#20: Chris Sander
Mapping the Labyrinth

Intro: "Be flexible to new methods. Like now, in Machine Learning and AI, there are all kinds of new methods. We're now using some of the latest machine learning to catch cancer not late, but early. And how are we doing that? We're actually analyzing clinical records in collaboration in Denmark, before people get cancer, using artificial intelligence methods to define people who are at higher risk to get pancreatic cancer, which is a very aggressive cancer, and so we can treat them early." AskDifferent, the Podcast by the Einstein Foundation with Nancy Fischer.
Nancy Fischer: Cancer is a diagnosis, not only a very scary one, but it also means for many people, this is it. My life will probably end very soon. But it doesn't have to be like that today. Research during the last twenty, thirty years made big progresses. And today, more than half of the patients who have cancer can be healed permanently. And the research goes on, of course, because there are still hundreds of open questions. How do cells in our body react to certain perturbances? How do cells interact with each other? And how does that help to treat different tumors? Chris Sander can give us the answers to questions like that. He is a Professor for Cell Biology at the Harvard Medical School in The United States and Einstein Visiting Fellow at the Berlin Institute of Health. Welcome, Mister Sander. Good to have you. Hello.
Chris Sander: Thank you very much, and thanks to the people listening.
Fischer: Let's start quite simple. If we put it in one sentence, you try, with the help of computer simulations, to predict how cancer spreads and how cells react. How do you do that? Can you try to explain it to me maybe as if I was a child?
Sander: Well, cancer is a very difficult problem, and it's not solved yet. And so, the question is to find it and then to treat it and then to prevent it escaping the treatment. And that's one problem. The other problem is to catch it as early as possible. And so I'm collaborating with people in Berlin that combine two different technologies together with us. One of them is the technology of doing detailed measurements in the laboratory about cancer like material, looking at all the molecules, doing the sequencing of the DNA, the memory molecule, doing the detection of the proteins, which are the working molecules in more and more detail. There's an amazing new technology that's been developed that's on the one hand. The other one is the technology of computation, not just hardware and simple software, but including the big data kind of computation called Machine Learning or Artificial Intelligence. So we're taking these two technologies and bringing them together to do what? To put a highlight on cancer in great detail to see what's happening in the body as the cancer evolves, because it's an evolving system. And then when it's treated, how does it possibly escape, and how can we treat it again to block the exits of its escape? Or how can we block the exits of the escape in the first place by treating, say, not just with one drug, with two drugs or three drugs?
Fischer: You once compared it with a labyrinth that the exits have to be blocked. Can you give us this picture again? Explain what you mean with this “labyrinth”?
Sander: I'll give you an example. There's a very, very aggressive cancer called melanoma that people get if they're exposed to too much sun radiation, you know, skin cancer. And at one point, about ten, fifteen years ago, there's a discovery there's a single mutation in those cancers that really makes the cancer grow in one protein, one of these working molecules. And now if you block this working molecule by using an inhibitor, a drug that blocks it, you would cure the cancer. And sure enough, many people who had that cancer retreated this way, and it was in the newspapers, and it was fantastic stories that people came back to life. But then a year later, a year and a half later, most of them had the cancer coming back because the cancer as an evolving system had managed to escape the treatment. So now the question is how to look at that material in great detail and find out how you now can find additional ways of blocking it. It's like a large labyrinth. We're trying to chase something that's not going to come out the main door. So block the main door, but then there's some side doors. So you have to know where the side doors are and where that thing that you're trying to catch is gonna go and block out the exits, and then you can catch it and defeat it.
Fischer: And how good do you know all the doors speaking, how far are you with the solution?
Sander: Well, quite far. We're getting closer. I mean, the biology of cells used to work on one or 10 or 20 different molecules out of 20,000 that are in the in the in in in human cells. And so gradually, we're getting a more and more complete picture of what's happening in the cells, how all those molecules are interacting, and how it affects what the cell is doing and the cell interactions and how that works in an organ and in a cancer tissue. But it hasn't been complete. So we have to get from the several hundred to 20,000, and that's a big jump. But all the new technologies since the sequencing of the human genome about twenty years ago have gotten better and better and better. But what was missing was to actually put this now into a computational engine that describes it, what's going on, using algorithms, using computation, using simulation - that's also done for airplanes or for cars or so on - and now predict what's gonna happen, using computation, because the human brain cannot figure out how 20,000 different things are interacting. So we need the computation. And then computation would not just to describe what's going on, but it would predict what would be going on if we come in with a new drug, and how we can do that as early as possible.
Fischer: It's interesting because if you mentioned the word prediction, for example, this is one of your main approaches to science, I read. Like you said, there are five terms that are very important to you. Correlation, causation, prediction, intervention, and engineering. Can you explain to us briefly how they all relate to each other?
Sander: It's like that in all fields of science. Science is meant to understand the world. But understanding the world means what? You think of a child that's learning about the world, he or she is touching things and is getting some feedback, playing with things, getting some feedback, seeing things and getting some feedback. And gradually is learning not just to understand the world, but to do something, to grab something, to eat something, to run, to play, to then later on build something, fix something. And so science in the same way goes from observation to predicting what's going on once you learn how the system works. And then not just predict, for its own sake, but for your own safety and to have some utility. I mean, human beings in the world are interacting with the environment, not just to understand it and not just survive. We are also building things and changing it. And so, hopefully, that spirit, will carry over into science in the future as well, because we have major problems to solve in the near future as you know.
Fischer: That's a beautiful comparison to see through the eyes of a little kid that is also curious. No? And also never wanting to stop discovering things.
Sander: That's right. In fact, there's another connection actually, which is that some of the new software that's being built in what's called Machine Learning or Artificial Intelligence, when you look at it in detail, it's starting to be somewhat like what's happening in human brain. And so we're getting closer and closer to actually modeling the processes, the understanding, all the way to prediction, that is happening in human brain. And probably, we will go beyond that. But for the cancer problem, it means in a very, very practical way, don't just understand it, describe it, but predict what's gonna happen and then intervene, do something useful, and perhaps also engineer something. We are, for example, also engineering protein molecules, the working molecules in the cell, by changing them and trying them out to develop new kinds of protein molecules that do useful things, both for medical purposes and also for degrading plastic in the environment by doing it more efficiently through protein molecules. So the application and intervention and then engineering are super important for science. And I think this should motivate kids also. It's not just about learning. It's about doing something in the world and creating a better world.
Fischer: You research in so many fields. Like, you started with physics - what you studied. Medicine is your topic. Biology, cell biology, of course. Also, IT, if we speak about computer simulation. Is one topic not enough for you? Or would you say maybe it's not even possible to only choose one of those fields?
Sander: This goes back to my education, actually. I went to a classical language high school in Kassel in Germany. And we had, you know, ten, twelve different topics. So as a teenager, I had the mindset of trying to go across the board, understand everything. And then I was advised to go to university. And what topic do I pick? And after long conversations, their answer was physics and mathematics, because that's the underlying basic science that can be used for anything. And sure enough, later on, after having studied mathematics and physics and getting a PhD in theoretical physics, I then used that to move into biology. So this was the way of, taking a very broad based knowledge and attitude, but then applying it to specific problems. But the other reason is that I'm just very, very broadly interested, and that's good news and bad news. The bad news is I sometimes get scattered. The good news is I connect very, very different things and bring them to bear onto solving these biological problems from physics, statistical physics, mathematics, some engineering. And I think it's hopefully very productive to do it that way.
Fischer: Well, I mean, we are very grateful for many things you researched, so I think it is very productive in the long run, definitely. We, also want to look at your own way, on your career. As a German, you teach at the famous, well known Harvard University. Many people want to achieve an academic career, but they never make it. How did you get to the point where you are now? What do you think is the secret of your success?
Sander: I don't think there was any secret, but there was a strong motivation. And the motivation started when I was a teenager, actually, because I was handed these books to read about famous physicists, not just Einstein, but also Heisenberg, and Feynman and so on and so forth.
Fischer: Who gave them to you, those books?
Sander: My parents gave them to me. And so I think that, and the school was also quite positive. But then I spent a year, as an exchange student coming from Germany, a year in Texas, believe it or not. But there, I lived in the family with a scientist. And, there was a very strong atmosphere of positive attitude towards science. And at one point, I asked him: "Oh, I think I want to go into science. What do you think? But to be a scientist, don't you have to be creative? What does it mean to be creative?" So I leaned back, and I was expecting the answer from the big Oracle of Delphi. And the answer came back. He said: "Well, you know, Chris, if you want to do science, stick with it, which means go for it." So this - I was very surprised. And that's what I did. I went through all kinds of obstacles up and down, you know, to to Berkeley in California, to, the Niels Bohr Institute in Copenhagen, to other universities and got my PhD a bit too late. But I stayed with it. I stayed on it because that's what I wanted to do. So the strong motivation really counted. And later on, I had a fantastic mentor when I went to Israel, to change from physics to biology. And, he taught me to ask good questions. Don't just do anything. What are the questions you're trying to answer? And make sure they make sense in terms of actually understanding it, but not only that, also being able to predict and then apply it. So my advice is go for it if you wanna do it. And number two, ask good questions and make sure they're good questions.
Fischer: How important is a good mentor in a career in science? Somebody that supports you or maybe even one or two more people?
Sander: I think it's subtle, but I think it's very important. And it typically starts with the parents, that can create a certain environment, teachers and schools, good teachers, or not so good teachers. And lately, I think what we should be thinking about also media, because the world is coming together in a way - a collective human brain of all of humanity, through media and amazing communication that was never seen before. And the motivation to do science actually should be propagated through these through these social media. And we're seeing some negative developments where bad news and made up things and anti-scientific attitudes are being propagated through the media. So my encouragement is - let's use social media and the amazing connections on the planet among all people potentially and promote science because I think that's going to be important to solve many of the problems we're facing.
Fischer: Because you just mentioned it, how risky, how dangerous, do you think the progress we have seen in the last couple of years against science, against facts, how dangerous is it for science and for researchers to really make their research known? Because I think in many cases, people just turn around and nobody is asking anymore.
Sander: I think it's very dangerous because, there's no guarantee that the last, one hundred years of scientific progress, Einstein and others, will just continue, in a beneficial way. We're seeing many, many tendencies, which counter that. I think it's extremely dangerous as a social and political development. And, it's not just about spending more money on science - that is also important. We're spending all kinds of money on all kinds of things. If you look at the the way money is being spent on consumption, that's increasing, increasing, increasing. We should spend more resources on science. But it's not just about spending more resources. I think we have to change the mindset of the population, worldwide, actually. Otherwise, there's a real danger that we might have a major crisis. Now in that sense, I think Germany, talking about Berlin, actually has a major advantage compared to The US where I observe things in detail. There's much more, understanding among the general population, not everybody, but much, much better than in The US where the educational system is not as well developed at the bottom ranges. So the masses of people, need, what we used to call enlightenment, and I'll call it, education about science and the benefit of science. And I think that is absolutely essential, and we should all work on that. Scientists shouldn't just be in the ivory tower. They should go out and help to propagate scientific attitudes. Otherwise, their own work will be negatively affected.
Fischer: Give interviews in podcasts, for example, first step.
Sander: That's a great way, and I'm really glad that, that we're doing this. Talking about podcasts, I think they're a very positive development. I see that also highlights the difference between, Germany and The US in the cultural sense. The NDR podcasts about Corona by Christian Drosten and Sandra Ciesek. It's fantastic. I think many people are listening to that. And that's a way of dealing with problems, about COVID-19. I think let's do more of that.
Fischer: Absolutely.
Sander: And that will motivate young people, to go forward and to grab science and do something with it. We don't need Greta Thunberg. She's great in terms of the climate crisis, but we need more movement from young people to actually elevate science into something that becomes more broad and to counter these negative tendencies.
Fischer: I would like to go back on your own way also for one more moment. We spoke about mentors, and you also said there were many ups and downs. I heard more ups, I think, mentioning all the great universities you could work on. But were there barriers? Were there turning points? And did you also sometimes maybe thought, oh, maybe I should go on my plan B or do something completely else?
Sander: Yeah. I had many ups and downs. Too many to describe. But at key points when I didn't know what I was doing exactly in terms of which direction to go to and where could I contribute and be creative and make a real contribution. Who was I as a 19 year old or 22 year old to make a contribution? And so I asked advice from people, and I had the courage to ask people who I knew had achieved something, had some wisdom. So at one point, when I was trying to go to graduate school, I took a bus from Texas to California to see somebody called Max Delbrück, a famous German scientist. In fact, in Berlin, there's a Max Delbrück Center that's named after him. And, I didn't have an appointment. No. I just took a bus for two days and went to Caltech, went to the basement, and saw this man, knocked on his door, said: "Look. Can I talk to you for a little while? I'm a student. I want some advice." So he talked to me. And thank you, Max Delbrück, for that. And he gave me advice. He said: "If you want to go into biology later on, if that's what you're thinking about, get some solid training in physics." So there, I had some advice, and that steered me in the right direction. There's another one example like that. I was in nuclear physics, and the down was that, nuclear physicics seemed to be becoming less and less interested. And all of my colleagues seem to know more about it than I did. Many problems were solved. There wasn't that much to be working on or to be solved anymore. So I went to somebody else. I took a train and went, to Göttingen and saw Manfred Eigen, another famous German scientist. Didn't have an appointment, only with one of his lab members. And I went to see him in his office and he finally talked to me. And I said: "I want to go from theoretical physics to theoretical biology. How do I do it?" And he leaned back and said: "Well, there isn't really anybody doing this, except maybe for me. And I point you to three people, that might give you some advice." So I sought out these people. I traveled to different places, and they pointed me to a problem that was really interesting, which is to go from the genetic information in cells to the three-dimensional structure of proteins, the shape of proteins. So I thought, oh, that's an interesting problem. So, you know, through all those attempts to find the right way, I got advice from some key people, and I had the courage to go to them and ask them. And that to my mind was the thing that actually, was a breakthrough for me. And then I asked that question of computing the shapes of proteins, which do all the work in the body from the genetic sequence and solved that problem many years later in 2010. Problem that had been unsolved for forty years, and we had a breakthrough together with my partner Debora Marks and colleagues in Torino. And we solved this unsolved problem after trying and trying and trying again and again. And that was, you know, the result of determination and a long way of getting to the goal.
Fischer: You remember that moment when the breakthrough happened? Like, was there one day or one certain hour when that happened and you were just overwhelmed by excitement?
Sander: Not really. But there were some key events where I had been thinking about this problem in the background, and I met some key people from statistical physics. And I knew how to talk to them because I had been in theoretical physics. And they said, wait a minute. Let's do it this way. And I thought, oh, yes. Of course. And then we collaborated, and that was a breakthrough. So it was more like the recognition. "Oh, here's the answer." But it only came from thinking about it for a long time and asking a pointed question. And the other thing that was really important is to be flexible in your own mind. Don't just try the same solution again, again, again. Be flexible to new methods. Like now in Machine Learning and AI, there are all kinds of new methods. We're now using some of the latest machine learning to catch cancer, not late, but early. And how are we doing that? We're actually analyzing clinical records in collaboration with Denmark, before people get cancer using Artificial Intelligence methods to define people who are at higher risk to get pancreatic cancer, which is a very aggressive cancer, and so we can treat them early. And this is a very ambitious program, but the flexibility came from listening to people who are in the field of AI, sitting around the table with them, and then defining a research program, which was thought to be very ambitious. But after two years, we've got the first results. Collaborating internationally, and this is my other encouragement - be open to international collaborations. And in the age of the pandemic right now, we're now used to actually pulling together people. I have meetings with people in Berlin every week in, with the collaborators in Copenhagen, in Toronto, in Oregon, in Texas. And if I want to have a meeting with these people, some of them are friends - I just call the meeting. Two days later, we're on Zoom. Or two hours later. And I think this is a great opportunity.
Fischer: Absolutely. Because we always talk about the pandemic and so many disadvantages. You just started the, Einstein Visiting Fellowship, but you could never go to Germany or at least not in this regular way of visiting Berlin from time to time and seeing all your colleagues. But I think if I read it right, it was actually in the end of the day an advantage that the pandemic happened for you and your collaborations, because they happened way more often. Right?
Sander: It is. And, you know, after a career, where I started the Department at the European Molecular Biology Laboratory in Heidelberg, and then I helped to form the European Bioinformatics Institute in Cambridge, England, and I started the department in New York for computational biology that grew to a hundred people. And, in that, I now am changing my ways by what I call network science. I'm more and more not working with people in my group, but I'm working more and more with people who are distributed. And why not? Because everybody has their own expertise, and I'm creating groups internationally, that work on particular problems, very carefully selected, and then very intensively work with them on two or three different problems. I think the network science is a new way of doing science transcending the classical laboratory, that we're all used to by the typical professors at TU or FU or anyplace else.
Fischer: So when I calculated right, you studied the physics in Berlin in the sixties. This is more than fifty years ago. And you seem to me as a very motivated and still very curious and engaged researcher and a scientist. So is there one day when you think this is gonna be the day when I just close all my books, put them on the shelf, and this is it - I'm retired.
Sander: No. I like hard problems. I can't even spell the word retirement. I like hard problems, and I want to solve the hard problems. And the cancer problem is one of them, where my collaboration with Blüthgen and Charité in Berlin, and others, we're trying to solve this problem, and we see the way. So the energy is there to actually make a contribution to solving this problem. And, that I think is only possible by connecting to all the new technologies, both the molecular technologies and the computational technologies, and then work together with the conditions to actually put this into practice. And it will take some time, but the feeling of being able to make a contribution to solving a difficult problem after having solved at least one other difficult problem, that's a major motivation. And that's the thing that keeps me going. And I think that anybody at any age, can attack difficult problems. And then if they're lucky, solve them.
Fischer: Thank you so much for your time and expertise. I spoke to Chris Sander, Professor for Cell Biology, expert in cancer research, and Einstein Visiting Fellow. This is AskDifferent. My name is Nancy Fischer. I thank you very much for listening. Thank you, Mister Sander.
Sander: Thank you.
Ask ifferent, the Podcast by the Einstein Foundation.