And we're starting, so what does that all have to do with drug discovery. Well there are a serious of nich- initiatives. How many people here have heard of Sage Bionetworks? Okay, so there, this kind of brings together the number of things that I have been talking about. You know the use, the make, the comments which puts into the public domain, a lot's of biological data, genomic data patient data and everything else in between. that there is an open framework, there are open licensing to make all of that accessible, and so it, these, this is disease models that are going online and being accessible. And so the idea, it's founded by Steven Fren who actually founded Rosetta Pharmaceuticals. which was a spinoff from Merck. So you know, it's someone who's been along the traditional route, and now he's dedicated to the idea of creating open drug discovery capabilities. Now I think that's, that's you know, that kind of example of bringing that together. When I gave this, this lecture last year we were actually just in the point when If you remember, Wikipedia was closed down for awhile. and there was a huge ruckus, and I think that was just another sign that the open access battle was being fought. And it's not been won yet, but it's kind of looking good from my point of view. Alright so, what's all this, we've sort of seen what, you know, what's going on we've seen the notion of a sage bionetworks as an open consortium for drug discovery what is, you know, how could this potentially impact drug discovery more directly. So, and I don't want to necessarily answer that question, but I want to go back and look at the question as why. Why is the drug discovery process, why are we not discovering many new drugs? [COUGH] There are economic reasons. I can't, I want you to tell me but you can't tell me. But you could tell me, but you can't tell me because of for whatever reason. [LAUGH] but, here's a couple of data points that gotta start, gotta get you to start thinking. Alright? Gene knockouts, when you knockout a gene, only in about 10% of cases do you actually see an effect on the gene or the phenotype. Why is that? So, because there's redundant functions. There's alternative network roots. There's robustness in the interaction networks. So that's already got to ask yourself okay you're knocking out a gene, what if you in some way perturb the system with a drug. So you hit a particular target. Well in many cases, because for these exact same reasons, you might be having very little effect. You know, we gotta, you can't ignore these things, alright. And then when you look at the fact that if you take all of the biologically active compounds and you start looking at them and you do it assay them in various ways across panels. You find just from that group 35% of them bind to more than one target. So this whole idea of the premise of the drug discovery. Paradigm for so long with respect to small molecules of one drug, one target, one disease. This kind of data just says, how can that possibly work? To me it explains exactly why, when I turn on the TV and try and watch the news, most of the time I'm hearing about the side effects of drugs. That you know, are, are being advertised during the commercials. And the, the reason there are these side effects, in large part, has to do with these kinds of effects. Alright, so what should we do about it? What is the drug industry done? In my opinion, for quite a long time it tried to ignore it. It kept trying to chase after these blockbusters that were these there and there are a few out there. Many being discovered serendipitously after all this massive amounts of research, but they're all out there, and lot of them are, actually, natural products, they're not actually synthesized drugs. So, you know, what, but if we accept this notion that there isn't one drug, one target, one receptor, what can we do? So here's something that actually drives this home even further, alright? This is the human genome tree. So you and I have about 550 human protein, we're humans so we have 550 protein kinases, approximately. This is the phylogenetic tree of those kinases. Each one of these instances is on the bulls eye kinase inhibiting drugs. They are on the market, FDA approved drugs. And so you can see, you know, Gleevac, which is a wonder drug, and the size of the ball here there is a reflection of the binding affinity to a target. A number of these drugs actually bind to off targets with higher affinity than what the, the known target was, believed to be. And then you take something like, like Staurosporine which is a natural product. That binds to huge parts of the, of the tree. So what does that mean? It means you're not inhibiting a single target, what you're seeing is a massive collected effect. So then the question becomes, well what, how can we use that, if we accept that notion, what can we do with it? So this Ehrlich's philosophy of the magic bullet is, you know, I think is not being realized. So it's just naive in a complex system in my opinion. All right, so how can we apply that? How can we accept that, and then what can we do about it? So I'm going to give you a couple of examples from, from our own work, all right? So, if we accept the fact that we, we've got this multiple binding. Let's, let's try and figure it out, what it is. Let's say I've got a drug, it's this right fist, I've got a receptor and these things are binding together. I know what that receptor looks like now, I know what it's binding pocket looks like, because I have the complex. So let me take that that binding site and let me search across the complete structural proteome, the proteome of the human looking for all instances of a similar binding site. If I find that then it tells me one of two things. Three things it could be may tell me nothing. But it might help explain the side effects of a drug, because I now know another target. But even better from a drug discovery point-of-view, it might actually allow me to re-purpose that drug. Because I've discovered a target that's actually associated with a different condition, and I have examples of all of those. I'm going to try and tell you a little bit about a couple of them. So these are the kinds of things it tells you. And I'm, I'm pretty much described that. So the basic idea is we take, there are in this database I mentioned, there are things like Lipitor. 100 billion dollar a year drug, it's sitting in a public database. An so you can go in this PDB I mentioned before. So you can go an look at it bound to its target, an you can play around with that. So you can take that target, an you can look for other similar targets. An so yeah, there it is. And so what you do, is you characterize the binding sites. And then you essentially search for those binding sites across all these other proteins. And you do this statistical test to see how valid what you discover is. You know, how valid is this alternative binding site, statistically? I'm not going to go into any of the science of this. this is work that's been done by Li Xie in the lab, I just want to say that. And you know so, that's sort of the technology. I'm not going to go into any of the details. So what can you do with it? What are the kinds of things you can do with that technology. So we happen to be interested in neglected tropical diseases. partly because of course[LAUGH] the, you know the reason neglected, neglected is hardly anyone is really looking at them. The drug companies. And not until recently haven't been looking at them in very much. Because there's no money in it. And yet on the other hand you know, there's 1.7 million deaths a year approximately from Tuberculosis. We also you know, people worry about it constantly, because we have multiple drug resistant. And we have extreme drug resistant tra-, strains. You know could eventually get into the, the, the these are third. This is in developing world disease. And unfortunately it doesn't get the attention it should. But if it gets into the developed world. And these strain start really proliferating. We've got a lot of problems. Because there hasn't been until recently. Particularly successful new treatments for TB for 40 years. So this is just happens to be something that we would focus on.
