 All right. Thank you for the invitation to present our work. Account is a newly funded collaborative center. And I'll describe kind of the workings of it. And we came up with the acronym account because we want to account for the variation that affects drug response in African Americans, but we had to stretch a little bit to get the acronym. So it's African American Cardiovascular Pharmacode Genomics Consortium. And so we've talked a lot about pharmacogenomics affecting the tails of the distribution to people in the tails, the ones that benefit from genomically guided therapy. And I want to kind of think about that in context of minority population. So this pie graph here was published in Science. They actually updated this, I believe, in 2016, showing the number of GWAS studies done by ethnicity. And as you can see, most of them are done in European populations with about 4 percent in non-European populations. The pie piece that would be African Americans would be even smaller than that. And what are we missing when we don't look at populations outside of European ancestry populations? Well, I'm sure all of you in this room, I'm preaching to the choir with this, is that African ancestry populations have more genomic variation than any other population on Earth. Therefore, they carry variants first, more specific to their population, and at frequencies that are quite different from European populations. And I would dare say that many times they carry actionable variants that are at higher frequencies. So that tail, perhaps theoretically and sometimes practically as we have found out, is broader for African Americans. More of them will have variants that affect pharmacogenomic response than perhaps the European population, which we have based a lot of economic modeling on as well. So with that, I wanted to just talk about the goals of account. This is a U-54 collaborative brand funded through NIMHD. We are not yet fully at a year of funding yet, so we just began this. And this is a collaborative center through two major U.S. cities, Washington, D.C., and Chicago. And so at the end of this, there's a short list of people, but I couldn't list all of them. And this is a very large grant with lots of people contributing. The goal was to accelerate discovery and translation in African Americans, and we'll think about that. I'll talk about that in a moment, to build infrastructure for continuing this work, which is very limited in this account proposal, but really needs to be broader. To provide publicly available data, this was a piece that I added because it's so critical for pharmacogenomics to move forward, that we have publicly available de-identified data on minority populations. And critical in doing the things we want to do in pharmacogenomics for a minority population that has previously been mistreated by the medical community is to have engagement of the African American community, and so that was built very purposefully into this grant. So much of what we've discussed today has been in this space right here, implementing all the findings over many decades of work. However, as I pointed out, much of these findings have been in European populations. And so the genomic studies, the outcome studies, these have really not been performed in African Americans. And we don't always know even what we need to genotype to implement in these populations. So we're still stuck in this discovery phase. And this requires that all the centers here that have talked about building platforms and genotyping, they're here. And we're going to have to fit in on the back end for African Americans to benefit from precision medicine if we do not accelerate the pace of discovery and translation for this group. So the center has two major projects, and as you'll see, there's also offshoot projects that will come out of this. The first is a discovery project. And as we said, this is about implementation, but we still don't even know what we should be genotyping in African Americans for many drugs. I can only talk about a few that we'll be recruiting for, but I'd say this is probably true for many of the things we already have drug gene pairs for. This will be an outpatient study. Again, this will be done across six medical centers, three in DC, three in Chicago. And these are the kinds of data that we'll be getting. So they're different drugs, but we'll get clinical response, DNA. And in the hope of building resources for people looking at minority genomics and pharmacogenomics, we'll be getting blood mRNA to hopefully do a whole transcriptome as well as splice variants. There's a side project built into the grant for splice variant in population-specific splice variants. And we've now tried to build in an IPSC library component as well. All of these will be deposited into an African American genomics commons, and I'll describe that in kind of broad strokes in a moment, but this will be an open resource that will be de-identified data that links clinical data, DNA data, transcriptome data, all together as well as tools that may be needed specifically for admix populations such as African Americans. The drugs we started with in the discovery project were Warfarin, the new oral anticoagulation agents in clopidogrel. These will be, again, recruited from six different centers with different phenotypes depending on the drug. As you'll see, there's a translational project here that we hope that findings, if we find something new specifically, will feed into this translational project. But this translational project is not dependent on the discovery project alone. As new discoveries come in related to African American pharmacogenomics, we hope that they will also feed into the translational project. And we wanted a translational project that was flexible in its ability to deliver pharmacogenomics to the patient. And I hope to show you that. So the translational project, which is led out of the University of Chicago by Dr. Peter O'Donnell, will be an inpatient cohort of African Americans that will, there's actually two different cohorts, I've only represented one here. What the specific cohorts will have standard of care or GPS guided therapy, and I'll show you what GPS is in a moment, with clinical outcomes which will depend mainly on the size of the population we are able to recruit. But there's a second cohort that will be a general hospitalized African American cohort, which will also implement GPS in. And this will be an ability for us to see the kinds of drug gene pairs that are specific for this population. So the GPS was built out of the 1200 patients project that Dr. O'Donnell led, and this is a graphic that I probably shamelessly stole from one of his publications, where he enrolls patients, takes blood, and then there's preemptive pharmacogenomic genotyping, which is then delivered back to the physician through the GPS portal, which I'll show you in a moment. And so the physician has pharmacogenomic information at the point of prescribing, which is really the goal for pharmacogenomics. And I put CPIC here because, of course, they helped us, the guidelines that were developed through CPIC helped determine some of the actionable items that go into the GPS. I should note that currently there are 51 drug gene pairs in GPS with greater than 250 CDS reports. This is the newest for a CPIC guideline for Warfarin. And this blue area here is actually what's specific if you are African American. So there are specific variants that affect African American dosing of Warfarin, which would kind of shunt you into this decision tree. And I told you about the tales of that distribution being bigger for African Americans, so this variant here down at the bottom, which was a part of a consortium effort that I was also a part of, discovered. This variant is at a minor allele frequency of 23% in African Americans. That means 47% of African Americans carry this variant. They have something that would predict a change in their Warfarin dose. So none of these variants have been accounted for in the clinical trials we've been talking about thus far. Part of dealing with an African American or African ancestry population is that race is a social construct. There is no real biological meaning to race. African Americans are admixed between Europeans and Africans. And most of the work that has been done to identify variants that affect African Americans have done some data cleaning prior to doing the analysis. Meaning that they wanted a group of African Americans with a certain amount of Africanness when they run those GWAS studies or even the Canada gene-based studies. But that is not how people identify themselves. They do not think of themselves as a certain amount of African ancestry. So the complexity of implementing pharmacogenomics once we go outside of a European population may involve figuring out the African ancestry of African Americans. And this is just some pilot data using the GPS panel of SNPs to figure out if we were able to correctly classify people as greater than 70% African ancestry, which is kind of the non-scientific cut off for most people doing trials in African Americans or pharmacogenomics in African Americans. So this data is, if you use all 1000 genomes patients, you can see the purple dots are the African Americans. And they fall on this axis between Europeans and blue and Africans and red, the green or Asians. This is if we only use the SNPs available in the preemptive genotyping panel. As I said, they're much fewer of them than in 1000 genomes data set. But again, we're still able to classify them or group them here, the Africans, the Europeans. The yellow dots are actually the white patients in the preemptive genotyping panel and the purple are the African Americans. And using a cut off of 70%, 70% African ancestry, we had a positive predictive value of 97.8%. So we are able to reliably pick out African Americans just using the SNPs on a preemptive panel. The negative predictive value is about 56%. So there are people right at the border of 70% that we will say with our preemptive panel are not 70% or greater African ancestry, but truly are. So we're airing on the side of caution using this method. I will say for the recruitment that's currently going on in account, we are using self-identified race with the hopes of implementing this sort of more computational method. As we get going. This is a screenshot of GPS to show you the interface with the physician. Once the preemptive genotyping results are back, so there's a traffic light signal. And green means the genotyping came back, but there's no variant that would affect the drug listed here. Yellow means there's some evidence. And as you click on these, you will get a short kind of pharmacogenomics consult report. And I believe Peter says 30 seconds to read it, Peter? 30 seconds, which is I'll insult all the physicians in here. Is the attention span of an MD doing this sort of work. Then for Warfarin, he invented this very new symbol here, which is a pill bottle with a decrease, a down arrow, showing that this patient requires a lower dose than average for the drug. There are also linkouts here that will take you to more of the literature supporting these recommendations. I just want to touch on the other pieces outside of the science for account. So we have an active patient engagement advocacy core led by Dorian Millard University of Chicago. There will be community-based research through pilot grant applications that are part of this center. We have interactions with PCORI Patient Clinician Advisory Committee. We have developed our own Community and Stakeholder Advisory Board. And we've sought representation from different institutions and entities in both DC and in Chicago. And this is a really vital piece for anything you do, for anything we do in an African-American or minority community. This is our website. It has not gone live yet. But I hope that it will go live in about two or three weeks. And it's part of the implementation core. So there is an implementation aim to everything we do. And that will start with education of both patients and providers. Those resources will be available on site as well as descriptions of the projects and the results and meetings and as well as a link out to the data, the common data that we will be making available. So we're developing the educational tools that will be, hopefully, for physicians, farm dees, as well as patient educational tools. This is the African-American Data Commons. This is done in collaboration with Bob Grossman. He also is the architect for the NCI Genomic Data Commons. So he has a lot of experience in taking genomic data and using structures to make them available in usable formats for the scientific public. So this is in collaboration also with Stanford, where we're building relevant data models and elements that then will be made available to the public as the data comes through. And this is just a mockup of a previous data model they built for a different consortium. So I feel that this is a very important piece of what we do as a person who works in African-American Pharmacode Genomics finding other data sets that have genomic data and clinical data, even to validate findings that we have, is nearly impossible. So as you can see, this is a busy graphic where the two projects, Discovery and Translation, are really in the core of what a count is hoping to accomplish, speeding up the process of discovery, translating into the clinic, and then hopefully building the components for implementation. Not just, I should also say, implementation to us does not mean just academic medical centers, but also in community centers. So we've made a real effort to include community leaders or community hospital leaders in our implementation core. The consortium core, which works with community partners to help us base our pilot projects and again, PharmGKB has been a member and an advocate with us. You can see the institutions that are involved right here. And I just, for sake of completeness, at Northwestern we are not only doing a count, we also have EmergePGX. These are, you'll hear more about that from Laura. However, there are some drug gene pairs that are now going to be implemented into the EMR, as well as there's actually clinical CIP-2C-19 testing ongoing. And so one of the interesting pieces of working on a count where we have three different institutions in Chicago is they all have different ways of reporting their pharmacogenomic data and then trying to harmonize that in a way in which we can run a study like this has been challenging for sure. So with that I will end and hear many of the main players from the different institutions that are involved. And thank you very much for your attention. Thank you. And while Lynn is going up, we have time for one or two questions. Terry. Thanks, mentally. Could you comment on where your African-American data comments? Where does that sit? Is that part of the NIH data comments? Is it a separate effort? How does that work? Yes, it will be a separate effort, but we're still designing how we're going to do that. So I think in collaboration with Bob, we'll figure out where that sits. As of right now, it would be separate. So it would be important then to have appropriate pointers among the various data points to point people to it. Absolutely. And this data will, of course, be available on the regular data sites, such as dbGaP and those places. One of the advantages of building that comments is that all of the data is together in one place so that people can incorporate all the different elements at once. Great. Thank you very much. Next, Lynn Dressler, PGX Implementation Research Programs at Mission Health System.