 Good afternoon, this is Dr. Jean Jenkins and welcome to our second webinar in the JNS series of presentations about the articles that you can find in the new genomic special issue published online, February 1st. We're very excited to have Yvette Conley talking with us about current and emerging technology, which is in genomics today. You have the opportunity to post questions, and those will be addressed at the end of the session. You are on mute, but you can also raise your hand and comment by way of typing in. We will be recording the session, and we will now start the recording. Also if you are representing more than just yourself in attending today, could you please note for our records by typing in how many people are actually in your room or working with you as part of listening to this webinar. And just a reminder that our next webinar will be March 5th. It will be a dual webinar from 3.30 to 4.30 with two topics, one by Dr. Xiufen Wang on cardiovascular genomics, and the other an overview of the genomics and metabolic syndrome by Anne Cashin, Jackie Taylor, and their colleagues who were also involved with writing the paper. So thank you so very much. I'm going to turn over the introduction to Dr. Kathleen Calzum. Well, good afternoon, everyone. We're delighted to have Dr. Yvette Conley speaking to you this afternoon. She has her master's in genetic counseling and a PhD in German genetics, both of them from the university to be her. And she's a tenured faculty member and associate professor with a primary appointment in the School of Nursing, as well as a secondary appointment in the Department of Human Genetics, all at the University of Pittsburgh. She's been managing and directing molecular genetic laboratories for over 20 years and is fully equipped to, and has a fully equipped laboratory that's actually in the School of Nursing that's focused on molecular genomics. Her research has focused on using a variety of genomic approaches to understand the biologic basis underlying variability in patient outcomes. She's had funding from the National Institutes of Health, the Department of Defense, the Alzheimer Association, and the Oncology Nursing Society. She's also director of an NINR-funded T32 titled Targeted Research and Academic Training of Nurses in Genomics, and continues to be a primary faculty member for the National Institute of Nursing Research Summer Genetics Institute. And she's a standing member of the NIH Nursing and Related Clinical Sciences Studies subsection. So there's really no one who's sort of more uniquely positioned to speak to technology and how it applies to genomics and its application for nursing practice. So we're delighted to welcome Dr. Connolly. Okay, thank you very much. I want to start off by thanking all of the attendees and also thanking my co-authors who's on behalf I'm presenting today. And the co-authors include Leslie B. Beesacker and Stephen Clesalves from the National Human Genome Research Institute, Carrie Merkel from the University of Arizona, Maggie Kirk from the University of Glamorgan in the U.K., and Brad Lizarat at the University of California, San Francisco, are my esteemed co-authors and without whom I would not be able to talk about all of these things. So let's move ahead. I'll tell you a little bit about what I'm hoping to talk about during the webcast. So I want to briefly discuss the approaches that are introduced in the manuscript. So those being genome sequencing, genome-wide association studies, epigenomics, and gene expression profiling. Now I do want to point out that I'm not going to be covering details on how to conduct these types of studies or the various platforms that you can use to conduct these types of studies. I think we just don't have time to cover all of those things, but the information I'm going to provide is more about how these approaches are emerging and being utilized and increasing in utilization for both research and clinical applications. So don't worry, we've provided in the paper and I'm going to address some of these towards the end of the talk. We included a lot of online resources in the paper to assist with some of this information. I'm also going to address implications for nursing practice and research, and then I will, time permitting, take you guys to some of the databases and resources that are out there online. And then we'll have a question and answer session. So as far as genome sequencing is concerned, what we're interested in with sequencing in general is we're interested in determining the order of nucleotides. So with that, the order of G's, A's, T's, and C's in the DNA. So in the past, sequencing has been around for a long time. In the past, what we did was we did sequencing. We figured out the order of G's, A's, T's, and C's for a very focused piece of DNA. So a lot of times what you needed to do is you had to know what piece of DNA you wanted to look at. And then we had technology to help you focus in on that particular spot in the DNA that you wanted to sequence. And you would collect that data, that focused data, one base at a time using methodology that gave you one base at a time information data collection. That's all changed. Now I have to say those old school methods of collecting, sequencing data, one base at a time, and focusing in on a particular piece of DNA that is still useful today. But now we do have, through advancement of technology, the ability to sequence many pieces of DNA simultaneously at the same time, so to collect these data in parallel. And it's also possible to not have to focus on a particular piece of DNA anymore. You can actually look across the entire genome and sequence the genome now. Now this idea of in parallel sequencing many pieces of DNA at the same time is referred to as next generation sequencing or massively parallel sequencing. Those are equally correct terminologies to use and a lot of folks will say next gen sequencing. And what they're referring to is the ability to collect DNA with this higher throughput technology that collects sequences, many pieces of DNA at a time. So as you could imagine, having the ability to actually look at greater sequencing coverage and to the extent that you could look at the whole genome from a sequencing point of view. And you can do this in a lot less time and simultaneously the cost of doing this has come down. It makes genome sequencing way more appealing for research and clinical utility. So it is something that's out there being used and it is on the forefront of being adopted by a lot of research and a lot of folks doing clinical research and then looking to translate that research to clinical applicability. We'll talk a little bit more about that. So whole genome sequencing basically is you're sequencing the entire genome. Now, there could be gaps in what you're able to collect from a data collection point of view, but for the most part you're capturing the sequence of the entire genome. Whole exome sequencing, so an exome is that piece of the gene that goes into the coding, that's the coding region of the gene. And your exomes, your collection of exomes across all of your genes across the entire genome is called your exome. And so we also have the ability to sequence the whole exome of the genome. So you're not doing the whole genome, but what you're doing is you're focusing in on the exome, which represents about 1% of the genome, but is likely to contain about 85% of the variability that influences the phenotypes that we're interested in to get a lot of bang for your buck out of whole genome sequencing. You're only sequencing about 1% of the genome, so it's a little bit quicker to do and a lot less expensive than whole genome sequencing, but you're capturing a lot of usable data. So why would one even want to entertain doing a genome sequence? Well if you take a step back and think about what your areas of interest are, probably the phenotypes or the traits or the conditions that you're interested in are not single gene conditions. These are probably conditions where there's multiple genes involved, maybe even gene environment interaction, and you would like to be able to look across all genes or regulatory regions of the genome, and you would like to know a little bit more about how that genome, the variability that that genome holds for telling you about your phenotype of interest. And because a lot of the conditions of high public health importance are going to be conditions where there's more than one gene, more than one variation in the genome that is impacting that condition, genome sequencing becomes very appealing. So one real advantage to genome sequencing is that it captures rare as well as common variation. So as far as changes to our DNA that can impact our health, there are, you know, there's a lot of folks out there who believe it's rare variation that's going to be what tells us, you know, the most about our phenotypes of interest. There are folks who believe that it's going to be common variation. There's a lot of us who believe it's probably a mix of both, but genome sequencing satisfies both camps of thought because you actually will capture all of the variability in someone's genome by doing the genome sequencing. So you can see where there would be appeal to that. So it's a very powerful approach. This is something that is, you know, very much if you can do it gold standard. You're getting the DNA sequence of an entire genome of an individual. So a couple of things that folks conducting research, folks who are looking at the research for potential clinical utility, and folks that might be using this down the road for clinical utility, some things that you need to keep in mind is that it is an error-free. And as the technology advances and as we start collecting more data and as we start to get better at managing the data, collecting the data and viewing the data, we'll probably have error rates go down. But right now, the average error rates for next gen sequencing is about a half a percent to two percent. So there's a lot of call in the genomics community for labs who are reporting this sort of data, this genome sequencing data, or exome sequencing data that you need to when you're doing these reports, you need to say what genes or what exons or what regions of the genome that were not adequately assessed while you were collecting those data. And that could be because you didn't get any data from a particular piece of DNA or you got data, but the quality just wasn't high enough for you to feel comfortable with that. The people who are getting the report on the other end need to know that that's something that came out of your laboratory. The other thing is the genomics community is calling for, because this is a newer technology. And anytime there's a newer technology, comparing that to our more traditional technology that's been validated is something that we try to do for a while until we feel more comfortable with the newer technology. So there is a push out there for anything that you find with next gen sequencing that it's a good idea to validate those findings with a more traditional method to validate those findings. The other thing that I wanted to make sure we brought up was the idea of incidental findings. And this is a pretty hot area in genomics right now. It's an ethical, legal, and social issue that has come up in the genomics community. And you can imagine how you set out to look at someone's genome. And you're looking for variation that's involved with your phenotype of interest. Because you're going to get a lot more information about that person's genome than what impacts your phenotype of interest. What if you were to find variation not involved with your phenotype of interest, but that has clinical importance not related to your phenotype? What do you do? When you uncover that information, what do you do? What are your obligations to the participant, the person whose genome you're looking at? And this is a very hot area right now. Over the past year, genetics in medicine has produced quite a few papers on this topic. And genetics in medicine is a product of the American College of Medical Genetics and Genomics. So I would bring you out to look at some of those articles about the topic. And they've even coined another ome. There's lots of omes and lots of omex out there. But the incident ome is basically the genome, the information that you find out about an individual by looking at their genome sequence. But it's stuff that you weren't expecting to find. And what are researchers supposed to do when they find an incidental finding that has potential clinical importance? But the samples that they're using are de-identified. The samples that they're using are from a biobank. The samples that they're using, those subjects were not consented to go back and give information to them. It's really raising a lot of interesting dilemmas. Because once again, what's happened in the genomic community is our ability to look at the genome, characterize the genome, our technology, to do all these things far surpasses and is far ahead of what we deal with, the massive amounts of data that we have. And so I think that once again, technology is much further ahead than what we can deal with when we're talking about doling out findings to folks. So very common issue right now that we're dealing with. And I would recommend that you look at some of those manuscripts that have been published in Genetics and Medicine. OK, so genome-wide association studies, or GWAS, as it's affectionately known. So this has, this is an approach that's been around a lot longer. You may have heard more about it. So genome-wide basically meaning that you're collecting genotype data for polymorphisms that span the entire genome. So up and down, every chromosome are polymorphisms that are spaced on that chromosome. And you collect genotype data for each of those polymorphisms that span the genome. You then take those data and you analyze it statistically using an association approach. So by association, this is no different than statistical associations that you would look for in data that isn't genetic data. So you're looking for a relationship between two variables that make them statistically dependent. Now in this case, it's your allele or your genotype of interest and your phenotype of interest. So you're looking to compare what someone's genetic material looks like to your phenotype and seeing if there's a relationship. So at the very basic level, what you could do and what people do do is they will analyze each polymorphism to see if there's an association with their phenotype of interest. And what you do is you run a multitude of simple association tests. Now we do have more sophisticated analyses that are available to us. For example, there are ways to look at gene-gene interactions so you can look to see not only is this particular allele or this particular genotype in this gene associated, but in reality it's an allele in this gene and an allele in that gene and an allele in this gene. And when I look at them in an additive way, I'm actually seeing that there are interactions between those genes and I have a strong association. So it's a little bit more sophisticated. Pathway analyses where you take genes that are known to interact with one another in a biological pathway and you look to see do those genes synergistically together have an association with your phenotype of interest. And so those are some very robust ways of looking at GWAS data. So the types of polymorphisms that are used most often in a GWAS are your single nucleotide polymorphisms. And most people today who are conducting GWAS experiments, they are collecting data on anywhere from about a half a million SNPs to about 2 million SNPs. So the reason why a GWAS works is this idea of linkage disequilibrium. So it turns out that if you think about how you get your genetic material from your biological parents, you do not get that material one base at a time. You get your genetic material in chunks called chromosomes. Now, it's true that when your biological parents made gametes, there was recombination and reshuffling amongst those chromosomes that they have. But once they donated that genetic material, then that material was inherited as a chunk. And so these chunks or blocks of linkage disequilibrium are very helpful to us. It turns out that let's say there's 300 polymorphisms in a particular chunk of DNA, a particular block of linkage disequilibrium within our DNA. And if you were to genotype all 300 of those polymorphisms, you would find that if you had genotype just one of those polymorphisms, you probably could have predicted the genotypes for almost all of those other 299 polymorphisms in that chunk of DNA. So it turns out, luckily for us, that we just look at one highly informative polymorphism in that chunk, in that block of DNA. And those are the SNPs that are then chosen for a GWAS. And so when you talk about a half a million or two million SNPs up and down all the chromosomes, those don't represent all the polymorphisms in the genome, but it represents a subset that help us capture the majority of the variability in the genome. And so you're looking for association with a chunk of DNA when you do a GWAS analysis. Now the most common design is a case control study. And so you could think of this like any old case control study, you have people with a particular phenotype of interest compared to a group of folks without that phenotype of interest. And you look to see, do those two groups differ significantly based on allele or genotype or haplotype frequencies for any of these polymorphisms? So the biggest advantage to a GWAS is that it gives the investigator a nonparametric way of evaluating the genotype-phenotype connection for that phenotype. So what I mean by this is you have all these polymorphisms that are up and down every chromosome. You then collect the data for those polymorphisms. You take that data, you analyze it, and you look to see what SNPs or what chunks of DNA look to be associated with my phenotype of interest. And only then do you look at, OK, well, this one is significantly associated. Let me see where in the genome this polymorphism that I have an association with. Let me see where that's hanging out. And if it's hanging out in the gene, then you can look and see, OK, well, what gene is that hanging out in? Sometimes you're going to say, oh, that's a very logical gene. But a lot of times, and the benefit here is it could be a gene or set of genes that you weren't even thinking about would be associated with your phenotype of interest. So that's what I mean by big advantages is you collect the data and you let the data tell you what might be interesting from a biological point of view for your phenotype of interest. And that's a huge advantage when you think about the fact that most of us, the phenotypes that we're interested in, we'd be fooling ourselves if we tried to say that we understood everything about that phenotype. So this gives us a way to circumvent the fact that we may not have all the knowledge that we need to a priori say what genes we would look at. We're just going to look at them all and let the data tell us what looks interesting. So that's a big advantage of a GWAS. So some considerations and potential drawbacks. So it does require a large number of subjects to conduct these studies. So if you think back to when I said folks are looking at a half a million SNPs, two million SNPs, you might be thinking to yourself, wow, that's a lot of data. And that would be very true. So multiple testing is an issue when you're looking at that many variables. And so you need to look at large numbers of subjects to offset the fact that you're going to have to correct for multiple testing if you're going to conduct these types of studies for the most part. And that large sample set requirement does turn some people off from doing this approach. The other thing is there are some phenotypes where you would really struggle to get large numbers of subjects. And by large numbers, I'm thinking most folks are saying you're anywhere between 1,000 to 2,000 cases compared to 1,000 to 2,000 controls. But the larger the better. You could think of some phenotypes where you would really struggle, and maybe over the course of your lifetime, maybe not be able to collect all those subjects. But then there are some phenotypes that are amenable. And this is why I see a lot of consortiums popping up for GWAS studies because people need to pool their resources. They need to pool their subjects in order to do some of these more non-parametric approaches. The focus is on common variants. So unlike the whole genome sequencing where you're going to capture rare and common variants when you do a GWAS by design, you are collecting data on common variants. But one thing that I do want to point out is when you have an association with a chunk of DNA that's in linkage disequilibrium with your polymorphism that you have an association with, be mindful that that chunk of DNA could house a rare variant. And what's happening is you have an association with this common variant, but it's a surrogate for a more rare variant that's hanging out in that block of DNA. It is an association. So sometimes you have to remind folks that an association is not mechanism. And so when you have association, you need to be mindful you have an association with a chunk of DNA, which may actually have an association with a surrogate marker for something else that's going on in that chunk of DNA. And so your work really isn't done. You need to characterize that chunk of DNA or chunks of DNA a little bit better. And you may need to figure out, well, what's the function of the variability that I'm seeing in that chunk of DNA if you're going to get a little bit closer to mechanism? But a lot of times when we talk about conducting candidate gene investigations, a lot of times the data that goes into what candidate genes you investigate, a lot of times that data comes from a GWAS and the data pointed to some very high priority candidates for us to look at for a particular phenotype. So gene expression profiling. So again, gene expression profiling been around for a while. And the whole basis of gene expression profiling is based on the fact that in every gene, I'm sorry, in every cell of your body, you have all the same genes represented. So every somatic cell in your body, whether it came from lung, whether it came from skin, whether it came from the lining of your mouth, whether it came from the lining of your gut, doesn't matter. If we took those cells and extracted DNA, every gene is represented in those cells. But not every gene is needed in every cell. And so not every gene is expressed in every cell. And so what we do with gene expression profiling is we actually look to see can we characterize what genes are actively expressed in a particular cell, a particular tissue of interest. Now why would that be interesting? Well, if you think about comparing genes that are differentially expressed between two different phenotypes of interest, that could tell you a lot about the underlying, the biology of the underlying phenotype. So what differentially expressed genes are there that you can characterize that can tell the difference between your phenotypes of interest could be clinically useful. So an example, so if you had a phenotype of interest, you could look at tissue from an individual that has the condition compared to that same tissue in an individual that doesn't have the condition. You compare the gene expression profiles from those tissues and genes that are up-regulated or down-regulated, differentially between the two, it's gonna say a lot about what might be, what concert of gene regulation might be going on that's leading to your phenotype of interest. For example, you could even look at the cellular level even within an individual. For example, you could compare healthy tissue within an individual to unhealthy tissue, so tumor tissue. So what genes are up or down-regulated in tumor tissue versus normal margin? And you could capture a lot of information about that tumor by looking at what genes are up or down-regulated. Now we actually exploit this gene expression profiling clinically already, and while I don't want to talk about a lot of different commercially available assays and things like that, suffice it to say that there are gene expression profiling arrays that are being used clinically. One particular example for risk stratification, for prognostication, for tailored intervention, a really good example is in breast cancer where a woman who has presented with breast cancer, her abnormal cells involved with her breast cancer, those are taken and clinically characterized using gene expression profiles. And being able to say what genes are up or down-regulated in that woman's tumor tells us a lot about what type of breast cancer she has. Breast cancer is very heterogeneous. So what type of breast cancer does she have? What is her potential prognosis? What therapy? What type of chemo or other adjuvant therapy might work best for that individual with breast cancer based on the biology of her tumor? We are already doing this. This is already being done clinically. For those of you who are involved in oncology, you're well aware that we are using gene expression profiling, not just in research to figure out what genes are important to what cancers, but we're actually using them clinically to make some very important decisions for these patients. So gene expression profiling relies on RNA. So the sequencing that I talked about, GWAS that I talked about, in those situations we are looking at the bases of the DNA. Here we're looking at RNA. So, and when you're talking about looking at RNA, you can look at a particular type of RNA from a particular candidate gene, or like a GWAS, you could actually look at the level of gene expression for every gene across the genome. And what you can do there is you can actually, at the most basic level, do what you did with a GWAS. You could look at every piece of data that's generated across the genome for gene expression. And you could look at each piece of data. You could say, okay, I'm looking at this level of expression for this particular piece of RNA, and I'm looking to see if it's associated with my phenotype of interest. And again, if it is, then because you looked at it, you looked at all the genes, you looked at it from a whole genome point of view, a nonparametric point of view. You might see expression of genes that are up or down regulated from that whole genome approach that you expected to be altered in its expression. Or you might find some novel genes or novel pathways of genes that you wouldn't have expected to be involved with your phenotype of interest. So that's, again, one of the benefits of the nonparametric evaluation. And just like GWAS, you can look at those individual pieces of data for gene expression, or you could look at how genes might, expression of one gene or a whole host of genes taken together might impact your phenotype of interest. So a couple of other things that I wanted to point out that are emerging. MicroRNAs are becoming very, we're learning more about them. We don't claim to know everything about them at this point in time. But they're very interesting. They are definitely involved in gene regulation. Not so much involved with, like messenger RNAs are with taking that template and translating it into a protein. These RNAs are not involved with making protein. They're involved with gene regulation. And what they are is small pieces of RNA that are transcribed and they are the antisense to messenger RNA for another gene. And the antisense RNA from the microRNA hooks up with the messenger RNA from that other gene and basically prevents it from being translated into a protein. So you can imagine how these microRNAs are impacting gene regulation. It's another level of gene regulation that you can assess. Now what's interesting is if you do these whole genome gene expression data collections, a lot of the standard platforms and technologies that are out there now for you to do the whole genome expression evaluation includes most of the microRNAs that we know about. You'll simultaneously not only evaluate expression of annotated genes, but you'll also evaluate expression of these microRNAs also. So again, lots of advantages to doing a whole genome expression. Couple of things that I did wanna point out because we're talking about RNA here not DNA. So things that you have to be mindful of when you are conducting this type of research or when you're evaluating publications of this type of data. And when you're thinking about clinical utility of this data, DNA is very hardy, extremely hardy. RNA is not. So folks who are working with RNA need to stabilize, appropriately stabilize their RNA. That's an important factor to consider because if you are looking at a publication and you're not, you don't feel confident that they stabilize their RNA, then you have to call on the question, how valid are their findings and then how applicable are those findings ever gonna be to the clinic? Other issues with gene expression. So gene expression differs across tissues. So if you remember what I said, take a lung cell, you take a kidney cell, you take a gut cell, take a skin cell, those genes are all there. The polymorphisms are gonna be the same amongst those cells. But what isn't going to be the same is gene expression. So when you're talking about doing a gene expression profile or you're talking about doing research, using gene expression, you need to give a lot of thought to what tissue is most appropriate for your phenotype of interest and that you are comparing apples to apples when you're comparing tissues because you can't necessarily compare a lung tissue to kidney tissue with gene expression because just because they have different jobs to do, they're gonna have different genes expressed. So you have to be mindful that there's gonna be variability from tissue to tissue. The other interesting thing about RNA that is distinct from DNA is that it is dynamic. Gene expression is gonna change. It's gonna change in response to exposures. It's gonna change in response to, there are temporal changes to gene expression. There are genes that are expressed in early development that are shut off and no longer needed in the adult. So there are a lot of changes in gene expression. It's very dynamic, which is not something that we see when we're looking at the basis of the DNA. So these are things that need to be taken into account when you're either conducting a RNA-based or gene expression-based study or evaluating those types of studies out there in the literature. Okay, so a little bit about epigenomics. So our sequencing and our GWAS, we were interested in DNA. Our gene expression, we were interested in RNA. With epigenomics, for the most part, we're back to looking at the DNA. So epigenomics is different. When you get back to the DNA for epigenomics, you're not interested in the basis, like you were with sequencing and GWAS. We're not interested in what the GSAs, order of GSAs, Ts, and Cs are. We're not interested necessarily with the polymorphisms that are there. What we are interested in is how that DNA is chemically modified or packaged. Now the chemical modifications and packaging impact whether or not a gene is expressed. So when we talk about gene expression, there's something going on mechanistically to impact that gene being expressed or not. And epigenomics explains a lot of variability in gene expression. So it's one of the mechanisms behind gene expression variability. Now, a lot of times people automatically associate epigenomics with DNA methylation. And I think that there's a good reason for that. And I think it's because looking at DNA methylation is something that we've moved along with DNA methylation is the form of epigenomics that has gained the most research in clinical applicability. There are some DNA methylation assays that are being used clinically. Again, mostly in oncology. But I do want to call your attention that there are many types of epigenomics, including histone modifications, chromatin structure, non-coding RNAs. So for example, some of the micro RNAs would fit here, and DNA methylation. And so I just wanted to point out that there are many different ways that the DNA can be chemically modified or packaged under this category of epigenomics. So I am actually going to spend some time talking more about DNA methylation because that is where most of the research seems to be happening. And it's certainly where most of the clinical utility for epigenomics has occurred. So with DNA methylation, you're talking about chemical modification to the DNA. You're talking about addition of methyl groups, but not just anywhere in the DNA. Methyl groups are added to GC-rich regions of DNA. So where you see a lot of repetitive GC, GC, GC, GC nucleotides, those are called CPG islands and CPG island-rich regions of the DNA are susceptible to DNA methylation. So when a region of the genome is hypermethylated, you're basically shutting down that region of the genome. So when the promoter region of a gene is hypermethylated, that gene is not gonna be expressed. When you get hypo methylation of a gene that's supposed to be methylated, you have activation of that gene. So you can see when we talk about DNA methylation being dynamic, you can see what we're talking about here. You add methylation, you shut down a gene. You take it away, you reactivate it. And because DNA methylation is key for temporal gene regulation, as well as tissue to tissue variability and gene regulation, you can see how genes can be turned on and off based on their chemical modifications. So epigenomics does share some of the methodological issues with DNA polymorphism-based approaches as well as RNA-based approaches. You're back with the DNA, so DNA is stable. And for example, if you are looking at DNA methylation, the methylation stamp that's on that DNA is also stable. You can extract that DNA and you don't have to worry about the instability that we have to worry about with RNA. But what makes it a little bit like RNA-based gene expression is that it does matter what tissue you're looking at because the methylation stamp that's on the genome of a particular cell is gonna be different than the methylation stamp on a cell from a different tissue. So like gene expression, you do have to be mindful of what tissue you're looking at. And you do have to be mindful of exposures because it is dynamic. So one of the things that can make DNA methylation evaluation a little bit messy is, it's impacted by exposures, both endogenous and exogenous. But I think what draws a lot of us to looking at epigenomics and in particular DNA methylation is that exposures do impact epigenomics and DNA methylation. And to that extent, we're interested in how our genetic material interacts with our environment. Now, be mindful that we can only talk about what we know about today. And when I was a graduate school, we did not know much about epigenomics. We knew that some DNA was methylated. We knew about imprinting, but we didn't know the extent to which our genetic material interacted with our environment. And that actually could impact methylation of our DNA and the epigenome of our DNA and how our DNA is packaged. So with an epigenomic approach, you could look at one gene of interest. You could look at the methylation status of one gene of interest. You could look at the methylation status of a pathway of genes. There are assays out there to do that. Or like GWAS and whole genome expression profiling, you could look across the entire genome. So there are technologies out there that allow you to look at methylation status across the entire genome. And in a non-parametric way, take a look at, well, what regions of methylation differentiate my phenotypes of interest? And so you can mix and match some of these approaches. And they overlap a lot, but there's little nuances about each of these approaches that the researcher and the person who's thinking about using these approaches clinically needs to be mindful of. So I do wanna talk a little bit about implications for nursing practice and research. And so I'm gonna pull directly from our paper. We have a table in our paper that deals with how across all four of these different approaches that have discussed how it impacts practice and research. So a couple of things that I wanted to point out, some of the highlights. So when we talk about genome sequencing, it is interesting to think about the fact that you could collect an entire genome's worth of sequence. And you could even do this with antenatal testing, for example. A lot of folks are thinking that what we could do is move towards whole genome sequencing clinically and have that whole genome sequence sitting there so that when it's time to prescribe a particular medication, our healthcare provider can take a look and see, okay, there's key genes that regulate metabolism of a drug that we're thinking of giving you. So in addition to all of the things that we normally would take into account when prescribing a medication, we're also gonna take into account your genomic variability. And so it would be there for us to take a look at and make appropriate recommendations for therapy, for example. And it sounds kind of futuristic, but there are a lot of folks who believe that at some point in time, we are probably going to have our genome sequences entered into our medical record and how we could utilize those data. Genome-wide association studies, again, particularly advantageous for common complex disorders, where you're thinking more than one gene is probably involved. Gene expression profiling, I already hinted at some of the things that we're using gene expression profiling for clinically, and certainly in the breast cancer population, we are using specific gene expression profiles of tumor samples to help us with accurate diagnosis, prognosticating, figuring out what would be the best treatment for that individual, and then post-treatment surveillance. For example, when some oncology patients go back in for follow-up biopsies and follow-up surveillance, we can look to see, are we seeing what we would expect to be a quote-unquote normal gene expression profile, or are we starting to see some genes abnormally expressed and is that cause for concern? And then with epigenomics, we have some clinical utility where we do have some methylation assays out there for key genes of interest. We know that loss of methylation and of oncogenes is a reason for some types of cancer and can help us differentiate certain types of cancer. So some of those DNA methylation assays can do for us in the oncology patient what we're doing using gene expression profiling in the breast cancer patient, for example. So lots of avenues for clinical implications and lots of possibilities down the road as these technologies continue to emerge and have more utility. From the point of view of implications for nursing, I think I'm probably preaching to the choir. You're all participating in this webinar, so chances are you need more education. But I think that you need to spread the word. No doubt there needs to be education efforts made. And I think that this webinar series is a huge step in the right direction. As far as research, clinical care, public health nursing, ethical practice, and then, of course, nurse leadership, there are already leaders amongst us that are looking to translate all of the new knowledge that is being brought to us through these emerging technologies and then how are we gonna use them clinically? And then what sort of ethical, legal, social issues might actually arise from conducting some of these technologies. So lots of implication for nursing, I'm sure this was the tip of the iceberg for the paper and I'm sure you guys could be thinking about additional implications. So for maybe a couple more minutes and then I'll stop and take questions, I would like to talk to you a little bit about some of the databases and resources. One of the things that myself and my co-authors believed very strongly was, if we're writing a paper about emerging technologies, particularly in the field of genomics, you know the paper's gonna be out of date, probably by time it's published. Or how useful is it gonna be a year or two years down the road? Well, to add to the potential half-life of this paper, one of the things that we decided to do was to embed in the paper a multitude of online, reliable web-based resources that are kept up to date. So that even if our paper gets a little bit out of date, readers would have resources to go to, to bring themselves up to date, to learn more about the technologies, not just how we describe them in the paper, but going forward, what's happening with those technologies, going forward, what data has been collected and stored in these databases. A lot of effort has been put into developing these databases and folks, we need to use these databases, we need to use these resources. So again, a couple of the resources that I wanted to point out, I did wanna point out Clint Seek, so Dr. B. Secker, who is a co-author on our paper, is the leader of the Clint Seek project at NIH, and so I wanted to give a call out for that because these folks are looking to develop clinical utility for whole genome sequencing. And so if you would like more information about Clint Seek, you can go to their website. I highly recommend you take a visit to the database of genotypes and phenotypes. So when we talk about GWAS studies, for example, that have been conducted, and I'll see if this will actually take me out to this website, when you take a look at what's going on here, you could go to DVGaP and you could type in, if your area of interest is cardiovascular, I'll just put in cardiovascular. You can see there are 95 studies that have dumped into this database, GWAS data that's publicly available. Now, some of this data's embargoed and you may not have access to it right now, but a lot of it is available and those that are embargoed will eventually be available. But you can take a look and see, your framing hand had 14,000 subjects in this study and the GWAS data and phenotype data is available to investigators through the DVGaP. The Jackson-Hart study, large epidemiologic studies that folks have seen in the literature for years, Women's Health Initiative, Cardiovascular Health Study, I would highly recommend that you guys go surf around on the DVGaP and see, put your phenotype of interest in and see what GWAS comes up that you might be able to tap into. Gene expression omnibus, or affectionately called GEO, is a clearinghouse for gene expression data. If you wanna know more about epigenomics, you can go to this website. If you would like to see databases of epigenomic data, mostly what's in there is DNA methylation data as well as some chromatin precipitation data the epigenome project is very interesting and contains epigenomic data from quote unquote normal individuals that you might find helpful to your research in particular for comparison group. And then just a couple of others that I threw up here and this is just a smattering of those that are in the paper. The paper has way more online resources for you to use. Genetic test registry and OMIM, very high clinical utility for you guys. I'd highly recommend you surf around on some of these websites. And just to put a plug out for the next webinar and I think we have time to take some questions. So thank you, Yvette. There has been one question posted so far if I can get to it and read it. Clarification, can you give an example of an incidental finding and how it was handled? Is it like LCM, is it related to research on that was not consented to? Okay, okay. So as far as incidental findings. So the idea of incidental findings, that's not a new concept. It's newer to us in genomics, but for example, in the field of imaging, incidental findings has been a problem. So if you do an image on someone because you need more data to clinically assist this individual, what do you do when you see something on an image that you weren't expecting and has nothing to do necessarily with the problem that your patient is having? It's the same situation. So an example would be, let's say I do whole genome sequencing because I'm interested in looking at an individual for, let's say I have a host of genes that I wanna look at variability in to say what their risk would be for cardiovascular disease or hypertension. And because I'm getting whole genome sequence, what if as I sequence through the Huntington gene, I find a run of trinocleotide repeats that are indicative of this person having the mutation that at some point in time if they live long enough, they're gonna develop Huntington's disease. What do I do with that incidental finding? It's not of interest necessarily to my phenotype, my index phenotype, but I know that there's clinical importance to that finding. What do I do? Now what do I do if this is a data identified sample? How much effort as a researcher do I now put into figuring out who is that individual? Because remember, a lot of the samples are de-identified for a reason. If you know, if you can trace back to the identity of that individual, would that individual even want to know that information? That's not necessarily what they signed up for. I have hypertension, I have cardiovascular disease and that's what you were looking for. Now you're telling me that I have this other thing. Well, I didn't wanna know that. So there's a lot, I don't know if that answers this individual's question, but that's the sort of scenario that we see playing out potentially with the collection of so much data about what someone's genome looks like. So the next question you've had is sequencing is the new light in the room of genomic research. Will future research require a blend of sequencing, epigenetic profile and expression analysis? If so, this could be very expensive and laborious. However, it appears to be the ideal. I think what is most desirable is usually most expensive. So more data is always better. And I think what you described where you're talking about looking at the sequence of someone's genome and then you're looking at the chemical modifications or packaging of that genome from an epigenomic point of view and then you're looking at what genes are expressed as a result of all of the above. That allows you to tell a nice story. It allows you to start from beginning to end. What do I see in the genome that could be associated with this phenotype and why? So you're getting the how and the why, which are very important pieces of the story. You're absolutely right. But one thing that I would say is we're using a lot of these more global approaches that are more expensive with the idea that as we start to catalog more phenotypes systematically, we're not gonna have to look at everything down the road. We're gonna know for this particular phenotype, we're gonna have to look at this particular group of genes from both variability and epigenomic and a gene expression point of view, but we won't have to look at everything. And so I think that's really what a lot of folks are hoping that we'll eventually get at. I don't know whose lifetime that'll be, but eventually we're hoping to get to the point where we have cataloged enough of these findings for every phenotype that we're interested in so that we can be more focused. And then the cost won't be so high. So I bet this will be the last question and just to let everybody know, this recording will be archived and available on this same Gov site that's listed on Ivette's slide. And this is also where you can register for upcoming webinars. And the links, obviously in the tables are available in her article, which is in the JNS issue. And so the last question, in addition to a very positive comment about how helpful this has been was that for the Jackson Heart Study from DB Gap, do you know whether it includes the genotype or SNPs associated with metabolic syndrome? So, okay, good question. So if you are familiar with the Jackson Heart Study or any of these large epidemiologic studies, if you're familiar with them from the literature, you have an idea of what variables they collected on these individuals. So metabolic syndrome, I can tell you that they have a whole host of laboratory measures on these folks. They, you know, so they have glucose. They have, you know, CRP. They have hemoglobin, HC1. They have BMI. They have, you know, lots of information on these folks. These folks have had a lot of data collected on them over time. You would have access to all the phenotypes that they have. You could then, now whether or not they have metabolic syndrome, I'm not 100% sure, but you could certainly look at the individual variability, individual variables that you would take into consideration to come up with metabolic syndrome. Look at those individually or look at them as a whole, as a syndrome. But those phenotype data as well as the genotype data, now you have to apply for these things. You have to apply to get access. It's no longer available to the public. These things have gone to the point where now you have to ask for permission. But very rarely is someone turned down. The main reason why someone would be turned down is somebody's already looking at that. But if you have a phenotype of interest, you can go and look and see what variables are available. You can go to DB Gap and it'll tell you what variables phenotypically are available and what platform they used for their GWAS data collection so that you would know how robustly covered the genome was in their GWAS study. Thank you so much, Yvette, for doing a wonderful presentation. I'll ask Kathy if you have any last comments. I don't think so. It was a great presentation of that. We really appreciate your time. Thanks, everyone. And for anybody who had additional questions, we will try and follow up with those individually. Thanks, Dr. Conley. We appreciate your time and the slides from Yvette will also be posted on the genome.gov site. Till next month, see you soon. Thank you.