 To determine the impact of the secondary pharmacogenomic findings for these patients, we decided to start this study, which was recently published in molecular genetics and genomic medicine, and really discusses the pharmacogenetic findings that we found using whole exome sequencing on diagnostic odyssey patients. Whole exome sequencing is a test that looks at all of the genes in your genome. So some of the benefits include really not having to know which gene in particular you want to look at. So we were in the early stages of starting to use exome sequencing for the care of these patients. And what we started to realize is that there's other information contained within that no one really would know about unless you're digging through it. Because you get a lot of results back when you get an exome result, right? And within those results, we were starting to find pharmacogenomic variants listed. And of course, that wasn't necessarily related to the disease that we were searching for an answer for, but it's an important result that can be used for patients' care. In looking at our cohort of patients, of the 94 patients we evaluated for pharmacogenomic findings, the majority were young, median age of 10 years at testing. The majority also had a neurologic component to their clinical presentation. Seizures was one of the most common phenotypes described in their reason for referral. And the majority of patients had at least one pharmacogenomic variant reported on their clinical west test and one-fifth of patients were taking a medication at that time that could be impacted by that result. I think if you look at our patient population, most of these patients were, I think the average age was about 10 years of age. Usually we don't worry too much about medications when it comes to this population because they're usually not taking a large number of medications. We're usually concerned about a larger number of medications in the older population. However, through this study, what I came to realize as a pharmacist is that pharmacogenomic matters in almost all age groups. Why? Because some of our patients were taking anti-depressants. Some of them were taking anti-seizure medications. Some of them were taking pain medications. Some of them were taking muscle relaxants. All these medications or some of the medications are metabolized by some of the genes that we identify in our cohort. Meaning that pharmacogenomics could, at a certain point, help these patients who are currently taking medications in our population either make changes with either the dose or change the medication to a different medication. So one of the important considerations as we were looking through the results we got back on these patients, the testing company is reporting back the specific variant that they identify in this patient, and then referring to as one of these star alleles. But to get to the actual patient's genotype, we have to then infer if no variant allele is identified that they are the wild type or star one allele. This is not reported. It's only inferred in the absence of any other genetic variation that you would detect. So if we look at these reports and know that they're only reporting on a subset of these variants, it may not be encompassing all of the actionable variants present in the population. The star alleles become very important in pharmacogenomics because it helps to predict how a patient may metabolize a specific medication based on their genotypes. So we wanted to assess how accurate we were at inferring this star one or wild type allele as it's really important to then take that genotype to establish the metabolizer phenotype that will then be used to suggest the proper medication dosing and medication used for these patients. And what's really of note is that in the African population, the reported alleles make up a far smaller proportion of the total alleles in this population. But that the non-reported alleles are significant. So for these patients, if we were reporting on the CYP2C9 gene and did not detect any variation and were to infer that they actually had the star one allele present, the wild type allele, we may be incorrect a significant proportion of the time. And that information would be used potentially to guide therapy for those patients and could be wrong. For me, operationally was troubling because I thought here we have some information that can help in this patient's care going forward either right now in terms of informing current medications, but also in the future they can take that information with them and make sure that the people that are prescribing medications for them understand their pharmacogenomic variants and hopefully can use that to provide the right dose or the right type of medication for them. As we move along with Holoxam sequencing, our ability to interpret this data is getting better and better. So as we get this genetic variation back, we're going to know better what to do with it, what it means for this patient with regard to a genetic condition they may have, with regard to pharmacogenomics, and our ability to manage this data is going to improve with it, especially as more and more patients are getting this kind of testing. We'll get it in the right location so the right physician or care provider can use this information to guide the patient's care decisions. One of the things that we really hope as an outcome from this paper is that people become aware of that and begin to build systems to be able to address these secondary findings and make sure that they're appropriately identified within the patient's medical record and appropriately communicated to the providers that will be interacting with those patients.