W. Daniel Hillis
W. Daniel Hillis Co-founder, Applied Minds
So I guess the theme of what I'm interested in is how these very simple things can make complicated things. And one of the most amazing examples of that is the human body, where we have genes making proteins, and those proteins are interacting. So at some level, it's all these little building blocks. And then, the result is, we're alive. We walk around. We eat. We sing. We talk to each other. How can you go from those little chemical building blocks to life? It's kind of an amazing thing.
And so I got very interested in – there's a lot of interest in genetics right now. Which I think is sort of like having the computer program. It's the computer program of our body. But what the computer program is doing is it's producing these proteins. And so that's what's actually happening in your body. So if you could really see those proteins, then you could understand what's happening in your body.
So why is that important? Well, right now, if you get sick, you probably don't even notice it until you get really sick. And one of the most dangerous examples of that is cancer, for instance. That every time somebody gets cancer, they usually – something was going on inside their body long before that, that if you had cured that, they never would have gotten cancer.
So if we could have a way of looking into the body, and seeing what's going on at the level of the proteins interacting, then we could tell that you were starting to get cancer, long before you ever started showing any symptoms. And that would be the time to fix it. So I got very interested in cancer for that reason because it's a breakdown of the system. It's not really an infection, where you have something coming from the outside, but something wrong with the system of your body. It's like a computer bug. Your body is cancering.
In fact, I think, we make a mistake when we talk about cancer as a thing. Because that makes it feel like it's something you got, like appendicitis. But I think, cancer – we should talk about cancering as a verb because it's something you're doing. And actually, probably, most people are cancering all the time. But your body is keeping it under control. And your body has lots of mechanisms for stopping it, keeping it from causing you problems, turning into a disease. And so if we could understand that process of cancering, and controlling the cancering, then I think we could stop it from getting out of control in people much earlier. And so we wouldn't even really treat cancer. We'd just stop you from getting it in the first place.
So what we can do now is we've actually developed a tool where we can see, for the very first time, all the proteins going around in your blood. So we can look at 100,000 protein markers, and look at the levels of them, and see how they change. And we can see, in somebody who's about to get cancer, can we see a signature and those changing? So that we can pick up precancerous conditions. And we've proven that we can actually see precancerous conditions early enough to fix them before you get cancer. So that's very exciting.
Now, of course, the puzzle is – now that we have all this fantastic information – for the first time, we see what's going on in your body? How are your cells talking to each other? We can listen in on the conversation. So now, we have the new avenue that's opened up, is listening in on that conversation of the body. So it's sort of like I've got the first telescope. And you're just peeking through it – wow. Look, there's the stars. There's galaxies. There's all this stuff that you never imagined there.
And that's kind of at the point that we're at, looking at the proteins. So there's a new field that's happening called proteomics, where we can suddenly see all of this stuff, and understand – well, hopefully understand. Right now, we can just see it. We don't understand it. But that could be the key to understanding what's going on in your body. And to noticing all kinds of disease states, and heading them off before they happen. Not just cancer, but autoimmune diseases, neurodegenerative diseases like Alzheimer's. So it's pretty exciting.
And right now, because we can see this for the first time, we're just kind of doing it at the level of recognizing patterns. We don't really understand what those patterns mean yet, but we notice they're associated. So we can get a blood sample from people who, we find out later, have cancer. And we can see, what pattern did we see that would have told us that was coming?
Genetics is fantastic because we can now read what your ingredients list is. So you're a meal, cooked up by evolution. And I can tell what ingredients you have in you, by looking at your genome. That's amazing. And actually, sometimes, people have certain diseases because they're missing a certain ingredient. And so there are many genetic diseases that we can just say, hey, that ingredient is broken, or you've got the wrong kind of ingredient there.
But that only tells you so much. I mean if I gave you a list of ingredients in a restaurant, you couldn't really tell if it was a good restaurant, or a bad restaurant. It's more, what comes out of the kitchen really matters. And that's what these proteins I'm studying are. They're what comes out of the kitchen. And so really, the difference between a healthy person and a sick person is usually not their genes. It's usually what their genes are doing. And that's the thing we can measure for the first time.
Well, so the first thing I'd like to be able to do, is get it so that every time you go in for a physical, we take a blood test, and we tell, do you have some precancerous condition that we can head off now, before you get cancer. That alone would just be amazing. So then, the next thing would be to do that for, say, neurodegenerative diseases. And then the next thing would be, sort of generally, be able to predict enough – use that information – to begin to build a kind of a model of how your body works, which we don't have right now. So imagine kind of a weather prediction model, so that we could tell not only are you sick now, are you healthy now, but we could sort of run that model forward, like we do weather predictions. Say, hey, you know, you're going to start getting a stomach ache in three days, if we don't do this. And so that's the farther thing where we really get a mechanistic, or predictive model of how the body works. We're still a long way from that.
But what we are doing, we're doing it in mice right now. So we're trying right now, to see if we can build a model of a mouse that we understand, if it eats this, or we give it this drug, that it will develop this way, or will get cancer, or it won't get cancer. We're trying it for the very first time in mice. And it's incredibly complicated because we have to model how the cells are talking to each other, the physical forces of how they grow, and stick to each other, and all of the signaling in the organism, how its hormones respond. Things like that. So there are many, many levels of modeling. So it's not just like a weather model. It's more like a global climate model for the body.
What are the major risks posed by epidermolysis bullosa?
Answered by Discovery Fit & Health
What is most exciting about working on the Ekso?
Answered by Katherine Strausser
Why do bacteria communicate and do we care?
Answered by Susan Sherwood and Discovery Channel