 So welcome this morning and thanks for coming out on such a beautiful day to Stanford and participating in our launch week here. My name is O'Toole Butte, I'm an associate professor and a chief of a new division called Systems Medicine and I'm gonna spend the next 30 minutes just explaining why this is the most amazing time to be in research in biomedicine and some of the more exciting things that we have going on in my lab. So if you haven't heard, if you haven't heard, we're in the middle of a data revolution today, a data revolution. The Economist magazine now says that we as a human species generate two zettabytes per year. If you don't know what zeta means, it goes kilo, mega, giga, terra, we just learned what a terabyte was, peda, exa, zeta, bytes. And of course next year we'll generate four zettabytes. Of course we're doubling this every two years every 18 months or so. In fact, we generate so much data now. We generate so much data in life science research and it feels like medicine and biology and botany and geology. We generate so much data that actually Chris Anderson at Wired Magazine predicts that science itself is obsolete. Okay, now why is that? Well, how can you say something like that? Why do I even agree with this? Because with the scientific method we ask a question and then we go find data to answer that question. But now what happens when in all these fields we have so much data, the new magic is in figuring out what's the question we wanna ask, given this data, right? We have so much data collected from mobile devices to how volcanoes erupt and of course in biomedicine as well. And the data is just there. It's not just there, it's actually on the internet and actually the kids and the students can get to this. In fact, I'm so excited by the kids and what they can do. This is one of my favorite quotes from the Harvard Business Review about two years ago. Think about the kids for a second, right? Because we're an educational institution for a moment here. What happens when a 12 year old can gather information faster, process it more efficiently, reference more diverse professionals and can get volunteer guidance from better sources than you can at work. How can you still be competitive with a 12 year old kid today, right? That's the world today. That's the world that they're growing up in. A kid today is used to free point to point 24 hour video conferencing with anyone in the world, right? That's what Skype is, right? And think about how hard it was to make a long distance call 20 years ago. Forget about international. And this is how fast things move. We know this, we know this. It's happening in life science research too. The Harvard Business Review again more recently says the next big scientific revolution. We went from the idea of experimentation to theory in the 1700s. And we went to simulation in the 1900s. The next big scientific revolution is in data-driven science. In fact, nature, one of our prestigious journals says it is shameful if we don't make use of the data that's out there. Science says we need to make data maximally available. And the Lancet says it is an impediment to public health if we don't figure out what we're doing with this research data. Where does all this data come from? The data sounds theoretical. Let me kind of make it real for you for a second. A lot of my data comes from devices like this. So in my hands here is something called an atherometrics gene chip. And I'll keep these around if you wanna come up at the end. The size of my thumbnail, here's my thumbnail. And this one chip lets me quantitate or measure the levels of every gene in the genome. 20,000 genes. I put a sample of cancer, different disease. I can measure how many of each gene is turned on and off on that sample. Now, in life science research, we love to scale. So once we get one of these to work, we quickly make 96 wells of them. Now, each one of these little squares is one of those chips. Now, and you can see if you couldn't see it, I'll blow it up here. Kind of amazing that we have a chip like this that measures every gene in the genome. It's actually kind of amazing. We've had this chip for 15 years now. Kind of amazing that this is actually a commodity item. Now folks at atherometrics don't like me saying this, but most researchers don't even care which company they're ordering from anymore. And it's amazing, it's a couple hundred dollars for these chips. What's absolutely astounding today is that we make scientists share their data on the internet. Now, when I started in this field, when I started in this field about, let's see if I can fix the mic here a little bit, when I started in this field about 15 years ago, we were working on the first 60 samples, the first 60 samples of cancer that we were measured with these chips. The data is so plentiful now. It's on the internet that we're just a few weeks away from having one million of these chips publicly available. Okay, let me kind of make this real. I can kind of take that for granted as a scientist, right? Here's a website, okay? This is run by the National Institutes of Health. And you can see here on the right there, 716,801 samples, it's about last week's number. And here's the European version of this, another 200,000 samples, where about 920,000 samples like these micros publicly available. And the growth rate of these repositories is just finally slowing down to Moore's Law, okay? Moore's Law, of course, Gordon Moore, founder of Intel, predicted in the 1960s a computational power will double every 18 months. Yeah, now we're only doubling every 18 months, only because the next repositories are growing like crazy, okay? And let's just make this real for a second, okay? And let's make this really fun. In the middle of that window there, you can see a simple search box, a high school kid today that needs to do a science fair project. She can type breast cancer and now find and download 31,000 samples of breast cancer as easily as she could find a song on iTunes today. 31,000 samples of breast cancer, type breast cancer on that little website there and you too can download 31,000. You know, 30,000 samples of breast cancer and there's more breast cancer samples than any breast cancer researcher will ever have in his or her lab. Because by definition, we're gonna have all of your samples and all of your competitors. Because you have to share your data. If you wanna publish a paper, you wanna get an NIH grant, you have to share the data today. And can I tell you how amazing it is that nobody actually even thinks about the data that we've already put out there? Because we're also really focused on getting yet another measurement in this field. Those of you who are kind of skeptical about the kids, these are the five kids my lab has placed in either the Intel, Westinghouse or Siemens, semi-finalists or finalists pools just in the past three years. These are all Pali kids. And they were from Pali to Central Valley. The kids say, so that means each one of these kids was in their year one of the top 300 science kids in the country. This is what we're doing with your either kids or grandchildren here in this area. Kids today can absolutely do this. A kid today can run magic with public samples like this. And if it's not breast cancer, it's colon cancer, it's prostate cancer, it's all the rest. Now, as a computational person, as a person knows data, I know that we have these repositories and we start to use them. We start to get ideas. Wow, this drug might work on this disease or this is a great blood test for this particular kind of cancer. Maybe we should use that. And so what do we do? We find our collaborators. Danford's actually a pretty small medical school when you think about the numbers. So we really know what everyone else is doing here, okay? And so, oh, we got a cancer prediction. Let's go find a cancer doc or we have a GI prediction. I'll show you some GI predictions here. But what happens, how do you scale from there? What happens if you don't have a collaborator on campus? Or what happens when you want to think about treating some of the rare diseases? I'm a pediatrician, I'm used to rare diseases too. So this is amazing. What I'm going to show you in the next few slides, if you're used to biology, the next few slides might be absolutely mind blowing for you. The other day we had a chemical that we thought would influence how people get diabetes, okay? And it's environmental chemical. And we said, you know, we need to test this and we're not going to do a human study. We need to test this in a mouse model for diabetes. And so we looked around on campus and no one had exactly the right model. But what does an informatician, a guy like me, do? We just Google for it and we find a website like this. AssayDepot.com, by the way, yet another Palo Alto startup company, okay? AssayDepot.com, think about Home Depot, but for biomedical assays, okay? Click on pharmacology. What type of mouse do you want to run today? Bone, cardiovascular, dermatology, diabetes, gender-neurinary problems, infectious disease, inflammation, cancer, eye problems, ear problems. Let's drill down a little bit for a second. Here's diabetes. This is called the OB-OB diabetes model. This mouse has been eating a lot and getting diabetes since the 1970s, okay? This is the model people use for diabetes. And here's a 16-mouse study and you can follow these mice for 16, these 16 mice for about 28 days. They'll do the fasting blood sugar, the insulin tolerance test, and the glucose tolerance test, the stuff we're used for diabetes. You can divide them up into two groups of eight or four groups of four, test whatever you want and these mice, they're going to do this entire experiment for you. Now, what am I covering up the price? $9,000 for this service, nine-week turnaround time, some of you are already chuckling, and literally add to shopping cart. An entire mouse experiment is purchasable with a credit card today. This isn't the hard part anymore, right? This isn't the hard part anymore. It's commoditized, right? It's absolutely commoditized. My credit card doesn't have very high of a limit. Now, some of, you know, but I have a stand for when they give you a little bit more there. Now, some of you have noticed this particular experiment's done in China and you can imagine we can outsource. Outsource sometimes means offshoring or not. For those that are really picky here are actually when you click out at the shopping cart, you get presented with this list of 133 vendors that want to buy your business here, that want to get your business. Summer in Wisconsin, Maryland, they could be all over the world. If you really need to have it FDA approved, you click on the right down the left there. If you need GMP facilities, you just click, it's like easier than Amazon, right? Amazon, you kind of lose that grid on the left as you drill down a little bit. This is just, I need it to do this experiment. Now we need FDA approval, I need USDA approval, and all of a sudden you get down to the vendors. We're literally able to do this for you. This is not the hard part anymore in biomedicine, actually, when you think about it. So now what I'm painting this picture is we have a million publicly available experiments, and if we just have a little bit of creativity to come up with an idea, we can have a credit cardable mouse experiment to test the drug or the diagnostic. All we need are smart people in the middle to kind of turn this crank. And let's take this to an extreme for a second. We're so used at Stanford to having our students start computer companies, IT companies, internet companies in the dorm rooms and the garages. College students start things like Google, Yahoo, Facebook. We have Microsoft started in dorm room, right? What's it gonna take for our students to start the next gen and tech in their dorm room? How much more do we need here? A kid in a dorm room has a million microarrays, any mouse model in the world, purchasable by credit card, all they need is a creativity here. What gets me excited is, yeah, it's kind of neat when students start yet another photo sharing app, okay? I want them to start the next gen and tech or the Amgen or even the Merck or Pfizer in their dorm room today. I think this is where we're heading, the publicly available data. This is why we teach the kids to do this today. So let's put it all together here, okay? Let me show you what we've been doing with this kind of data here. My hero for the last few years has been this guy, Linnaeus. How many of you remember, still remember because you all had to remember at one point, kingdom phylum class order family genus species. Remember that? Everyone in the room had this on a test. I don't care what school, what, where in the world you went to. You all had mnemonics for that, right? King, Philip, whatever. Some were clean, some were dirty. You don't have to repeat them here. Everyone had to know kingdom phylum class order family genus species. It's on a test. The high school kids today still have to know that. Linnaeus in the 1700s is known. It's actually just celebrated the 300th anniversary of his birth. He's known for putting the species into a taxonomical tree. And we still use that taxonomical tree. What everyone's forgotten about Linnaeus, what everyone's forgotten about Linnaeus is that Linnaeus was also the first guy to put diseases into a taxonomical tree as well. He published this book called the Generomorborum, the 1700s, and the word for this was coined in the 1700s. The word for this is called nozology. Nozology, it's not about noses. Nozology is the systematic classification of diseases. And in the Generomorborum, you can see in the original Latin, if you can read it, the 10 classes that he thought the 300 diseases of the world should be classified under the first ones, fever with skin problems. Then he got fever with urinary problems, fever with pain. I mean, this is the best they could do in the 1700s, just to put it in perspective here. This is about 80, 90 years after Galileo was under house arrest. We're saying the sun is the center of socialism. It's not bad progress in 80 or 90 years. But nozology is still with us. It's not this bad. We don't remember Linnaeus for this because this is horribly wrong if you haven't figured that out. I mean, he got the species so correct. Let's just remember the guy for this, right? We forget this all the time, but we have nozology today, okay? Now, we at Stanford have computerized systems when you go see a doctor, right? But some of you might still remember going to see a doctor in the old days that had a piece of paper. And they check out these boxes and say, please hand this back to the receptionist. Remember that once in a while, you would dare to peek on the back to see what is that? And there's a grid of numbers there, right? And these terms, those are called ICD-9 codes, the international classification of the disease. 9th edition kind of already sounds old, okay? Actually, even this week, we're still arguing. Do we stick around with ICD-9? Or do we move to ICD-10, the 10th edition? Even that's already obsolete out the door, should we wait for ICD-11, the 11th edition? We're still arguing about nozology, believe it or not. But you know what? It's lost all connection to science. Nozology today is just to make sure we get paid the right amount here with insurance companies, right? That's about billing. So we said, you know, let's think about the science aspects for a second. Now, this was in the 1700s. Let me just show you what ICD-2 looked like, okay? ICD-2, the second edition. Now, just for the historians in the room, this was never a United Nations standard, okay? Because remember, this predates the United Nations. This is 100 years ago. This became a League of Nations standard. Anyone remember the League of Nations, right? This became one of those standards here, right? And so 100 years ago, this is 1909. 100 years ago, the book is written the same way our current books are written. Well, Code 39 is cancer of the bucal cavity, cancer of the cheek. Then you got cancer of the stomach and the liver and the intestines, cancer of the skin. And then this book does what we still do today in our code books. We're running out of numbers enough with the cancers. Put all the rest under 45. Including lung cancer and brain cancer, okay? What they're saying 100 years ago is that it's not even worth distinguishing lung cancer from brain cancer. Lung cancer is the number one cancer killer today. 100 years ago was cancer not otherwise specified. But that's funny, but that's not as funny as what my favorite part of this book is. This international standard literally has to say, if you run into a patient that died because of a visitation from God, be sure to use code 189 here. This is the international standard. Now, the same way you're laughing at this, imagine them laughing at us. 50, 20, 10 years from now, right? Those silly fools didn't know this bowel disease and this joint disease were the same thing, right? Even though they had different kinds of doctors taking care of them, it doesn't matter, right? Because underneath the molecular path of physiology, the stuff that we study, the science we do at Stanford, that's how we figure out the connections between these diseases. We don't put them in silos, we gotta figure out what's the common principles here. So, we take the world's data and we do this. We come up with grids like this. You can't read this, but one of my first big grants at Stanford was just go scour the internet for every disease that's been studied with these kinds of chips. And you can see a whole bunch of diseases up the columns and you can see all the different genes on the rows. We can start to paint kind of tapestries like this, which genes go up, which genes go down. In fact, we can start to see patterns. This disease leprosy is just like the next disease over or this disease obesity, it's like this cancer. We start to get patterns like this. We can put those patterns into a tree and you can't read this, but I'm just gonna highlight some of the interesting parts here. You see at the bottom here, I'll point it out just with my hands here, the tree works, the branches that are similar towards the bottom here. And at the bottom here at the left, on the right here, from your perspective, is colon cancer and colon polyps. And what I'm saying is that at the molecular level, it looks really similar between colon cancer and colon polyps. And many of you seem like you're over age 50, just a handful, I'd say. And several of you over age 50, you probably hopefully had that preventive colonoscopy. If they saw any polyps there, they're gonna buy it soon because colon cancer is a major disease that we need to try to prevent. If we get it early enough, we at Stanford and elsewhere can do something about it. So it's kind of a no-brainer at colon cancer and colon polyps look like each other. Now over on the left here, we have cervical cancer and we actually have four worsening stages of cervical cancer. Why is that interesting? Because as the cervical cancer gets worse, it actually starts to resemble this incredibly rare disease called apocet. Apocet stands for something. I'm not gonna explain what it stands for. It's a rare disease when kids are born having autoimmune problems. Autoimmune problems are things like lupus and inflammatory bowel disease, but these kids can't train their T cells to recognize themselves and because of that, they start to reject their own organs. It's not a great disease to have. They need support for their glands and hormones and things like that. Why is this interesting? Because just by looking at the diseases, this cancer, cervical cancer, is more similar to this rare, incredibly rare pediatrics disease than this other cancer, colon cancer. Why did we ever think all the cancers look like each other? Just because the same kind of doctor takes care of them. At a molecular level, you realize that the disease is gonna sort every which way and I think this is what we do at Stanford. We don't just try to put diseases into silos like that. Think about patients from the whole perspective. We think about the science from a whole perspective like this. We don't just say, this is the cancer's gotta be studied this way. I'm a pediatrician, I'm an informatics guy, but I also belong to the cancer center here. This is what we do here at Stanford. We think about it in a much broader way than this. Now, instead of making just pretty pictures, we start to realize, boy, you know what? A lot of times we get a disease over here where we have a lot of therapies next to a disease over here that we don't have any good therapies for. So why don't I try some of these drugs on this side on the diseases on this side? Ah, even before I put it into a patient. I just showed you how with a credit card, you can get all these mouse models done here, right? Why even try it on a patient? Let's just see if it works in the mouse models here. So this is what we did. And so the first one of these predictions we had was actually the seizure drug to pyramid, topomax, maybe you've heard of it. It's a generic mouse, it's pretty cheap, but predicted the seizure drug should work on inflammatory bowel disease, which is down here. So seizure drug up here, your inflammatory bowel disease down there. And you can see some pictures here. You could, we make these rats get Crohn's disease by giving them this chemical, TNBS. Give them steroids, which we know works on Crohn's disease. It's not a great therapy. And then to pyramid is this seizure drug and you can see it works about as well as the seizure, as steroids. Now, let me show you to it this way. One of the nice things about being able to choose where you get your experiments done is now this is actually a rat colonoscopy. Okay? Some of you are chuckling. And I know the ones who are chuckling are the ones over age 50 who know how hard it must be to do a colonoscopy in a rat. The rat on the left, you could see it has a pretty clean bowel, except for the brown stuff, I won't explain what that is. We're a little bit before lunch. You could see the red scarring and inflammation in the rat in the middle. That rat has the Crohn's disease. And the one on the right has the same chemical but now has the seizure drug to make the Crohn's disease get better. So I'm not just cherry picking. Pictures for you can see the colonoscopy here. Can I tell you, this is the kind of thing you wanna buy with a credit card, okay? So this is exciting to me because I think the future for us is in finding new uses for drugs. We're gonna come up with new drugs. We're great at doing that at Stanford, certainly in the Bay Area, in the peninsula. We know how to come up with those and it's getting harder, but we're struggling, we'll get to them. But I think there are other uses for drugs out there. If there's a drug that's safe and has other uses, if we don't look for these, I don't think we're gonna find them. I don't think anyone's gonna find them. So we gotta make an effort and try to find these new uses for drugs out there. Now, let me switch gears a little bit and talk about what's also really exciting to me. And some of you know about this revolution that we have in terms of being able to look at one's DNA and one's genome. Now, even before you read the slide, I'm gonna explain to you what is your genome? What is this genome, right? Some of you might have had pieces of your genome sequence by companies like 23andMe. They've been on Oprah. They've got 150,000 people who've, and that's also Stanford students who are founders of the company, Stanford medical student alumnus who are founders of the company there. And so what is your genome, okay? Your genome has six billion base pairs. Three billion came from your mom, three billion came from that, billion, billion. That's too hard to remember. Here's a simple way to think about it. Your genome has as many letters as, do you know the joy of cooking by Irma Bombeck? The joy of cooking. I'm gonna stay away from religious books. The joy of cooking. Imagine 4,000 copies of the joy of cooking and that is your genome. Now we have a bunch of restaurants. So these restaurants are doing well and these restaurants are doing poorly and some part of why they're doing poorly may be because of their recipe books, okay? Maybe some of the words are misspelled. Maybe some of the pages are missing. Of course, it could be other things. It could have bad ingredients, but we think the same case in control. If we have patients that do well and patients that do poorly, we think some element of that has to do with your DNA and the recipes and all your cells. That's what the joy of cooking is. We cook omelets and French toast and the cells use the DNA to cook up all the different proteins we need for life. Original genome project finished about 2003. That took about 13 years, cost $3 billion. Remember this $3 billion because you can see the price curve in a moment. Where's excitement now? I'm just gonna mention a few companies that really formed out of Stanford, Helicos. Steve Quake, who's a Stanford engineering and medical school faculty in bioengineering started a company called Helicos. They've been charging on the order of tens of thousands of dollars for their genome sequences and already they're gonna be obsolete pretty quickly. Pac-Bio, which is over in Mendel Park, they're on track to sequencing the entire human genome in 15 minutes. Prompting Discover Magazine to publish this article a couple of years ago on the Jiffy Lube of the genome, okay? Those of you who drive know that you need to change your oil once in a while and Jiffy Lube promises to do that in 30 minutes and now we're on track to doing this in less time and probably cheaper. Complete genomics down in Mountain View. They're on track to doing 80 genomes per day. That's if you do the math one every 17 minutes if they're running that in day. And Ion Torrent, which is now this up in Foster City, Illumina down in San Diego. The current price leader is about $1,000 for your genome and these two are just competing like crazy, getting the price down lower than that. This is the price curve that everyone really shows. In fact, it's dropping so fast. Even the Genome Institute at NIH can't keep this up to date fast enough. As of July 11, you can see what the current price is per genome. If we were following Moore's law, we would be charging about $5 million per genome today. And already we're down to a thousand, which is the bottom line here. Again, they can't even keep this up to date fast enough here. And just think and imagine for a moment. All the revolutions launched by this curve, the internet, personal computer, iPhones, Facebook, Google, you just have to imagine what are all the revolutions that are going to be launched by this curve. And obviously, a big part of that is biomedicine. What am I going to be at risk for? Who is my family? All of these questions are askable and answerable now with this kind of technology curve. Now, of course, if you just linearly extrapolate, even a thousand dollars seems like a lot. The president of complete genomics said the other day in his blog post, yeah, we're heading towards $33 per genome by the end of this decade. Right, I'm willing to bet more people in this audience are going to pay more than this for brunch today, okay? $33, why? Hell, we'll just make it free. Hell, it'll just be negative cost, right? Some companies, some hospitals will pay you to get your genome so they can figure out what to do with you, right? Yeah, they're car insurance companies that make you put a GPS if you want to reduce your car insurance rates to make sure you're driving well. You can imagine something similar. The price is irrelevant now. The magic is figuring out what to do with this. So in the last few minutes, I'm just gonna present our first patient presenting with a genome. Steve Quake, obviously he started Helicos in 2009. He paid $35,000 to his own company to get his genome sequence. And he published it. He put his whole genome sequence out on the internet back in 2009. But what was left unanswered when he published his paper was what's the medical use of having your genome? What's the medical relevance for this? So a team of us, you and Ashley in cardiac genetics, myself and my team, Russell Altman in pharmacogenomics. He gave some talks earlier this week as some of you might have seen. Steve Quake, of course, himself, Hank Greeley from the ethics side, all came together to work on this because Steve Quake presented as a patient. What does that mean? Well, I'm a doc, let me present a patient to you. Here is my patient I'm presenting to you. Let's pretend this is Grand Rounds for a second. He's a 40 year old male who's presenting in good health to his doctor. He exercises regularly, takes no medications. Why is he coming to the doctor? Because Steve Quake here, his nephew here, his nephew at 19 years of age, died in his sleep, sudden death. Just didn't wake up one morning. Now Steve Quake goes to his cardiac geneticist. You and Ashley says, doctor, doc, am I at risk for sudden death? Here is my genome, okay? He becomes a patient. Patient has a heart rate, blood pressure, blah, blah, blah. Patient presents with 2.8 million snips. That means he differs from the reference at 2.8 million positions. He's got 752 copy number variants. That means he's got an extra copy of pieces or missing a copy of other pieces. And by the way, doc, your next patient is already here in the waiting room. No one's ever gonna give us more than 15 minutes to deal with this, okay? So this becomes a prototype. What's it gonna be like for the doctor of the future to deal with patients coming in with their genomes? This is a prototype here. Now of course, in rooms like this, we are teaching the medical students how to do this, okay? The medical students today take electives if they want. They get two weeks of ethics, trying to tell them not to do this, but then eight weeks of go get your genome and let's go see what we can do with it, right? So they're fully informed participants here, but we're doing this, nobody else is doing this here. Nobody else is teaching the medical students, right? We're hitting the ground running here because the technology moves so fast. What does it mean for Steve's cardiac risk? I can tell you, we looked at those genes, that set you up for sudden death, it's not really clear. We don't get crystal clear answers. Well, Steve Quake doesn't have any of the known mutations that give you sudden death, but he had some other weird things in those genes. So we said, Steve, instead of extreme exercise, just limit yourself to moderate exercise. That doesn't sound wishy-washy to you. That's kind of wishy-washy to me. It's also, we know this other gene called LPA. That gene, basically, if you have this variant, you are at higher risk for having sudden death by cardiac risk or atherosclerosis. So we said, Steve Quake, even though all your lipids and everything are fine, we recommend you start on statins. Many of you in this room are probably on statins. And we wouldn't otherwise start them on statins, but we said, Steve, you should start on statins. It's been written about in many newspapers, Steve Quake has decided not to start on statins. We've yet to find the gene in the genome for compliance with medicine. And he's got the risk, Elio, for that one. So someday he'll start with statins. Even after you talk for an hour on this, I'm just gonna show one slide. I'm gonna borrow another slide from Ron Saltman. The long story short from the pharmacogenomics, now we can say something about 150 drugs. If he decides to start on statins someday, he's not gonna get this rare muscle side effect. If he needs blood thinners, he needs less than most people. He's likely to respond to certain drugs. He's not likely to respond to other drugs. And we can say something about 150 drugs now. In my lab, we had to figure out what to say about all the rest of medicine. And so what we realized is that we needed a master list of just every single spot in the genome and every single disease. But we started rereading the entire world's literature on this. Now, believe it or not, we started early. We knew this day was coming. We started about five years ago. We started with a Palo Alto High School kid. We said, let's just get a prototype going. Let's see how many papers this kid can read over the summer. He read about 50 papers and we learned a lot. And we said, we gotta just industrialize this and we did. So papers like this, we record more than a hundred different items, what kind of disease, the gender breakdown, the ethnicity. And now fast forward, we got about seven, 8,000 papers like this red. We got about 2,000 different diseases, maybe a quarter million such connections between a spot in the DNA and a particular disease. We come up with nifty graphics like this. Now, Steve Quake being a 40-year-old white male Californian by just being just from that, he's got a 9% chance of having Alzheimer's disease, for example. Okay, we can get that from the statistics. But now I'm gonna add back all the different spots in the DNA. And you know what, the ones at the top are the most believable. They're in six papers, five papers, four papers. These down here are only one paper. And now if you're a conservative doctor or a conservative patient, you could say, well, I only believe if it's in three or more papers, you draw a line here. Well, say, well, Steve, you only have a 3% chance of Alzheimer's disease. You're a young whippersnapper doctor or patient. You believe everything in the genetics literature. You can follow it all the way down and say you got only a 1% chance of Alzheimer's disease. Kind of doesn't matter, both of those are lower than 9%, right? If you could do this for one disease, you could do it for all the different diseases. And from there, what we do is we show it in a graphic like this. The font size here represents what is my new risk of having that disease. And it's not gonna be about 15 minutes of bad news, right? What am I gonna do in that 15 minute encounter? I'm gonna say you're gonna dive this, you're gonna dive that, you're gonna dive this. No, that's not what we are gonna do. That's losing the humanity here. The humanity comes in front of the other magical part of the equation of diseases. That's the environment. Now we think about the environment because of how bad it is. We have kids drinking soda in schools, 1,000 channels of TV, kids not exercising. We've got the pesticides, we've got the toxins. I think the environment is the new drug, the magic drug here. What do I need to do to my environment to compensate for my genome, to compensate for these predictions here? Well, Steve Quake, you shouldn't smoke, should exercise, you should watch your diet, or I shouldn't drink too much alcohol. Probably didn't need a genome for those four. But Steve Quake, if you love to garden, don't hang around the pesticides, because you're at slightly higher risk for Parkinson's disease and 10 papers have shown have connected pesticides to the etiology of Parkinson's disease. This is extremely crude, this is first pass, this is qualitative, but I think this is a new way, how docs and patients can talk about their genomes. You're doing really well with the exercise, Steve. Why don't you come back in six months and let's see how you do with your diet, right? I think this had docs and patients talk to each other, maybe newly genome enabled here. So just in the last minute here, I just want to thank you all for coming. I think this is the most amazing time to be a student, this is the most amazing time to be a medical resident or a fellow in training here at Stanford, and this is just an unbelievable time to be a biomedical researcher at Stanford, and I'd love to take some questions. I'm gonna save time for questions instead of showing you more slides here, so thank you. Can we have microphones for whoever would like to? Don't be shy, Shirley. Primarily statistical inference rather than deterministic, in other words. That's a great question. So question is, from that genome, is it a statistical prediction or is this really your destiny in some ways? Let me rephrase your question for a second. Remember, the genome is your genome from birth to death, okay? So you think how potent it is or how potent it really isn't to know your DNA from birth to how you're gonna die, it's a long road between those two. So the predictive power of your genome, whatever, just not really strong, okay? It's not, it's gonna be one piece of the equation here, okay? So if I have, for example, a propensity to be obese or to have diabetes and I'm also in the wrong influence or the wrong environmental factors, maybe those things happen then. So there's statistical predictions, but we never know how potent these are gonna be in the end, okay? Why am I saying that? We have, for example, that patients come to Stanford, that let's say 40, 50, 60, 70, 80 year olds and they were born 40, 50, 60, 70, 80 years ago and they have disease today. How much does that help us predict in the future? The future kid today now has 1,000 channels of TV, cheesecake factory size portions. So we try to learn from where we've been to figure out what the future is gonna be like for people. But until we get there, we never know if we were right or not. So we try our best. So the DNA is one piece of the equation here, but it's no smaller than thinking about the environment that people live in. We could have a genome sequence on a person, but if that patient can't afford the meds, we give them, they're gonna have a disease, right? So the genome is amazing. It's part of a personalized equation, but it's not the whole equation here. Does that answer your question? Ah, okay. Behind, how does one find a genome-trained physician? What are the initials behind their name? That's a great question. That's a great question. How does one find a genome-trained physician? That's a great question. I think one finds them at Stanford, okay? Ha, ha, ha, ha. We're training them, okay? So I think I can just honestly say there's no other medical school that's putting out graduates that have had, they're 23 means done. Now if the public is getting this done, you'd think the doc should do this too, okay? So we're starting with that, but we go from there, right? I think the letters that one has in the future are gonna be different. I don't think any field of medicine really covers this today. It could be a branch of the pathology pathologist, it could be the geneticists. We don't even have enough geneticists in the world to actually take care of just the people who've had 23-me profile study, okay? There are only 40 or 50 geneticists trained every year in the United States, 40 or 50. I didn't say 50,000, I said 50, okay? So we don't even have enough, so it's gonna be a lot of fields, but what I'm gonna actually answer is I'm gonna answer it in a more provocative way. I think every doctor will be genome-enabled in 10 years, okay? You're not gonna go to a specialist to deal with your genome, okay? You're gonna be going on a flight to Australia, and you're gonna be going to get your shots. You don't need shots for Australia, but that primary care doc is gonna say, well, have you looked, maybe you're at risk for deep vein thrombosis. You should be the one that goes, walks around the aisles on that flight. Make sure you don't get a clot in your legs there, okay? You can't, you don't have time to go see a specialist, so that every doc is gonna need to know what you're doing and how to take this piece of information and fold it into context. So I'm just gonna have to answer it in that provocative way. You got a question here? So I'm a practicing psychiatrist, child psychiatrist. Charlie. And people come in with various elements elements to what we call depression, what we call anxiety, et cetera. We have classes of drugs, antidepressants, anti-psychotics, anti-anxiety drugs. It's relatively, though I'm convinced naively, relatively easy, to say I wanna put this person on antidepressant. Right. But there are now lots of studies that show in populations that the antidepressants are all pretty much equivalent in their efficacy when you look at them in populations. They all have potentially different side effects. Right. But that's also statistical. I've now heard of a company that you can take a swabs and the swab and among a limited number of drugs, they'll tell you what the likely side effects are gonna be if that's medication. But it's expensive. Who's gonna pay for it, number one. And all of medicine seems on an economic level to be set up in terms of treating disease. We get paid to treat disease. How are you all impacting these things? Because to move from what I was trained to do to what I would like to do, I feel like I'm buried under financial situation. That's right. You're asking me to solve all of the medical problem in one minute. So let me try here, okay? Okay, let me try. 60 seconds. 60 seconds here, right, before the coffee break, right? So it's a tough problem, right? We don't have much of a system at all today in the United States, right? It's evolved to what we have and we could call it a healthcare system, but there are a lot of competing interests. I think the one fair thing to say is that we all want people to do better in this country and in this world, but we all have different ideas of where that magic is. I think Stanford is just unique in really investing beyond even just the walls of a hospital, right? We have a hospital CEO sitting in your audience here who's saying, well, it's okay if we reduce the number of people we actually see in the hospital because maybe that means they're doing better. And how do we invest in actually getting them treated outside? 30% of people have a weight loss app on their iPhones, okay? You could say that's wishful thinking. You could say this is how people think about healthcare today, okay? Let's install on your mobile phone. So what I'm gonna say is we're used to thinking in that encounter in the exam room with what am I gonna recommend? What prescriptions I'm gonna write? I think we need to think broader than that, right? That's the point of imagination. This is what Ron was saying upstairs. The encounter between a hospital and medical center and a patient can be beyond just that 15-minute visit, right? We're consultants. We wanna help you. And it doesn't mean when you're sick but to keep you out of the hospital. Now there gonna be companies that are starting innovation. There's gonna be innovative products here. They're gonna need to make money. Otherwise, you know as well as I do, they don't get funded on St. Louis Road. Eventually others come, there's competition, price goes down, there's value there and people show how this could be done. So I'm not against these companies making money either. This is sometimes how it works in the United States. But I think there's competition quickly and what gets me excited though is I think the kids can do this today, okay? The minute you get all of these kids or developing iPhone ads to think about biomedicine, I think it's a whole new world for us. The whole new world. With that, any other questions? We could take one more, early. Do you think the day may come when a physician may prescribe having your genomic chart just the way he does for your blood panel? Absolutely. So questions, are docs gonna order a genome test? Can I tell you, it's already been done, okay? Just in the past year, we've been working with this incredible family. A mom, a dad, a ninth grader and an 11th grader, okay? This is the most unusual family. They're from Cupertino already, they're out there. This kid, as an 11th grader, analyzed her factor five clotting gene in Excel and presented her results at Cold Spring Harbor, which is where a lot of molecular biology gets presented here as an 11th grade student, okay? I'm gonna spend the last minute talking about this family. Why? Because this family, the father's well-known, he sold a sequencing company to Illumina. This family's really genomically aware. But they said we wanna get our genomes done to figure out what to do. So they stage a picture of a prescription and literally, family of four, get your genomes done, come back in six weeks, literally. We've seen the prescriptions already written. Now, why is that interesting? Well, the father's had a problem with clotting and he was on blood thinners and had a clot again on a flight. And now he knows he's got two clotting mutations, not just one, so now he knows why. But the kids were always wondering, am I gonna get this? And that's why the daughter analyzed her genome and she knows she's gonna get a clotting problem someday in the future, okay? Fine, 11th grade girl, now she's 12th grade, now she's actually in college this year. But as a 12th grader, she started to have acne. And as a pediatrician said, well, go see a dermatologist. A dermatologist says, well, you're going to college, your hormones are all out of whack, why don't we just start you on oral contraceptives? She'll just regulate your hormones in your period and of course it takes care of your acne. Now what does this 12th grader say? Doc, my genome tells me I'm gonna get a clotting problem in the future. Oral contraceptives are contraindicated for me. This is the model, this is how we teach the undergrads, this is how we teach the high school kids, and certainly this is how we have to teach docs. The simple answer to your question is yes, we better be writing prescriptions for this because even the kids today are gonna figure out how to make use of that kind of data. And that's the magic of Stanford. So with that I'll end and you guys can go on to the copyright, thank you. The preceding program is copyrighted by the Board of Trustees of the Leland Stanford Junior University. Please visit us at med.