Preview Mode Links will not work in preview mode

Getting Personal: Omics of the Heart

Feb 21, 2018

Transcript of the February Podcast,

“Getting Personal: Omics of the Heart”, Episode 13


Hosted by Jane Ferguson


Assistant Professor at Vanderbilt University Medical Center & Associate Editor of the Circulation: Precision and Genomic Medicine journal of the American Heart Association

Jane Ferguson:             Hello. This is episode 13 of Getting Personal: Omics of the Heart. It's February 2018. I'm Jane Ferguson, an assistant professor at Vanderbilt University Medical Center, an associate editor at Circulation: Precision and Genomic Medicine, and an occasional podcast host. This month, I talked to Kiran Musunuru and Svati Shah about how they spearheaded name changes for Circulation Cardiovascular Genetics and for the AHA Council on Functional Genomics and Translational Biology, and Andrew Landstrom talked to Kaytlyn Gerbin and Brock Roberts from the Allen Institute about some extremely cool work they are doing with CRISPR and IPS cells to create fluorescently tagged maps of live cells, which allowed them to image and better understand the structure and function of individual cells.

                                    I'm delighted to have two guests on the podcast today, Dr. Svati Shah is the current chair of the AHA Council on Genomic and Precision Medicine formerly called the Council on Functional Genomics and Translational Biology. She is an associate professor of medicine at Duke University Medical Center. Dr. Kiran Musunuru is editor in chief of Circulation Genomic and Precision Medicine formerly named Circulation Cardiovascular Genetics, and he's an associate professor of medicine at the University of Pennsylvania Perleman School of Medicine.

                                    Dr. Shah and Dr. Musunuru were kind enough to take time out of their busy schedules to join me for a joint discussion on the recent enhancement of name changes for our council and our journal. With tight schedules and last-minute flight cancellations we didn't have ideal settings for recording, so apologies in advance for a little more background than usual.

                                    My instruction highlighted a number of name changes and astute listeners will have noticed that the new names for both the journal and the council are very nicely aligned, so I know this was not a coincidence, and I'd love to hear from both of you, what prompted the decision to change the respective names of the council and the journal, and how did you come together to streamline these name changes?

Svati Shah:                   Well, maybe I'll take a first start, you know, we, we're really lucky in our council, we have a very, you know, certainly one of the smaller councils [inaudible 00:02:26] we have a very collegial spirit that wants to get things done. So, these conversations actually started probably three years ago, umm, when Jen Howell was chair of the FGTB council. And we realized that not only was our constituency broadening in expertise and breadth and depth, but also, umm, the desire to kind of move beyond the really wonderful work the council is doing around technology platform, genomics, genetics and you know important advances in many of our council members have made in the translational biology field and really thinking about the fact that we have this amazing expertise that can come together across a wealth of disciplines to really translate what's being done in the omic space, and apply it in this new world of precision medicine.

                                    And so, umm, that is what stirred really thinking about a name change so that not only would it reflect this expanding constituency in expertise and hopefully draw even more people, across the, you know, wide expertise. But also to harmonize more with people who are in other councils, including clinical cardiology, and people that, really, in the end we are actually quite allied with scientifically, but perhaps those councils didn't recognize really what our council was about because of our previous name. So in that context, you know, it's been wonderful. Kiran has been a wonderful partner in all of this, he's been a real leader in the council and over the past two years we have had many conversations across council leadership and the entire council, and thinking about what this name change would be. And actually, it was almost a consensus amongst council leadership to choose Genomic and Precision Medicine as the name, really to reflect our core beliefs and our core science in genetics and genomics, but also to reflect the expanding expertise of all the different omics platforms, our expertise in clinical genetics with more genetic counselors joining our council, and our expanding expertise in computational biology. And this really allied nicely also with the American Heart Association building a very important institute, the Precision Medicine Cardiovascular Institute. So, I'll let Kiran go from here but again, Kiran has really been a great partner in this and he can kind of expand on that story and how that led to the journal name change.

Kiran Musunuru:         Sure, so, with respect to the journal, I think these changes have been growing for a while. I think a lot of the same considerations came into play, the feeling that the journal with the name Circulation Cardiovascular Genetics was perhaps too narrowly defined given how the field, how the science was evolving. And the other consideration is that the Functional Genomics and Translational Biology Council has had a journal, a companion journal if you will, all of this time with a fairly distinct name, Circulation Cardiovascular Genetics, and so it wasn't necessarily obvious to those who are not on the inside so to speak that there was supposed to be a very tight connection between council and journal, that the journal really was the journal of the council and so in the process about deliberating about a council name change, it became natural to think that, "Wow, wouldn't it be nice if the journal could execute a similar name change", and separately, even though this predates my tenure as editor of the journal there had been conversations going on separately or independently that perhaps the journal would benefit from signaling that it was not just about cardiovascular genetics in the very narrow sense, but was really about a much larger area of science. And so there had already been contemplation for quite a while about a name change and so when I assumed the editorship I didn't really have to do much to convince anyone that this would be a useful thing.

                                    The scientific publishing committee of the American Heart Association and all the various people involved publishing the journal were already sort of primed for a name change and then it just ended up being a nice convergence of opportunities, Svati with her work in the council and really showing the leadership to lead the transition from Functional Genomics and Transitional Biology to Genomic and Precision Medicine.

                                    That really laid the groundwork, and because it was such a deliberative process, such an inclusive process, involving dozens of people on the leadership committee of the council as well as general membership of the council, it was really a no-brainer. The hard part had already been done, the thinking had already been done and I was straightforward to say that we should change the name of the journal to match, Circulation Genomic and Precision Medicine.

Jane Ferguson:             Have there been any logistical difficulties in getting this name change through, or has it all happened very organically?

Svati Shah:                   The American Heart Association has been a real partner in the name change, sometimes things require many layers of approval and in fact, it has been a relatively seamless process. We came up with a consensus around the name change and later applied formally for that change in the council name, and that was pretty quickly approved by the Scientific Advisory Committee, within a few months really. Our name change became official and we are in the exciting time now of advertising and kind of marketing the name change and appeal to a broader constituency and really reach out to group that perhaps wouldn't have realized that this council is a great home for them again thinking of genetic counselors and computational biologists. So, it really, you know, has been a surprisingly seamless and fun process.

Kiran Musunuru:         As I mentioned before it was already kind of in the air that a change was imminent and so when I posed the name change to the Circulation Genomic and Precision Medicine it ended up being a very smooth transition. It was timed so that the volume change, that is changing from the volume associated with the calendar year 2017 to the volume associated with the calendar year 2018, January first ended up being a very logical transition time and so that's when the change occurred.

                                    And happily, the council name change ended up occurring almost in lockstep; whereas, you know within a few weeks of the journal announcing its name change the council was able to announce its name change as well. I think that has had a reinforcing effect across the American Heart Association and its membership. It really signals that the council and the journal are tightly tied together, are partners in moving in lockstep.

Jane Ferguson:             Svati, this question's probably more for you, so what does the name change mean specifically for existing FGTB council members, and what if anything will change, and then what might it mean for potential new members who are trying to decide what council to join?

Svati Shah:                   That's a great question, Jane, you know I am a pragmatic person and I think our council also reflects that pragmatism. We get a lot of things done and we, I think, spiritually all agree that we shouldn't just change the name just for the sake of changing the name. And so we really, actually the name change followed [inaudible 00:10:18] were involved in, these discussions are a year and a half of really thinking about what direction we wanted the council to go and then what the sort of short and long term goals of our council are and then how does the name change effect the long term goals.

                                    So, we have a lot of great initiatives in the short and long term, which again will capitalize on our broadening expertise in these different clinical genetics and precision medicine and really, translating genomic and omic findings into, into important patient care. And so, we have several things coming down the pipe that are sort of proof of principle examples of what the name change reflects. 

                                    So, one example is that we are now working on developing a certificate in medical genomics with the idea that we really need more genetics education. Our council has been very much embedded in genetics and genomics education, Kiran being a key example of that. And now we are expanding that into thinking about how genomics is applied to clinical medicine but making it at the level that is digestible and understandable and is easily applied by a general cardiologist and even primary care doctors will be able to use that resource. And the idea is this will be your self-sustaining certificate that's given through the American Heart Association, so we have a group that's been working on that certificate and hopefully that will be coming out soon.

                                    Another key component of what we're doing is trying to reach out more and partnering with other associations including the American College of Medical Genetics and the National Society of Genetic Counselors, again really thinking about how we transition our important scientific discovery work into translation implementation science around patient care.

                                    To give you some examples of what that means in terms of what the name change is reflecting, I think with the right use, for the second part of your question, which I think is a really important part of your question is, we want to attract more people in the computational biology field, in the precision medicine space, in the clinical genetic space and again reaching out to genetic counselors through some of these societies, because we, just the wave of precision medicine is here, is going to expand even more and the expertise within our council that was already there but that now we can expand. I think it will be leveraged to really make important contributions to making sure that those efforts in precision medicine are done well, or done responsibly and are done with the patient in mind because in the end the American Heart Association is at the forefront of patient advocacy group.

                                    This is a really exciting time, I think that, you know, however you want to define precision medicine the bottom line is precision medicine is here and we can't have, it's not going to be a single faction of individuals or a single expertise that is really, is going to be able to leverage fundamental scientific discoveries whether its genetics, genomics, metabolomics, proteomics, and really translate them responsibly into patient care, so it's going to involve an interdisciplinary and multi-disciplinary effort.

                                    I feel really proud that I'm part of the AHA and that we sort of have this perfect storm between Kiran's leadership in the journal, our council changing, you know, its goals and its name aligning with the Institute for Precision Cardiovascular Medicine within the AHA. And I think that, you know, it's not all rainbows and sunshine. We have a lot of work that is cut out for us in the next few years to figure out ways that we can tangibly and concretely, and again responsibly, work together across each of these three components of this perfect storm to make sure that it’s not just a glitzy name change and that there is actually substance and behind all of it, so, you know, it will be, there will be challenges, there will be obstacles, but I think that the amazing people within each of those three components, I feel very confident that we are going to be able to do it well.

Jane Ferguson:             Yeah, I agree, as a member of the council, if anybody can do it I think this group of people can do it, so it's very exciting to see, so thank you both for joining, and congratulations again on the new names. It's really exciting to see these, you know, new directions for the council and the journal working together. And I really look forward to seeing all the great initiative that will be coming out in the next few years.

Svati Shah:                   Thank you, Jane.

Kiran Musunuru:         Thank you, Jane.

Andrew Landstrom:     My name is Andrew Landstrom, and I'm an assistant professor in the department of pediatrics section of cardiology at Baylor College of Medicine. I'm a member of the early career committee of the American Heart Association Council on Genomic and Precision Medicine, previously the Council on Functional Genomics and Translational Biology, and I'm joined today by Brock Roberts and Kaytlyn Gerbin, who are scientists on the stem cell and gene editing team at the Allen Institute.  Here to discuss a little bit more about CRISPR editing and what they have done for live cell imaging using fluorescent proteins.

                                    So, Brock and Kaytlyn, I'm hoping you can discuss a little bit about what the Allen Institute is and your overall research mission and goals. 

Kaytlyn Gerbin:           Yeah, great, so this is Kaytlyn and thanks Andrew for having us on, and we're really excited to share a little bit of the information about what the institute is doing, because we're building a bunch of tools that we think would be really useful for the research community. So, we're excited to get the word out there. And so, the Allen Institute is a non-profit research institute, and we're based in Seattle, Washington, and, essentially what we're trying to do is better understand the cell.

                                    We want to understand the various states the cell can take based on structural organization of how different organelles work together. And so, we're doing this, essentially by live cell imaging and also combining that with predictive modeling so that we can build tools to be able to understand structure-function relationships and how cells behave in a healthy state or in a diseased state. So, you can kind of think of this as, we like to say sometimes like a Google Earth for the cell, so if you kind of think about it in that context, a lot of times, you know you could look at the cell at a really high level just like you could look at the Earth at a very high level. Then you could zoom in further and you could look at an individual pathway maybe that you're interested, or perhaps, as an analogy, like a different highway within a part of a city.

                                    But you don't really understand how all that works together and how the city functions together until you start to put in things with spatial organization, or maybe temporal dynamics, or how different parts of the structures, or different structures and organelles work together to form the unit that is the cell.

                                    And so, essentially, we're trying to generate a bunch of data so that we can build predictive models to help us understand that better. And, we're doing this with human induced pluripotent stem cells, and the first cell state or cell type that we're studying is cardiomyocytes after differentiation.

                                    And so, yeah, as we're kind of generating this data we are a non-profit institute, and all of our lines and our plasmids, protocols, data, pretty much everything that we make is becoming publicly available as it passes QC. And so, yeah, we're excited about that, I don't know if, Brock, you have anything else to add.

Brock Roberts:             Yeah, just I think an important concept that we're often working with is scale. And, biology exists at certain scales, and that's certainly true for cells and the Google Earth analogy holds.

                                    You know, at some level if we want to understand the cell at the scale of its entirety, but we have to kind of cut that down and understand cells at the level of its parts.  And, they're working together as we know, and can infer, but we try to find a way to look at the part one by one and then put it all together in a model that's predictive. And the predictive part is going to be really important. Much like Google Earth can allow us to, you know, look at a traffic pattern in the city or something like that once the data is filled in. We hope to fill in enough data by looking at the cells constitutive parts to make the predictive model.

Andrew Landstrom:     And not only looking at, sort of, constitutive parts, you're doing this in a physiologic live cell, so really it's Google Earth, but it's Google Earth in real time as cars are driving down the freeway and people are walking down the street.

Brock Roberts:             Right-

Kaytlyn Gerbin:           Yeah, exactly.

Brock Roberts:             Yeah, that, that's where the dynamics of the cell can really come to life if you've prioritized looking at live cells, which are obviously incredibly dynamic.

Andrew Landstrom:     And so, you know to be able to accomplish this, you all have come up with some pretty novel methods. Would you talk a little bit more about your CISPR editing approach, and how you've applied this to different lines and to get, sort of, different markers into cells?

Brock Roberts:             Right, sure, the, we should say that we owe a lot to the development of CRISPR-Cas9 editing, which preceded us by a few years, but we've tried to kind of scale it up in some important ways.  And, really the important thing to appreciate about this process is it's a way to make a very precise, precisely guided DNA break in the genome of a cell. And we do this in human induced pluripotent stem cells, and so we can quite precisely choose a position or location in the human genome and trigger DNA damage, trigger breaks in the DNA molecules that make up the chromosomes. 

                                    And we can do this with, kind of a highly specifically guided RNA molecule that we complex with this Cas9 nucleus molecule, and these are, very famous molecules now, over the last few years they've become very well known.

                                    And the upshot of this is we can, sort of trick the cell into repairing that DNA break using the processes that are always at play in living cells to resolve breaks in DNA, but we can sort of trick that process to add something additional at a specific site. And the additional sequence that we use is a tag sequence that corresponds to a fluorescent protein after the DNA is expressed and translated. And so, what we can effectively do is tag proteins that are produced in a highly endogenous, natural fashion within cells.

                                    And the proteins that we can tag in this way, using this method, correspond to some of the most canonically recognized structures and organelles within the cell. And so, at this level we try to choose proteins, tag them in this manner, and take advantage of the fact that they will localize predictably to some of the dozens or hundreds of structures that make up cells.

Kaytlyn Gerbin:           Yeah, and a key thing I think to add that Brock kind of mentioned was that this isn't any over expression, we're doing all this endogenously. So it’s really like, pretty, I think that's a big advancement over what has typically been done in the past with a lot of fluorescent tagging of proteins within the cell.

Brock Roberts:             Right, but what's important to appreciate is that we're using the cells endogenous copies of each protein, expressed from the genome. We've done it in about 30 different genes so far. And we have a high success rate in accomplishing this process, all the way through to completion, which is to say that we know that we can introduce a tag onto at least one copy of each gene that is, that encodes a protein that can be tagged this way, and then we can monitor the cells over several months and ensure that this doesn't have a negative consequence on their growth or on their ability to differentiate or something like that. Our quality control process. We have a high success rate so far.

Andrew Landstrom:     And that's really, in my eyes, one of the key, key sort of, innovative factors of your work, in that these are endogenous proteins that are able to be expressed and then to be imaged in real time without really disrupting the underlying cellular physiology.

Kaytlyn Gerbin:           Yeah, and we do care a lot about what you just said, that it doesn't have any negative effect on cell behavior because we are using these as a surrogate for understanding cell behavior in, hopefully, a normal context. And we do an extensive amount of quality control work and all of that QC data is available on our website, and then you can actually access all of our cells through Coriell and all of the QC data for all those cell lines is made available, and we've also done a pretty extensive job outlining the QC that goes into this process so that, hopefully, people will take a look at that when they look at our cells and understand what we've done, but we also hope that this will kind of help set a standard for things that other people should be looking at when they're doing editing on their own.

Brock Roberts:             And we really hope that people take these cells and do experiments that we don't have the bandwidth to do, and test them in ever expanding ways and let us know and report on it. Let us know how the cells perform and their unique assets.

Andrew Landstrom:     Yeah, and I think all that sort of transparency with the quality control really makes it user accessible and just sort of invites that degree of collaboration, that's great.

Kaytlyn Gerbin:           Yeah.

Brock Roberts:             Yeah, we hope so. Yeah, we think so, too.

Andrew Landstrom:     So how many cell lines do you have available?

Kaytlyn Gerbin:           Yeah, so, currently, and again you can access all these lines on the website, but we have 16 lines that are released that have gone through the full QC process. Those are available now, and we have another six that are listed as in progress, which means that they will be released very soon.

                                    Just to give you a few examples, so again, we're tagging proteins to label organelles in the cells. So, a lot of times, you know there's a lot of different kinds of proteins you could use to tag an organelle, so we've chosen a subset of those. So, we've tagged, for example, Tom20 to label mitochondria, Lamin-B1 for the nuclear envelope, alpha tubulin to look at micro tubules, and we also have started doing a lot more endosomal trafficking pathways, so like the endosome, lysosome, peroxisome, for example, and then a few other epithelial markers such as tight junctions, desmosomes, and actin. 

                                    And so, there's a kind of a bunch of structures. Those are just some examples of what we've been starting with tagging, but one of the reasons why we chose to use induced pluripotent stem cells for this whole model is because they do have the ability to differentiate into many cell types. And, I mentioned earlier that we chose to start with cardiomyocytes as a key cell type to look at, and so all of our cell line, as part of the QC process go through a cardiomyocyte differentiation protocol. And that kind of helps us ensure that the cells are pluripotent and that they can become a defined cell type and that the structure that we've labeled still is present in that differentiated cell. But it also means that we can start looking at some really interesting things in terms of how these structures change during differentiation and change from the stem cell state to the cardiomyocyte state. And so, one thing that we really started doing towards the end of last year, and we have lines coming out, hopefully soon on some cardiac specific tags. And so, to give you a few examples of things that we're working on, we have cardiac troponin I 1, and this I think will be available, I think it's passed QC and will be available pretty soon. And then we also have, we're working on sarcomeric alpha-actinin, titin, some gap junctions so that connexin 43, and then also starting to do a few signaling pathways and one that is of particular interest for the cardiomyocyte field would be beta-catenin for Wnt signaling.

                                    So, we are kind of expanding on that list as well. So, we're really excited to start looking at these cardiac structures in the cells.

Brock Roberts:             One way to summarize kind of, our strategy and one thing unites all of the different gene and protein targets that we have produced and focused on so far is to really think about the product gene or the protein as a reporter for an organelle or a structure in the cell. So, there are of course an extraordinary number of genes and proteins using this method, and there are many different justifications that would fly for why you would target a particular molecule, a particular gene, a protein of interest, but, what we really try to focus on are proteins that serve as a reporter for a structure. 

Andrew Landstrom:     So, have you tagged any ion channels?

Brock Roberts:             We have several targeting experiments that are, that take advantage of tagging the transporter molecule. One that is available is a transporter in the mitochondria, a transporter to the outer membrane, Tom20. And we're also making connexin 43 available for gap junctions. These proteins that function as trans-membrane transporter molecules accommodate the approach quite well.

                                    Another that is a bit further behind, but we hope to make available before too long would be marker of the sarcoplasmic reticulum and cardiomyocytes. This is a serca protein.

Andrew Landstrom:     So, with all these cell lines at your disposal, you've spoken to, sort of, the dynamic changes that occur both in differentiation of cardiac myocytes and cellular development and cell physiology, what are some other thoughts that you have that these lines might be able to show us? What are some fields that might be immediately informed by these models?

Kaytlyn Gerbin:           I mean, I guess just kind of on a big pictures I think that having the ability to study live cells and look at different structures in the cell will help us better understand structure-function relationships. So I think that in cardiomyocytes that, you know, makes a lot of sense, but I think even just in the stem cell field, being able to understand how localization of a particular organelle corresponds to a different state that the cell might take.

                                    And so we kind of are thinking about a lot of these different stages and states that the cell can pass through and how do we characterize that based on just kind of at a healthy or just kind of quiescent state, and then comparing that to different protivations, so looking at disease or maybe change in time, change in mutations, drug response, response to stress and how are the structures changing and how does that kind of dynamic integration effect how the cell behaves as a whole?

                                    And I think that that's one thing that we're really trying to do at the institute that is out of the scope that a lot of federally funded academic labs can do. A lot of times people are focusing on specific pathways or a specific molecule or a specific protein and don't necessarily have the bandwidth to look at the cell on a systems level. And so, kind of as Brock mentioned, with doing these different proteins as tagging the organelles we're hoping that being able to integrate that and generate enough data where that starts to become predictive I think can be really, really powerful. So...

Brock Roberts:             Yeah, and there's another thing to add that's is kind of a larger thought that we are very preoccupied with and interested in, which is to take kind of a post genomic view of biology and cell biology in particular. Genomics has been so explosively successful in allowing us to document and document the state of cells at the level of which genes, which of the many, many, hundreds and thousands of genes are active in a particular cellular state, in a particular cell type or particular state that that cell's in.

                                    We can easily get lists of genes that we know are functioning and turned on. What we want to do is take that to the next level and start defining a cellular state as a combination, a particular combination of dynamic behavior of those molecules which we can actually see. So we want to be able to see the parts work together. Not just have a list of the parts, and define states in that way.

Kaytlyn Gerbin:           Yeah, and I think you kind of asked about what kinds of communities might find these tools useful and I think lot of the disease, we're thinking about how this might apply to disease modeling or drug screens or even developmental biology and kind of studying things like that, so I think a lot of our, we have some collaborations, and we've also been really trying to expand what kinds of groups and communities are using the cell lines.

                                    There's been a lot of, kind of positive feedback on people taking, you know, a highly defined cell type that has a lot of QC done, and then having the right tools to be able to start to look at things like that. So, I think we kind of mentioned some of the tools we have, but I just to kind of restate that, all of the cell lines that we listed, along with many more, are available at Coriell. 

                                    And then, in addition to that you can get the plasmids that we've used, which have gone through also an extensive QC, so if people are working on patient derived, on their own patient derived IPS lines, you know, you could get the plasmids for whatever reporter and then put those in to your own cells. And we do have protocols available that describe our whole process in a lot of detail for how to do that and kind of different QC steps along the way.

Brock Roberts:             Yeah, we describe each targeting experiment in enough detail for it to be, we hope, recapitulated in any human cell line or cell type without too much strain on behalf of the person that's doing that work. So we hope to kind of inspire people to try this, even if they might not be familiar with it.

Kaytlyn Gerbin:           Yeah, and another thing is that our data is also available, so all of our imaging data that we're doing, and then, you can actually find that we have a website called the Allen Cell Explorer. And from there you can go through and look at all of the imaging, no, pretty much all of the images that have gone through our pipeline are now on the website. So you can go through and actually look at individual cells that were imaged during what, you know, as live cells, and then look at different structural tags that are in there.

                                    Another thing that you can see on that is the predictive modeling, and so what we're able to start doing is predict the structure of, let's say, mitochondria based on the nuclear shape and another organelle that's in there. So, we're able to start doing a lot of that. So that, I think, will be really useful to people.

                                    We going to add about the label free...?

Brock Roberts:             Yeah, and some of the more interesting results that have emerged recently are, are the ability to infer through machine learning approaches and neural network approaches the status and state and sub-cellular localization of certain organelles in the cell and structures in the cell that are actually unlabeled. Those can be inferred from the sort of sophisticated analysis of bright field, you know, images that are not displaying any particularly obvious properties, any tags or anything like that.

                                    But because the work has been done in the background to train these models and deep learning approaches with individual cell lines that do have these very specific reporters of distinct structures and organelles, because that data set exists, our modeling team and imaging team is able to appear actually quite deeply into the state of cells that are actually not labeled.

Andrew Landstrom:     Wow.

Brock Roberts:             So, it's pretty, pretty interesting.

Kaytlyn Gerbin:           Yeah. Yeah, we don't have that up on our website yet, but that's in the works to get that actual predictive model. So essentially what that would mean then is that you could take a bright field image in your own lab and then put it into this model, and then get information about maybe where the nucleus is or where the mitochondria are or where the actin is predicted to be.

                                    And all that is actually trained off of thousands and thousands of images that have come through the imaging and then the modeling pipeline. So, I think that that tool itself, once that is out and fully QCed I think could be, have a big impact right away. At least we're hoping that it will.

Brock Roberts:             Hoping. And those computational algorithms are among the publicly available tools that we have that can be found through our website, and our publications that are coming out.

Andrew Landstrom:     That's absolutely fascinating. Are you able to provide a specific example of how you've used, sort of, artificial intelligent, deep learning predictive modeling to infer a physiologic sale or response that was not directly observed?

Brock Roberts:             Well, I think the response is, we're really hoping to go in that direction. To use this to, I guess, if you will, take shortcuts toward a response in the form of a state change after we alter the environment in some way, or perhaps alter the genome to mimic a disease, mutation, or something like that. Right now, we are building the relationships. So, we can, we know, and I guess one example we can give is progress through the cell cycle.

Kaytlyn Gerbin:           Yep.

Brock Roberts:             That would be one kind of clear example that we, we haven't done a lot yet to manipulate the cellular environment or trigger cells to go through different states, but obviously cells that are in culture and proliferating alter their state by progressing through cell cycle. So that's one example that we can detect. We can clearly look at how the morphology of cells and different cell cycle states that emerge that are their chromosomes are compacted or dispersed as they undergo synthesis or undergo division, progress through metaphase and so forth.

                                    We can look at those cues and connect the state of the cell with respect to the cell cycle, to the state of some of the organelles, with the state of the mitochondria, for example. And we're hoping that same approach will hold up when we trigger, in some cases, more subtle changes to the physiology of cells.

Andrew Landstrom:     That's particularly fascinating. I think the, you know, the ability to leverage that in the setting of, like you were mentioning, patient derived IPSCs from heritable diseases. You know, these sort of monogenic disease models that impart a biophysical defect in the cell could then be modeled and not only directly observed, but perhaps indirect cellular physiology might be inferred in a way that we really haven't been able to do so previously.

Brock Roberts:             Absolutely. Yeah, there would be, in some cases there are monogenic disease mutations and pathologies that we know ought to have an effect, and we're really excited to see if that holds up, and how that holds up and what their phenotype is when looked at in this sophisticated way. And then there are other, more mysterious mutations that would be really excited to see a phenotype in.

Kaytlyn Gerbin:           Our goal at the institute is to build the tools and provide the resources to the community to be asking these kinds of really detailed, very interesting questions. I mean, I think there's definitely interest in doing some of that work here, but our main focus is to design the tools and the methods and make that all available to the public as soon as it passes our QC. So, that's the, those are the kind of thing that I think the community will have a big impact on, testing these kinds of things in their own systems given you know, new tools and ways to do it, so.

Andrew Landstrom:     Right.

Brock Roberts:             Our whole, our whole ethos is to cooperate and to facilitate. And rather than compete with other investigators, we want to make things possible and that are shared and open. For example, our list of genes that we went with to target, that was on open collaboration. We asked as many specialists in the academic community as we could to develop a consensus of what would be the most useful markers for different organelles. And we chose those proteins and genes. So, we're really trying to be collaborators, as best as we can.

Andrew Landstrom:     Are there specific examples of collaborations that you've felt were particularly productive or yielded some new exciting insight?

Kaytlyn Gerbin:           Mm-hmm (affirmative), yeah. I could give you a few examples. So, Doctor Ben Freedman, who is at the University of Washington, he is working on kidney research. And so, he has a few of our cell lines, he actually, it's convenient because we are located right across the street from each other, so we'll see them fairly often, but, yeah. So, he works on kidney research with the different cell lines and he really wanted to get the cells into a 3D context. So, he is working on a lot of different tissue engineering to study developmented disease.

                                    And so, he's also starting to make his own, their own mutations in the cell line, and so that's been, at least so far, that's been one collaboration that's I think has really been very powerful. And it's cool because we don't have the bandwidth right now to be looking at kidney organoids, but I think it's showed, kind of, the power of these kinds of cells and tools that, you know, when you have that you can do the live cell imaging with different structure within the same kind of organoid and you can get a lot of information, and so ...

Andrew Landstrom:     Yeah.

Kaytlyn Gerbin:           That, that's been fun to see develop. Another one that I know, Chris Chen at at Boston University is using our lines and is making cardiomyocytes with them as well and looking at the effects of patterning. And patterning is something that we also planning to do here, but that collaboration has been great to kind of get things going.

                                    And we've also been working closely with a group at the University of Washington, Georg Seelig's lab, who's developed a new way of doing single cell RNA sequencing. And so, that's been fun, we've been looking at that with stem cells and then cardiomyocytes to, kind of help, help us figure out what the different states that the cells are in. And then that is going to help and form, kind of, what future tags we might do or when, when to do imaging or kind of what protivations we want to put the cells through.

Andrew Landstrom:     That sounds like you're spanning the gamut really of downstream experimentation on these lines.

Brock Roberts:             Yeah, and we've also had a lot of people buy the cell lines.

Kaytlyn Gerbin:           Yeah.

Brock Roberts:             Acquire them through the Coriell, we hope that each case of their productive application toward different research questions could be defined as a mini collaboration. Maybe we'll hear from some of these people. And in some cases we have, and there may be more things that spring out of that.

Kaytlyn Gerbin:           Yeah, I think like, because the lines are available through Coriell it's, it's a little early to start seeing publications from the stuff, because we're a pretty new institute, but we do keep track of where the lines are going and, I mean it's exciting to see, I mean, pretty much all throughout the world there's people ordering the lines and starting to do research in a lot of different kinds of systems.

Brock Roberts:             Right.

Kaytlyn Gerbin:           So, we don't always necessarily directly collaborate with the people that are using the lines, but a lot of times we do hear from them or we'll run into people at conferences or something who have been using our lines. So that's really fun to see that its, our, you know, the work that we're doing here is actually producing things that people in the community are finding informative and useful. So, that's always fun.

Brock Roberts:             It's still so early in this project. I mean we're just at the beginning of a lot of collaborative potential. So, we really hope to see this take off.

Andrew Landstrom:     Yeah, and if people listening want to collaborate or want to learn more, how can they learn more and how can they get ahold of you all?

Brock Roberts:             Oh, hold on, I think, first of all, we really want to funnel people to our website. We think it's a really great resource and at that you can contact us through that link. We look forward to hearing from you.

Kaytlyn Gerbin:           Yeah, so you can start with the website there. And as we mentioned before you can find all of our cell lines, plasmoids, protocols, etc. on this site. And we've also started to do a few more instructional videos, and so those are coming up on the website, too. So, some things, you know, especially as more groups are starting to use the lines, we do have the detailed protocols, but I think groups that maybe haven't done stem cell culture before or haven't worked with these kinds of IPS lines before, we're trying to provide as much content for people to make it easy for them to do the research. So we're starting to do more, sort of instructional videos.

Brock Roberts:             Yeah, and we seek this out. We want to hear from people. It's not a bother. We're trying to get as much, we're trying to get the, we're trying to branch out and communicate as extensively as we can.

Kaytlyn Gerbin:           Actually, one thing I just thought of that I want to add in her is that we have started to work with a few stem cell cores. And so, right now, I mean-

Brock Roberts:             These are core facilities at universities.

Kaytlyn Gerbin:           Yeah, stem cell core facilities at, yeah, exactly. So, part of trying to distribute the lines is that if we can, you know, individual investigators could get our lines from Coriell and get the licensing and everything to do that in their own lab, but it's, I think, going to be really great if we get some connections with the stem cell cores because then once we can provide the lines to them, they can distribute them to investigators that are part of the core.

                                    And so, so far, we already have agreements in the works with University of Washington, UC Berkeley, and then the Salk Institute, but this is something that we're really hoping to expand this year. So I think, in particular, you know definitely contact us if there's questions about the lines or anything, but if you are part of a stem cell core at a university and you think that people at the university would be interested in using our lines we're working really hard to make, you know get, kind of, packages, protocol packages and everything available so that we can get these lines set up in the stem cell cores.

Brock Roberts:             Right.

Andrew Landstrom:     Well Brock and Kaytlyn, thank you so much for joining me. What an incredible resource that you all have created, and I especially appreciate how open and transparent you are with your lines and your quality control and how you just really, you know, try and strengthen collaborations and to start new ones.

Brock Roberts:             Thank you very much for the conversation.

Kaytlyn Gerbin:           Yeah, this has been fun, thank you.

Jane Ferguson:             I hope you enjoyed listening to this episode of Getting Personal: Omics of the Heart. Let us know how we're doing by leaving a comment or tweeting at us at @circ_gen. We love to hear from you.