy>And then Gary Beter who just visited IU, also gave a data science presentation, he his talk here at IU will be rebroadcast at 5:00 a.m. and so he presented on work to basically simulate time resolved cell linages for human, but that’s a simulation, this is not experimental data. But I think we can come from these two different angles and try to see what we can do this way. Ultimately, I think we are many of us are interested in having that grand map of cell lineages from the very beginning of an adult human being, and then also understanding how often certain cells take over right – and blood cells, cells in your intestine Chenchen just presented, they turn over quite often, others they stay with you or they die and they never get recreated. So, I think it’s really important to map what we know so far, but also to map the known unknown. So, that was the second takeaway here.

<Jan> I totally agree, is there questions in the in the audience? I have not followed much here, so I don’t know.

<Katy> Yeah, so Stephen has joined. He’s going to be moderating the next panel. Mariel is our tech support here and the audience members, I think they would have to raise their hand to speak. I do see quite a few experimental lists in the attendees list and I think they are the ones that could actually test some of your cell, right? And there are more and more teams that we connect to that do organ printing, often times just part of organs, not an entire organ. ARPA-H just committed 200 million, I think well big pot of money, to 3D printing of organ. So, I think the time is ready for this and there will probably be more and more printed organs and then using human reference atlas and using the human cell atlas data to confirm that this is the organ, you miss the right cell type.

<Jan> It’s super important, the referencing and the standardization that it has to go up against the normal human biology. It’s very difficult to, we, if you don’t have that reference and you begin to print together, you get an imbalancing. The functional level, the function at the cellular level may not be where you need to be, and your clinical outcomes will suffer. So, it’s from a reference standpoint, not only as I’m saying you know, we needed to know the exact individual cell before we begin to mix them together, but then after you begin to mix them together then the anatomical referencing and the cellular referencing becomes, it’s the only yard stick that you really have, it should be the normal human. So, it’s certainly fully interconnected, and to me it’s remarkable how, you know, over time you know technology is all of a sudden coalesce in time, right? We have the, back in the days with the Human Genome Project sang sequencing as manual emerged, and we were walking along the chromosomes, but it was painfully slow, and yet all of us sudden at that right time the DNA sequencing technologies emerged as that took us to where we are today. And then, when I look at 2024 – 25, we have the single cell that moved through its initial starting days and is now turning into a spatial technology, which is absolutely remarkable and I don’t think that, you know, the first single cell data that were done on say nuclei or just individual trypsinized cells, would have imagined that it would be possible to actually pin it into a spatial domain. I didn’t think it was possible, certainly not within that time frame that it has happened. So, it has been almost a vertical advancement on that technology development, and then what I’m seeing with our DoE technology is also a convergence on technology availability at a time that’s ripe. And then, when you combine that with the current AI and generative model training that is emerging, and has come, the type of data we’re going to generate will be even interpretable at ways that hasn’t been before. And that is that sort of convergence right now, I think that we have that we need to capture to really go and get some breakthroughs done.

<Katy> So, I see a number of students also here in the list of attendees and I wonder what you would tell them what they need to get hired.

<Jan> They should, we would love to have the best talent at Trailhead, so send – always! I mean, if you really think that this is you, do not hesitate to call me directly or contact me on LinkedIn and so forth, like just, if what I was talking about is exciting to you, you tell me what excites you – that’s the most important starting point. And I would also say that I’ve met a lot of students that, right now, are really worried whether or not academic careers are the right thing because they see a lot of suffering from, you know, faculty. It’s very difficult to maintain productivity when hyper competition is where it’s at. And it shouldn’t be that we train ourselves to become scientists and then most of the time is actually out asking for funding through grants. I mean we should do the work that we were trained to do, we should do the science. And I don’t think industry necessarily is the one and only opportunity here, but it is, I think the total research budget in the US is only 10 to 15% that’s actually run through the government funded research, but that’s what people see. 85 to 90 is run within companies. So large pharma companies have enormously active research programs and there are many other places that research is ongoing where there is a need. But I would rather than be depressed about, you know, where you are as a young scientist, if that’s what you are, which I hope you’re not, that you take a look at this to say it’s one of the most exciting years that biology has ever faced. Because this is the time when we finally get through, and that you move from, you know, you think biology is a big black box for which you’re going to learn this much over a lifetime to where you can now go in and actually begin to grasp it, but you will need the machines to help you – because a single human brain cannot do that. So, to me, that’s a fascinating world that we are at right now, which I find to be very grateful to be present in.

<Katy> And I think we will have to bring in AI and large language, and all of it and, it has to be the text that we can get out of papers, and out of experimental metadata, but also the 3D spatiality – because all of that unfolds in 3D space, be it the human genome that we saw in the first hour of this event, to protein folding, to cells that work together in assemblies like what Michael LaVine showed us, to ultimately human organs which are touching each other and impacting each other, and ultimately human beings – which are also distributed in a 3D space.

<Jan> and you mention the machines to just, as we close, I can see I’ve only got a minute left or so, that to me one of the things that that I’ve sort of developed this thing that I call the anthropocentric concept. And the anthropocentric concept gets at this, that you know we tend to think that we’re like the Apex species. We’ve got it made that we’ve got to have the best and the brightest to do a certain thing, but the end of the day we’re humans – and that’s a good thing, but we’re humans. So, our brain is meant to work well for us in the environment we are and when dimensionalities creep above the three that we’re living in, being human is actually a limitation. So, anthropocentrism is essentially where, if you really think that you’re always going to be the best, don’t – get the help you need and get to the machines and then be liberated that way. And you’ll get much further, I think, so don’t put yourself in as the as a bottleneck in your process on the problem that you’re addressing.

<Katy> Thank you Jan.

<Jan> Thank you for the event here on a Saturday night in Cleveland and Katy, the fact that you can go through all of this and moderate all the sessions is a testament to both your resilience and your excitement of all things in science. So, again thank you so much, I’m going to drop off and look forward to catching up.