 So the big picture is trying to figure out how we can actually do predictive genomic medicine. Of course there's lots of different kinds of predictive or precision medicine. Genomic precision or predictive medicine is one kind of precision medicine. And what we're trying to see is if we can use genomic sequence data from individuals to predict diseases that they are liable to have or to get in advance of them actually getting those diseases. So we decided to take our thousand sequences from our volunteers and search for the most severe, most obvious, most pathogenic mutations we could find in their DNA as a pilot to start this process. That is we wanted to use the most severe mutations because those should be the best for making predictions about diseases that they might have. So we went through and found hundreds and hundreds of these kinds of mutations which in itself is pretty surprising. And then filtered through them for the subset that were in genes for which we had a chance that we could see a potential disease, susceptibility in a person if we knew that there was a mutation there. So what we basically used is the entire world's literature of all diseases that are inherited in what we call a dominant pattern. That is they're transmitted from an affected parent to an affected child because when that inheritance pattern is the case, then it's one of your two genes that you have two copies of has an alteration the other one doesn't. And of course it's much easier to find mutations in genes when you only have to have it in one of the two chromosomes mathematically, statistically. It's much less common to have mutations in both copies. So we again made it easier by just selecting genes that we only have to have a mutation in one of the two copies. So we filtered through all those genes and we ended up with a hundred mutations amongst our thousand volunteers a variance that met these criteria in genes that could potentially cause a disease or a disease like trait in an individual. Then what we did is used the genomic data, the sequencing data to tell us okay in this patient check for this disease and in that patient check for the second disease and in the third patient check for that disease. So what we're doing is not looking at the patients based on what disease we think they should have based on them or their family history. We turned it completely on its head and we used the genome to tell us what we should be looking for in the patient and ask the question does the patient in fact actually have that disease or does someone in their family actually have that disease? And the result was very surprising in that about half the people in fact had that disease and many of them who had the disease weren't aware of it until we pointed it out to them. That is it was sort of cryptic or occult or in apparent disease and them or their family members and they were quite surprised by us going to them and saying we think you might have this disease and in fact it turned out to be the case. Some of them are potentially quite serious. Some of them are cancer susceptibility gene mutations and some of those were in families where there was in fact a family history of cancer and the family was well aware of it. Others though were in families where there wasn't much of a family history of cancer and it brings up an interesting paradox in medicine right now which is our current way of finding families at risk of such diseases is to wait until multiple people in a family have suffered or died from some cancer. That's how we identify them currently and then we do a gene test. What our approach does is turn that completely on its head and say we're not going to wait until you have people who are suffering or dying from this disease. We are going to find you from the genome and find you and identify you before that happens and in fact we did exactly that by finding individuals who have cancer susceptibility checking their families and in one particular case it actually went so far as a person having preventive surgery and in fact a cancer was found in that person and they had no idea that it was even there at all. So it proves the notion that you can take an individual, sequence them, search their genome for severe mutations in disease genes then you can implement predictive precision medicine and prevent a disease from occurring and causing suffering or potentially even death. So it's a pilot that actually demonstrates that you can really do precision medicine from sequencing. So this is in fact a pilot and it's a small set of mutations and a small set of people so what we have to do is see if we can actually scale this notion. Can we do more genes? Can we do this with milder mutations? Can we do more mutations that might not be so easy to find or predict and see what our predicted power is to see if we can expand it? Our overall yield in the study of a thousand people was that we found 30, between 30 and 40 people who have a disease susceptibility predicted from their genome which is a lot, that's three or four percent and that's a subset of mutations and it's a subset of genes. So we know there's more out there. What are those mutations? What genes are they in? How can we find them and how well can we predict what diseases people might be at risk for? And it's not perfect and we know that about half of them had a disease and half didn't and actually the half that didn't are as interesting as the half that did because the half that didn't they're telling us something interesting about that disease and that gene also which is that you can have such a mutation and not have disease. We need to learn from those people why that's true because that's telling us something about disease resistance as well. One of the main challenges is to begin to change the culture of medical practice and we don't want to change or discard the notion of the physician's primary duty being to take care of a patient who comes in when they're sick, make a diagnosis and treat a disease. If that occurs every day it is the right thing to do and it will be the main way that disease is managed by physicians. But we want to add a new tool to the physician's armamentarium which is to say it's fine to take care of patients once they're sick and that's appropriate but wouldn't it be better if we could find some patients before they got sick and take care of them so they didn't get sick? And adding that to the armamentarium requires us to change how the physician thinks a little bit and it isn't just that their practice is limited to people who come in sick but their practice can also include patients who come in apparently healthy but actually have a susceptibility. The genome tells them what is wrong with that patient and they can use that to make the patient healthier or prevent them from getting sick in the first place. But changing that mindset in medical practice will take some time and people will have to get used to that notion and will have to train physicians on how to do that responsibly and properly.