 All right, thank you so much for having me. Before I get started, I'll just mention that both of these projects are very, very large. And to talk about both of them in the time allotted, it's gonna be very challenging. So realize that both of these projects go much deeper and much broader than what I'm able to touch on today. And if there's anything else you'd know about, I'm happy to take any questions later. All right, so one common question that both ClinGen staff and ClinVar staff hear a lot is what is the difference between ClinGen and ClinVar? And so hopefully through our time together today can help you understand the differences but also understand the ways in which they work together and are complementary. So ClinGen and ClinVar work together again to provide a complementary resource to support genomic interpretation. ClinVar is a database that's funded by intramural NIH funding and maintained by the NCBI. And its goal is to be a public archive of reports of the relationships between variants and conditions. ClinGen, on the other hand, is a program which is funded by NHGRI. And its goal is to identify clinically relevant genes and variants for use in precision medicine and research. And I added the NE in brackets to the ClinVar goal there to kinda help you see this subtle distinction between the two. ClinVar is really an archive and they can and should accept information about any kind of variants from anywhere. But what ClinGen does with its focus on clinical relevance is put a layer of curation on top of information that might be available through ClinVar. So I'll start with ClinVar, what is it? It's a public archive of variant phenotype assertions submitted from a variety of sources which include clinical labs, research projects, expert panels, and other databases. And it's different from other NCBI resources such as DBSNP or DBVAR. Those two that I just mentioned, they primarily catalog information about locations of variants and just the fact that they exist. The smaller variants in DBSNP and the larger copy number variants in DBVAR. And the way ClinVar is different from those two is it catalogs the existence of the variant but also adds on to it an assertion of what that variant means. This variant is pathogenic for this condition. This variant is uncertain in this particular setting, that kind of thing, that is not to be found in DBSNP or DBVAR. And I will say that in my personal opinion, I do think that ClinVar is one of the most important resources available to people that are doing genomic variant interpretation. If you just think about the context in which ClinVar was developed, back several years ago, many clinical laboratories were kind of niche laboratories working in a certain space, looking at a certain set of genes, maybe in the same disease area. And as such, they got to develop a really good wealth of experience on this particular disease area or this particular gene. And so when they did see novel variation in those genes, they were better able to deal with it because of their experience. But as next generation sequencing technologies came about and whole exome, whole genome came into play, many labs were called on to be able to interpret variants in any gene at any time, regardless of their experience in that gene, or the information out in the literature available on that gene. So when you're faced with a situation like that, often what you have to do is see, well, has anybody else ever seen this variant before? And if they did, what did they say about it? And in the past, all you were able to do is, you could look in the literature, but as we heard this morning, not everything of importance makes it to the literature. You could also phone a friend and maybe ask, hey, have you ever seen this variant? What did you think about it? But if you do that, kind of the buck stops there. You know that information, your friends know that information, but nobody else gets to see that information. So when ClinVar came along, it really kind of changed the game because now we have a place where people can do this. They can quickly and easily access information about who has seen a variant, what did they think about it? Without having to rely on kind of just side channels to get that information. And I think there have been databases before ClinVar and there won't be databases after, but I think one of the most unique things about it is that it is free, it's publicly available, and it accepts submissions from everyone, so it really creates a unique resource. So what does it do? It facilitates the evaluation of variant phenotype assertions by archiving the submitted interpretations of gene disease relationships, aggregating data for multiple submitters, and determining if there's a consensus about the interpretation. And the most important thing that it does not do is interpret variants itself. So a lot of times, you know, we hear people talking, they say, well, ClinVar said the variant is this, or ClinVar said the variant is that, ClinVar said nothing. ClinVar is showing you what the submitters all provided them about that particular variant. So what's currently in there? Well, as of August 2018, there are over 700,000 variants from over 1,000 different submitters, and over 430,000 of these variants are unique and actually have clinical interpretations associated with them. This also includes about 10,000 variants coming from expert panels, and we'll talk about those in a bit more detail when we get to ClinGen. But there are also about 18,000 variants in there that actually have conflicting interpretations associated with that. And I point that out to you, not necessarily as a bad thing. Some people might look at that and say, oh, you know, that is a marker of poor quality. There's variants in ClinVar that, you know, different labs don't agree on. I point that out as a potential positive. This is just simply highlighting for us a problem that has always existed. Laboratories have always interpreted certain variants differently, but we just didn't realize it until people started routinely putting their information into a publicly available database. And this is really an opportunity for us to see that problem and to work on addressing that. And that's one of the ways ClinGen works with ClinVar is to try to encourage these submitters to resolve these discrepancies. So one thing that's important to keep in mind about ClinVar is that it's a submitter driven resource. There are many, many pieces of information that ClinVar can accept on a variant, but if a submitter doesn't provide them, then they're not there. So, you know, one complaint that we often hear is that people say, oh, you know, I see a lot of variants, but they have no supporting information available. That is really on the submitter that is not necessarily a bad thing on ClinVar itself. The quality of submissions absolutely vary. We have people that submit very comprehensive interpretations with all their supporting evidence in a very nice summary. And then we have people who submit the bare minimum information. And I would argue that these all have a place. These are all at least somewhat useful in that you at least know that a variant has been seen before, but of course the ones with the more comprehensive information are gonna be the most helpful in your practice. So when you're assessing the information you find in ClinVar, you really need to look at the quality of the submitter and the submission itself. And within ClinVar they do have some marks to help you look at the quality of a particular submission and one of which is their star status. So you may, if you are a regular user of ClinVar, you have noticed that each submission has a star attached to it. In general, one mark of quality is its review level, which is what these stars kind of stand for. You can see that the one star level is usually a submitter that has provided the criteria they use to actually determine if a variant is pathogenic or benign or what have you. So they actually say, all right, we need to see the following piece of evidence to call pathogenic. We need to see the following information to call something benign. So that kind of gives you a sense of how they actually looked at particular variants. That's not to say that variants that have no star associated with them are ones that you shouldn't use. It's just to say that you should maybe take those with a grain of salt because they haven't provided that transparency that we would like people to. There are other review levels. As you can see here, two stars if submissions have multiple concordant assertions, three stars if it has been reviewed by an expert panel, and four stars if the particular variant has been subject to practice guidelines. But kind of beyond the star level, users really need to look for other measures of qualities themselves. That's really the one measure that ClinVar puts forth, but could we maybe put forth other metrics? So this is one of the ways ClinVar comes in and really partners with ClinVar in this. As an archive, ClinVar really shouldn't be the one that says, okay, we're only gonna accept submissions from you and not you. We don't like your work, this kind of thing, especially if you really wanna encourage people to continue to submit information to you. But ClinGen as a separate entity who is really focused on clinical relevance can do something like that, and so we have. So here this slide is just showing you some criteria that we have put forth as a measure of quality. And you can see here that a number of laboratories have actually worked to uphold those standards, including some of the most proficient ClinVar submitters to date. So now that I introduced ClinGen a little bit, I'll step back and kinda walk you through exactly what we are. So again, the clinical genome resource, or ClinGen, is an NIH-funded effort looking to identify clinically relevant genes and variants for use in precision medicine and research. And in a very small nutshell, the way that we do this is that we encourage our different stakeholder groups, which includes patients, clinicians, laboratories, and researchers to share their genetic and health data. And once this data is publicly available through resources such as ClinVar, we're able to use it to answer a number of critical curation questions, such as which genes are actually associated with disease, which variants are actually causing those diseases, and is any of this information medically actionable? Once we have answers to those questions, we put it back into the public domain, either through ClinVar itself or through our own website, clinicalgenome.org, trying to build a genomic knowledge base that anyone can access. And hopefully, in so doing, we're able to improve patient care through genomic medicine. We have been funded since 2012, and in those years, we've been developing our various processes and the infrastructure to support those things. And now we're really at a place where we can put some of these results out into the public. And so what I'm gonna spend the next few minutes doing is just quickly walking you through each of our curation activities so you can kind of get a sense of what they are and how you might use them in clinical practice. So since we were just talking about variants in relation to ClinVar, I'll start with the variant pathogenicity efforts. So what we're trying to figure out here is which variants in gene actually cause disease? And there are several different efforts going on in this space. We're trying to attack this problem from two different fronts. The first is how can we address the existing classification differences? So this 18,000 number that I showed you earlier, how do we get at those interpretations that are already out there that we know disagree and see if we can make them agree? And we do this in both the sequence variant community and the copy number variant community, which I'll tell you about in a moment. And then the other way that we address this problem is to try to prevent future classification discrepancies by making general and quantitative specifications of the current ACMG AMP sequence variant guidelines and also encouraging groups to come up with disease or gene-specific modifications to those guidelines. When groups do that, they actually apply for expert panel status in ClinVar and that is how we get these three star expert panel records that we mentioned earlier. So in terms of addressing those existing discrepancies, again, as I mentioned, we have efforts both for sequence and copy number variants. Both of these efforts just finished pilot phases within the sequence variant community amongst the participating laboratories, they were able to resolve 87.2% of discordant sequence variant classifications, again, between those labs that participated. And within the copy number variant community, we took a slightly different approach in that we wanted to first look at any copy number variants that overlapped known dosage sensitive genes, but weren't originally interpreted as pathogenic or likely pathogenic. And so doing, we were able to update classifications for 63.8% of those CMVs that we evaluated as part of the first pilot effort. So these things are underway, they are ongoing, they will take time, but this is helping to bring that 18,000 number down. Again, the other way that we are trying to address variant discrepancy issues is trying to prevent them from happening in the first place. So we do have a group called the sequence variant interpretation working group that's working to further specify the ACMG AMP guidelines for sequence variant interpretation. But then we also concurrently have our domain specific groups. So for example, cardiovascular neurodevelopmental disorders, hereditary cancers, et cetera, working within those spaces to say, are there specific modifications to these rules that we would put forth in these particular settings? So we've talked a lot about variants so far, but I would point out that those are not the only things you need in order to be able to do genomic interpretation. I think curation activities related to genes and diseases themselves are also important. So to that end, we will move on to our gene disease validity process. So the question that we're trying to answer here is does this gene, when significantly altered, cause this disease? And this is, I would say kind of the first step in the process, because before you can even start to interpret a variant, you have to be sure that the gene you're in is actually one that's related to the particular disease. So what we've done is defined the criteria needed to assess this, some genetic evidence and gene level experimental evidence. We described the strength of the evidence that supports a gene disease relationship in a semi-quantitative manner. And this system allows users to methodically classify the validity of a given gene disease pair. And here I won't spend too much time on this, but this is just to kind of give you a sense of the categories that we use to describe this evidence. We have categories that describe when there's positive evidence, and we also have categories to describe when there's maybe evidence to refute a given gene disease relationship. How might, what might you use that for? What good is that? Well, I think one immediate application of this process is for laboratories and test design. So a laboratory might decide when they're trying to design a multi-gene panel to only use genes with perhaps moderate and higher levels of evidence. One might argue that these genes with limited or lower levels of evidence we don't know enough information on and variants in those genes might more often be classified as variant of uncertain significance, which is difficult for both clinicians and patients. So that might be one potential application. For clinicians, this might actually help them when they are getting results back. So, you know, say they get a variant back in a gene of uncertain significance and they may be faced with the question of how aggressively should I follow up on this? Should I pursue additional testing for family members, things like this? They might, you know, choose to do more if the gene has more evidence or emerging evidence and may choose not to act as aggressively if the gene has less evidence. Another curation activity that ClinGen is involved in is dosage sensitivity. And the question that we're trying to address here is, is a gene or genomic region dosage sensitive or haploinsufficient triplosensitive? And really this process was created as a resource to help in the interpretation of copy number variants as they are identified on the chromosomal microarray. And we have developed an evidence-based process to assess genes and regions for both haploinsufficiency and triplosensitivity. And our goal is to create a genome-wide dosage map. Again, similar to what I just showed you for gene curation, we also have categories describing the strength of evidence for this piece of information. And how might you use this? Well, if you are looking at a copy number variant that you are unsure of, you might put the coordinates of that copy number variant into our dosage map and see if any of those are dosage sensitive. And I see that my time is just about up. Very quickly, we also have actionability curation, which is looking to see which genes confer a high risk of serious disease that could be prevented or mitigated if the risk was known. And the final slide is really just to say that neither ClinGen or ClinVar could act in a vacuum. We really need both pieces. Without that shared data that's available in ClinVar, ClinGen wouldn't be able to do any of its curation work. And without some of the support that ClinGen provides, the quality of ClinVar might go unchecked. So I think both things are important. Both things work together. And thank you. Any questions? Otherwise, we'll try to hold off comments until after the next presentation. All right, thank you very much. Next we have Kristen Weitzel presenting from Ignite.