An Interview with Justus Kebschull, Postdoc at Stanford University

Aravind Krishnan


Justus Kebschull is a 29 year old Stanford postdoctoral researcher. Born and raised in Germany, he completed his undergraduate education in England and his PhD in New York. Kebschull, during his undergraduate years, was interested in gene regulation and not very interested in neuroscience. While trying to attain his PhD, he discovered the idea of mapping and producing a computer model of the human brain with the help of a professor.


Q: If you are comfortable with sharing this information, what is a typical work-day for you as a neurologist?

    A: I am lab-based.  I get up at eight, go to the lab, and then I work for about somewhere between 10 to 12 hours, go home, do other work, and fall asleep. The amount of hours depends on your own choice, as you are very independent. You just need to get your work done. That may take 5 hours or may take 14 hours. Also, depending on who you are, you may want to get up early or late. And accordingly, at work, I set up my experiments. Right now, I am working on animal research in mice. For example, if I am doing a tracing experiment, I do surgery on it to inject a tracer, or virus, in a specific area of the brain of the animal which is to be used as a landmark where the area of interest is, and you can precisely inject a teeny bit of virus in that area. You can then express whatever protein you want to produce and let the mouse run around while you are doing that. I, then get the brain, section it, isolate the RNA from it, and the brain can be subjected to further investigations using other regular techniques.

    The other thing you do a lot of aside from experiments is writing and reading. Obviously, none of the work you do is a vacuum, as there are countless people doing neuroscience research both now and in the past, so you have to be on top of what other people are studying in your own area of research. Even unrelated things are valuable, to get a sense of the approach. You can both gain data points and inspiration from such unrelated things. You have to sit and read those papers, and you get much faster at doing that, after which you write abstracts, papers, poster material, and other scientific publications. One major way scientists communicate with each other is writing these papers and publishing them. This takes a long time, getting your experimental results, putting them in line, getting a logical argument, and making interpretations from your data, all of which takes months to make better, and then edits, illustrations, and the like. And that also takes a lot of time.

    Towards the end of my PhD, I did much more writing than experiments because I did the experimentation and reading beforehand, of course. But then you never stop reading and writing about experiments. Lab life is very impressive. You guys will hopefully get a chance to experience that in a lab in the summer, perhaps. It’s a very pleasant life because you chat with your colleagues what you find out, bounce ideas off people, and it’s all very related. You never realize the right explanation for something until you discuss it with others, so it is a very collaborative environment. Many of the projects I have worked on have many authors on the paper working together. This helps everyone contribute in their area of expertise, such as math, surgery, or the like. It turns out that science is getting more and more complicated as you go on. As you look at more and more papers, you see that the author links go on and on in recent times. Compared to papers in the ‘50s, seeing 20 authors on a paper is nothing unusual.


Q: What challenges have you encountered in your line of work thus far? In your area of work, what challenges or obstacles have you encountered thus far? How did you overcome them?

    A: Lab life can be somewhat challenging on a personal level since it is not like school where you take a test and get a grade and know how well you did. You don’t get satisfaction from getting good grades. Most of the experiments you do are going to fail, for one reason or the other. It’s usually for a technical thing, but sometimes your hypothesis is wrong, and you have to start over. But that’s what science is. Often you have to try out different possibilities and see what works. So more often than not, your experiments will not work out. It doesn’t mean you are a bad scientist, but you have to be perseverant to get to the bottom of your research question. PhD is a bit of a shift even from college, as there are no more grades, classes and the like. Now, you simply are focusing on one main question, and have to work on that. Science can be hard at times, but it helps to have people around that are also doing experiments, so it can be a shared thing. And obviously having a good mentor and advisor is somewhat important, as they can give you assurance that you are on the right path. If you give up on your first try, you will never get to the answer, as people would have figured it out already if it was that easily.

    So, frustration in the lab is the most challenging, but the reward of discovering something new that no one else has ever seen before is a great feeling to work towards. You just need to learn to sustain yourself in the disappointment in between. It’s not bad, and I enjoy it, even though I spend many weekends and nights at the lab. I like being at labs, so I spend a lot of time there. I don’t have a family or kids yet, which is a limitation some postdocs have in terms of the hours they can work. Those are the main challenges of lab research.


Q: What area of research do you work in, and what current research projects are you working on?

    A: In my PhD, I worked on mapping the brain, trying to find out systems and figuring out how different parts of the brain are connected. Now, I am shifting my focus to try and figure out why brains are connected the way they are. So I am looking at Evo-Devo, evolutionary comparisons of brain circuits and cell types we find in different areas, and comparing very simple work and complex brains. I compare their dichotomies in lab computers, by adding computational units. It’s kind of like improving a computer by adding more transistors. I am also looking at mice, frogs, and birds, and comparing those to see the differences.


Q: What opportunities are there for high school students interested in neuroscience to research?

     A:  It’s possible you guys can work in a lab during high school. It seems to be popular, and it is worth the experience. It’s often difficult to get into these programs in high school, so don’t worry if it doesn’t work out. At college, as a neuroscience or biology major, it’s better to spend your summers working at labs. For preparing, it’s important, even though you might be neuroscience or biology majors, it’s important to have a good mathematical base for computational analysis. It’s very difficult to pick up math skills in college, so it’s best to have a solid foundation in high school. If you don’t have this, you’ll do just fine, but your limited to the subsect of things you can do in neuroscience, and focus on different questions, which is fine, if that’s what you like. You have to remain curious. At the beginning, I had no idea what I wanted to do. You might want to do neuroscience, but in college, you might get exposed to something else, in biology or elsewhere. You just have to keep an open mind. The most important thing to make a good scientist is to be curious about questions, which will get you through the difficulty of conducting research and answering questions. So take the opportunities. I’m not an American, so I don’t know the college admission process, but it’s always helpful to have lab experience on your CV [curriculum vitae]. It all starts from high school, after all. Your club is a great example of this, science competitions, science fairs, and other random science things you do are what sets you up to be a good scientist: thinking critically about the questions you want to answer, and thinking about how to obtain answers to questions. And don’t be discouraged if things don’t work out, because most of the time, they won’t.


Q: Dr. Zador mentioned that you won the Katz prize, can you tell us a little bit about that?

    A: Katz was a famous neuroscientist who died 10 or 20 years ago of cancer in the middle of his career. He is famous for visual neuroscience and olfactory neuroscience, doing very important work. And so, in his memory, they started the Katz Prize for PhD students to recognize their work. I got to go to a meeting and give a 40 minute lecture, where I got to present my work to everyone (MAPseq: Multiplexed Analysis of Projections by Sequencing). We’re essentially trying to map out how single neurons and their axons connect around the brain. The idea is that mapping the brain is very important. If you have a map, you can orient yourself, derive how things move about the brain, kind of like the useful GPS on your phones, like paper maps before. In neuroscience, we don’t have good maps of the brain, and the ones we do have are only in certain areas and are not very detailed. Every area contains a million neurons that have stimuli and are wired up in different ways. What we had to do to map these out was microscopy. The idea there is very simple. You image the brain to see the neuron. To see if it is sending the processes, begin at the neuron and follow it through the brain. The problem there is that all these processes are extremely tangled up, and it’s very difficult to clearly follow them. To address this, people color different neurons with different tracers, like GFP, or green fluorescent protein, which is produced by jellyfish. Then, you see one glowing neuron, which stands out from the other neurons, which you can follow through the brain. You can do this one neuron at a time, and it takes a week in total. If one area has 100,000 neurons in it, tracing 1 neuron at a time is just scratching the surface. You might only trace about 20 neurons in an evening, which is probably only certain types of neurons in the area. So what we did in Dr. Zador’s lab is to instead label neuron not with colors, but with these DNA barcodes. The idea is the same like in a supermarket. You can identify each neuron by these DNA barcodes similar to the products in a supermarket. Every neuron gets a different combination of letters, A, T, C, and G. So you have these different nucleotide combinations. And what’s neat about that is, with these 4 different bases together, the number of different sequences you can make grows exponentially. For a 30 letter “barcode” there are 1018 possibilities of combination, which is the number of grains of sand on all the beaches in the world. It is 10 billion combinations more than the amount there is in a mouse brain. This allows us to label many neurons in the same brain to differentiate them. We use an empty virus to put that sequence in a neuron.

    This sequence then moves from the cell body to the targets, wherever the axon leads to. If you inject it in a neuron and you find the “barcode” in your ears, then that’s where the neuron leads. So we can now trace 100,000 neurons in the same timeframe that we used to be able to trace 1 neuron. This showcases the interconnectedness of the neurons. If you have 10 different neurons in area A, they are each specialized. In the visual cortex, you might have one for up-down, side-side, and the like. Each of these types only has one peripheral target area. You can imagine that each of these areas has a distribution of different strings of information across the brain. If the barcode is present at all downstream areas, that could mean that each of these areas gets the full complement of the neural signals. This can be applied to computing, because specific types of information are needed. In my lab, I am now comparing how patterns of project compare in birds, frogs, and mice, to see if these processing pipelines are present and how they compare.


Q: How did you first get interested in neuroscience?

    A: I wasn’t really interested in neuroscience at all in my undergrad. I was this network guy studying how genes regulate each other and how cells conduct this regulation. When I got to Cold Springs Harbor, different professors came in to talk about what work and research they are doing. One came in and had this idea of actually mapping out and producing a map of the brain. Being a molecular biologist, having the mouse genome and human genome available, I was surprised that there was no high resolution, detailed map of the brain. And so, I wanted to help get this fundamental information for neuroscience, and apply my work in gene regulation to the connection and regulation of neurons. Right now, in Dr. Zador’s lab, it was a very fundamental idea to find out about how animals do things.


Q: How does work at Cold Springs Harbor Laboratory compare to work at Stanford University?

    A: The work itself is very similar, as in I go to lab, do experiments, talk to colleagues, and go to my professor. As a postdoc at Stanford, I am even more independent than I was at Dr. Zador’s lab at Cold Springs. The weather is also a lot better at Stanford, but Cold Springs is a lot more focused research institute, where there can be up to a thousand researchers, and a PhD class can have maybe 10 students. Stanford is huge, with so many undergraduates and graduates, computer scientists, mathematicians, and the like. You don’t really know everyone on campus anymore and that is one aspect I miss about Cold Springs.


Q: How does computational neurology factor into brain imaging?

     A: Computational biology is extremely important. A lot of what I do would traditionally follow in computational biology. We have new technologies and are collecting data a lot faster than we used to be able to. Imaging methods like microscopy are much faster and are of higher resolution, with terabytes worth of data. Neural recording, such as sticking devices into the brain to record a thousand neurons over time at the same time. It is your job as a scientist to interpret this data that you collect. I think the most successful people are able to conduct experimentation as well as data analysis. There are people working in both fields, to supplement each other, so everybody should learn how to code, in any real programming language, such as MatLab, and just be comfortable around data. It’s very useful to work with large data sets. Maybe you’ll have a collaborator who will do the computational biology, but especially in neuroscience, even if it is not your main thing, it is extremely useful to know how to code. At least, you have to be able to look at someone else’s code and verify their work, to avoid just blindly believing them for what they said they did. If you have any opportunity to do some coding, it is certainly wise do some coding.


Q: What programming languages would you recommend learning?

     A: Neuroscientists, at least systems neuroscientists, use MatLab a lot, which is not a real programming language, but is useful. Also R, which has a lot of support for genomic studies, such as sequencing. As for real programming languages, Python is good for most things. If you are better than that C or C++ is even better, but R and Python are sufficient for what you need to do.


Q: As a neurologist, would you recommend studying abroad and where?

    A: I don’t know what opportunities you have as high school students. But for neuroscience, Switzerland, Germany, England, and Lisbon are all good options.


Q: How does life differ from Germany to life there at Stanford?

     A: I grew up in a regular household in Germany and went to a regular school there. At some point, I decided what I wanted to do with my life. I always knew I wanted to do science and research, but I had no specific area in mind, like curing cancer, or the like. I didn’t know what I wanted to do. In Germany, you pick a major right from the beginning, like biochemistry or zoology. I looked around and found Natural Science at Cambridge University in the UK. At least I started with Chemistry and Biology, and ended up doing systems biology BA and Masters, which are combined in the UK after 4 years of University. So then, I studied how network structure impacts network function, which has applications from protein function to how Facebook works. I was actually primarily interested in studying gene regulation. I am studying snRNA, which are small pieces of RNA that regulate genes after they are transcribed, differing from traditional mRNA. I wanted to work on this topic, so I went to Cold Springs Harbor, where I found a researcher working on the concept. I ultimately ended up going through through various rotations at this laboratory. Cold Springs Harbor works where you start taking classes as an undergrad and choose labs you want to work in. Most places in the US allow you to try out different rotations in different labs to choose what you want to because you only get a theoretical education in undergrad. It was when I worked with the snRNA researcher when he convinced me to abandon this whole gene regulation idea and become a neuroscientist, because I had this networking idea of making a big connectivity network of the brain as a systems biology. So I joined Dr. Anthony Zador’s lab in 2012 and then started working on these various brain mapping techniques, publishing a few papers. After years of enjoyable toil, I graduated last summer, as PhD takes 4 to 6 years of work, including the first year of classes and things like that.

    If you want to become a professor in biology at university, you have to join another lab, more independently, under a professor, and do a postdoc, which is between 2 and 5 years generally. And so that’s what I’m doing now, in neuroscience, hoping to do some work, find something out. I want to apply for professor positions in the US, Canada, England, Germany, Europe, or elsewhere. I was never interested in the brain as a child, so you’re way ahead of me!


Q: Do you have any closing remarks?

    A: I guess the message I would give you guys is to maintain your enthusiasm. Don’t just do it for college, your CV (curriculum vitae), or PhD. Science is only rewarding if you are actually interested in the questions. Things you may think are simply cool may be worth researching and could even yield new discoveries.

    Aravind Krishnan

    Aravind Krishnan

    I am a 11th grader from Hillsborough, New Jersey with an avid interest in neuroscience and biology. Neuroscience has the vast potential for new discoveries and betterment of lives, and I am looking forward to pursuing the field in the future. It must be a core component of any education, and I hope to help create the next generation of neuroscience professionals.