The interns had the amazing opportunity to interview Dr. Nitin S. Baliga, MSc, PhD, the senior vice president of ISB and a director and professor in the Baliga Lab. In this interview, the intern cohort got to learn about Dr. Baliga’s personal journey, his tips and tricks for managing his time, and his biggest pieces of advice:
Can you talk a little bit about your personal journey?
I’ve been at ISB for 21 years now and I was one of the founding members of the Institute. I came here in 2000 and I was in Massachusetts before that. I came here to help start the institute and develop the framework that can be used for predictive modeling of biological systems. The implications can range from personalized medicine-figuring out how patients differ from one another so we can tailor therapies to the unique aspects of a disease-to understanding mycobacterial resistance-what are the drivers of the problem and how we can be much smarter about devising treatment regimen that can prevent the resistance from emerging and maybe also reversing some of the resisting phenotypes of some pathogens. We also applied the computational framework to understanding environmental issues including issues that involve ocean acidification where the dilution of carbon dioxide in the waters of the ocean make them acidic which can be problematic for a lot of sea life. We also looked at soil contamination at different sites, particularly an Oakridge in Tennessee where a large amount of waste was generated in the Manhattan project that is now leaking into the soil. We were trying to understand the implications of this action on our microbial communities. Those are some of the major areas we are looking at.
I know I said computational analysis, but I should probably say systems analysis. I am a firm believer that you should have a cross disciplinary approach that takes a problem, and in a hypothesis driven manner, race to get all the data types that exist and need to be generated, with new or existing technologies, so that we can then begin to build models that explain phenomenon that underlie the problem we’re trying to solve and those models can help design the next set of experiments to the best of your understanding. That’s the systems analysis approach.
If you look at the cycle, there’s different kinds of expertise. One is the domain expertise, depending on the problem you’re trying to solve. If you’re trying to understand issues related to personalized medicine and cancer, you need to work with oncologists, get patient samples and biopsies. Then you need to have people who have expertise in generating molecular information regarding those samples, including RNA sequencing or single cell analysis, etc. For that, you can use existing technologies and in some cases we develop new technologies., so you need to bring in people who have expertise in engineering or different kinds of technology development such as chemistry, physics, etc. Then, computational biology is needed to analyze all the data that came out of the application of the technologies. This will help build predictive models that take large amounts of data and organize them into models that can be used to understand the different drivers of the phenomena. Further, you can then develop new experiments and then predict what might happen if you were to change some of the key players you identified by either knocking them out or increasing them and things like that. Ultimately, you then go back to design experiments to test whether your predictions are correct. Along that predictive cycle you have different kinds of capabilities that are necessary to be successful and that is often done in a team science approach.
I would encourage you to think about other datasets, whom they are invented by, and what technologies are used. Also, when you have computational models, I encourage you to think about who would benefit from that, how do you know you’re correct, and how can you test it, who do you need to partner with, etc. Ideally, you would be working with a team that does all of those components and more. More work that I’ve seen has been done through programs that allow you to train in multiple disciplines. Maybe you are already exposed to it. In the past, you would specialize in some domain. For example, if you like physics, you could choose physics as your major and specialize in it. I think it still happens, but I think there is more appreciation for project based learning. I think more and more training programs are forcing students to look more broadly. While you may specialize in one area, you recognize that you need to work with others and use concepts from other disciplines. I think that’s a really good thing to keep learning as you progress in your careers! Always think about the problems. Then, when you try to solve the problems it becomes clear what kind of capabilities, collaborations, and teams are necessary to solve the problem.
Can you talk about a project or two that incorporates an interdisciplinary approach with a lot of different people?
Let’s talk about one of the first projects we did. When I first came here, frankly, most of us were inspired by the vision of building predictive models for biological systems and using it to solve important problems. But, most of us, if not all, did not have a clear understanding of how to do it. The journey begins with a question and problem that inspires you, that has not changed. At the time, there were really no technologies that were available like now to monitor an organism’s response to environmental changes. This meant profiling all of the expression changes and genes encoded in an organism’s genome.
Although there were some technologies, we had to build some of our own. That required to code technology from microarray technology-which is not as widely used today as it used to be. That technology uses a robot to print DNA molecules onto a glass slide, and those DNA molecules are essentially small fragments of the genome of an organism. If you go across 10,000 or 100,000 spots, you can essentially have different parts of the genome presented in each of those spots. In a single slide, you can have the entire genome of an organism printed, not the whole thing, but pieces of it that are relevant to understanding how it responds to environmental change by up or down regulating genes.
To build that technology, in the beginning we had to figure out the chemistry of glass surfaces. So, if you put a spot on the slide, the spot stays confined and doesn’t stray afar and mix with the next spot, because you need to have clear dedicated spots that it can know the identity of. We had to work with companies and chemists to figure out a number of chemical surfaces that kept those properties. The spots also couldn’t be too big or too small. Once the spots were in a format that was useful, the next thing you have to do is take these organisms and subject them to different environmental changes that are relevant to what they experienced during evolution, in the natural environment, etc. so that you can trigger responses in the organism. The organism can now say “oh the temperature is rising I need to turn on or turn off some genes.” When it does that, we can over a time course, take samples from the culture of the organism, break open the cells, take the RNA out, and figure out which genes were up or down regulated over time. We can then take the RNA molecules and label them with some fluorescent probe to get those results.
First we had to work with biologists to design the experiment so that it could reproduce the natural environment of the organism and administer controlled perturbation. We then had to have good reactor systems that could subdue that and then you needed to have a good biological understanding of phenotype changes of the organism and so on. We went from biologists doing the experiments, generating all the RNA samples, to working with chemists and other companies to generate the fluorescent probes that we could use to label the RNA, to finally taking the RNA molecules and staking some probes using the added chemistry at the time. There were also many other technologies that also worked, but we had to choose among ones that worked with the types of samples we had.
After all of that, the next step is to take that table of samples and throw it on the glass slides so that the RNA molecules can hybridize. If you have taken some classes in molecular biology, you know that DNA has 2 strands. There’s the complementary strand that hybridizes, so the strands are printed onto the slides of typically single stranded DNA that is complementary to the RNA that is made from that DNA in the cell. That molecule goes and hybridizes, so we had to get the hybridized chemistry figured out. Then, once you do the hybridization, you put a sample after you increase the heat and also another sample labeled with a different fluorescent probe before the temperature is increased so you have a reference. Right on the glass slide, you have RNA molecules from two different samples that go and hybridize in the same substances. RNA molecules from the same gene, but from two different samples. One is labeled green and one is labeled red and they go to hybridize. Once you have done that hybridization, you then go scan the slide. You do that using a laser scanner that goes and activates/triggers the fluorescent of the molecules in each spot. Then there is a reader that reads the fluorescent that is omitted. Essentially, it can scan the entire image and it can get each spot’s relative amounts of green and red that are being emitted. So, again you had to deal with people that built scanners and understood optics.
You get these images of green and red in three; one is of the glass slide, one is in the green color channel and one is in the red color channel. Those images will then have to be interpreted, meaning you need a computational biologist who can look at those images and then using a segmentation technique, find the spots. Remember, for each slide, it’s not going to be identical due to dust specs or somebody’s hair. Although you’re really careful, sometimes you have unintended noise, so you need good algorithms that can distinguish signal from noise. The people that really helped us were astronomers and astrophysicists because they had spent their careers looking at the galaxies figuring out what’s a star and what’s a dust speck. A lot of the techniques they used could be applied to this problem, so then you get these spots identified and you can now quantify how much green and red there is and then all of that data comes as a matrix.
There’s the identity of the gene that is being probed, the value of the green channel and the red channel for T0,T1, T2, everything is red, green, red, green. We then had to figure out what this information meant. We had to work with statisticians who could then analyze this data and under rigorous statistical analysis figure out how to take away the background noise. For each spot you could have a range of possible values, at very low abundance you could have small amounts of signal and at very high amounts you could have high amounts of signal, but it doesn’t go just linearly. You can have a certain amount of signal that is on the further end of the linear scale and then once you reach a certain level, you back out because once you saturate the scanner sensitivity, you have this kind of curve which has to be accounted for. This is all the signal processing and then you have the statistical analysis where if a 2,4 change in one condition versus another is significant or not significant. We had to develop maximum likelihood statistics to figure those things out. We haven’t yet gotten to building the model yet, this is still just kind of figuring things out.
Once you get all the data out, these matrices of data are then fed into machine learning algorithms where you are try to interpret what are the different interactions among different proteins, DNA, RNA, and the cell that were responsible for driving those observed changes. Basically, how did the organism sense an increase in temperature and how did that communicate a signal through a network of interactions that then gave you information on the change of expression on certain genes etc. That’s all the computational modeling that gives you a hypothesis about “hey this is the gene that is interacting with these genes, and these genes are the main drivers that can knock this gene out and then subject the organism to increase its temperature.” Then you go to the molecular biologists and the microbiologists that generate mutants that are missing those genes and you run the experiment and collect the data again.
When we started off, everything was new, so you had to really figure out a way through this path. A lot of the steps were not anticipated and when we encountered them, we realized that we need certain kinds of expertise, we need to talk to someone, or develop something new to solve the problem. It took us about 7 years to go through this entire cycle multiple times to build the first predictive model about how this extreme organism responds to changes in its environment. That was the first model ever built for any organism at that scale, but it was 7 years of work by many many people, with everyone playing a different role. Each role equally as important. To do this, you had to have a change in your mentality about how you did science. That would be one example, but the way science happened after that has been different because you learned that this whole process is cyclist and that different capabilities are required. You established collaborations, built teams, and you gathered people who are inspired by similar problems. This helped the next step of problems become relatively easier to address because you are getting people trained in different disciplines to come join your team and do similar kinds of science on other problems. That covered a lot of ground there.
How did you gather everyone over 7 years of work?
When you do team science, everyone has to have an incentive to participate. Being inspired by a problem is one really important thing, but everyone is motivated by a different goal. Whether that be career, professional, personal and so on and so forth, so you have to respect that and the fact that they speak different languages, that they were trained in different disciplines, and that how they might speak about the same problem might be very different if they are looking at it through a lens of physics versus chemistry or microbiology. It definitely took some time for us to understand each other. There was a lot of mutual respect and patience, and everybody had to leave their ego at the door when they entered the room.
There was no place for anyone to think that what they did was more important than what someone else was doing. This was a key thing. This is true even now. When you do team science, mutual respect is where it begins. If that is not there, team science is very difficult to be effective.Nitin S. Baliga, MSc, PhD
The second is because there are so many problems, everyone gets a chance to take a lead on some aspect of a problem and others have to then play secondary roles. Those are always interchanging as you go through different parts of the cycle. Recognizing that and knowing that if everybody gets a chance to get recognition for the work they’ve done, they’re more likely to put all their energy and enthusiasm into solving the problem.
There was a lot of social engineering that was necessary and that is something that is always evolving as the problems you’re trying to address are different, the training people are getting tends to be different, and what excites and drives people can change based on those things. People who are more experienced are a lot more comfortable being in team science because they know how it works and have a lot of experience behind their back, but newer people are at the risk of not getting the recognition that they deserve or trying to take over and champion their own cause rather than others’ cause. So, you kind of have to have training in the culture that is constantly reset and revised to keep with the times. It’s complex and that’s the fun part. As people are seeing this happen more and more, there’s a lot more acceptance that this can be done. However, there are still a lot of arguments about the best way to navigate this space of collaborations.
Intern cohort with Dr. Nitin S. Baliga!
How do you balance your work and personal life?
I’m busy, but I’m also really fortunate to have a good team that I can lean on! I do that as much as I can so I delegate my work to people who are better at doing certain things than I am. In fact, all of my team are better at doing certain things than I am. It makes more sense for me to understand that as a leader such as who needs to be involved and what roles they need to take on to make it an effective collaboration. If you have a good consensual approach to doing that, then you can allocate your time effectively.
I have certain rules for making sure that I devote a good amount of time to my personal life as well. In the evenings and on the weekends as well, I decide to not do any work. I do tend to respond to emails from time to time, only if they are important or when there are deadlines, but by and large, I take joy in separating my work life from my personal life. It’s good for your brain as well to get that rest. I mean everybody is different, so you sort of need to figure out how you work best, be good at communicating expectations, and making sure that you have the right people beside you. I have some amazing people in my lab that I can rely on, and oftentimes, once they understand that they have the authority to take lead on certain things, they do! Once you trust them and recognize their efforts, then they are incentivized to do more of that. You definitely have to build peoples careers and let them run with it. Which in turn, frees up your time to think about other things.
How does your previous work and research in Massachusetts differ from the systems approach at ISB once you came here?
There was a major fork in the road once I finished my PhD. My PhD looked at a very specific set of genes for this organism I was studying in response to oxygen and light. It was a set of five or six genes that I analyzed and it said a lot about the organism. When I got my PhD, genome sequencing became more accessible and a lot of genomes were being sequenced. Genome sequence of bacteria was something we did in partnership with Lee Hood and that opened up my eyes to what was possible because I was part of the team that did sequence annotations. When you do genome sequences it used to be done by large teams and it still is, but it used to take years to sequence the genome of an organism and computational interpret what those sequences encoded functions where and so on. I found that the organism I was studying while I was focused on 5ish genes, the organism had actually close to 600 genes. It was a very complex genome made up of 3 different DNA molecules and there were a lot of genes, maybe ⅔ of the genome, that didn’t have any known functions or were poorly described functions.
The fork in the road that i’m mentioning was the choice I had to make between going deep into characterizing individual genes down into the atomic level, which is a reductionist approach that is still very valuable, versus taking a step back and looking at all the genes in a much more predictive level to explain phenotypes of the organism and how you change certain behaviors by altering its genes networks. I had opportunities to follow both career trajectories and I chose the one that I did, which is to take a step back and look at the whole network of the organism. It was a risk because the other opportunities I had were in well established places such as Harvard and UC San Diego, etc., but I also had the offer from Lee Hood in Seattle. I really had to spend some time thinking about what I’ll been doing in 5 to 10 years and where the field will be. The other 2 paths seemed safe because they are established paths that people have taken in the past and they know what opportunities are there at the end of it. This was completely new as you’ve seen how it took 7 years to build the first predictive model. This path had to be paved by us and it was charting new territory. There was something exciting about it in an early career stage, you have a lot more leeway in taking risks. In fact, you should take risks throughout your career and I think failure is an important part in discovering big things. You try something risky, you fail, you learn a lot, and you can apply it to something new.
I decided to take the risky path and it was risky for several reasons. Firstly, it was not clear what the job was about. The vision was really inspiring, but systems biology was just a thing written on paper, it didn’t exist when I accepted the position. When I came here, they didn’t even have real estate! They just signed a lease for a small space and had a paper that was taped with scotch tape on a glass door. That was the institute. There was nothing there. But when I moved here there were another 40 or 50 people from different parts of the world with the same excitement and enthusiasm! Nobody knew exactly what we were going to do together, but we were inspired by the vision and those were some of the best times I’ve had. There were a lot of people from many different types of backgrounds failing all the time, but knowing we would keep evolving.
I think that everybody gets the chance to take a risk like this and I would say if you encounter such an opportunity, you may want to think forward about big visions and big ideas when you make the decision. Don’t be afraid of failing. Of course make an informed decision, but you can’t have the idea of failure because of the risk you’re taking scare you away from making the right decision.
Can you talk a little bit about what you saw in the world or just in your own high school education that inspired you to create an interdisciplinary approach in STEM and science curriculum and creating this internship opportunities for people that come from underrepresented and disadvantaged groups?
I did all of my schooling, until I came to the U.S. to do my PhD, in India. I was born and grew up in Mumbai, India. I got my masters and PhD, so my experience was very much from India. We had a series at ISB where people gave presentations talking about how people paved their paths from wherever they were to how they came to ISB. I did that presentation as well, and I had been reflecting a lot about what experiences in my life were instrumental in the path I took and who were the most influential people and events. There were a hand full including my mothers interest in education and how she taught us when we were very young, forcing us to think outside the box and trying to explain things without memorizing it.
Even though memorization could give you really good scores on tests, she insisted that to truly test your understanding, you need to be able to explain phenomena. There was also a professor in college that did just that! They kept saying “I first have to make you unlearn everything that you learned, so you can actually build the right foundation for solving real world problems.”
So, there’s this project based thinking that is really important and I could praise it to a couple of people and events in my life, but I also remember there was an internship opportunity that I had. I spent a few months in a laboratory where I didn’t really do much other than break glassware and run some chromatography, but I learned quite a bit beyond just those few activities. I got to see real scientists, I saw how they worked, how they came to work, how they went home, etc. I just saw professional life in a way that made it all real for me. Whatever was in the textbooks and things that we learned just came to life! Those were a few experiences that I felt were influential. When I was doing my PhD program, I worked with this organism that made all these colors. It made purple and pink colors, and overall it was really just a beautiful organism. I thought it would be a really exciting thing that would get students excited, so I kind of put those things together.
In 2002 I thought “hey I should do something” and applied for a National Science grant for my research. I got the grant on my first try and a part of the grant was for education and broader impacts, which I had no clear idea on what I was going to do for that. I proposed to bring in high school students to the lab and give them first hand experience because in my personal experience this is going to be life changing for them. The reviewers loved it and I got the money. Then I went around trying to get students into my lab. At the time, my wife’s cousin was working in a program that was for adverse Latino youth aimed to give them further constructive opportunities instead of taking up paths that can steer them into trouble. I spoke to her and she said I have a couple of students who would be perfect for that, so she recommended a couple of students for the program. This was in the first or second year. That’s how I used to get students, through word of mouth and begging students to come work with us. They came, spent some time with us in the lab, and I learned enormously from them! I learned that I did not remember high school and what I remembered was actually from my undergrad because high school was so far back that I didn’t remember what experiences I had. I had unreal expectations of the students, but I had so much fun though.
One day I went into the office and found that one of the students crammed herself into one of my bookshelves. When I walked in, another student said. “Woah you did it!” I was amused and asked what was going on, and the other student said, “She challenged me and said I wouldn’t be able to get into the book shelf. I wanted to prove her wrong.” To me, that was just refreshing to see how enthusiastic students are and how they got joy out of the little things. I felt that we needed to have a much better structure to leverage that enthusiasm to do more fun things and give them experiences that could be exciting! I knew I couldn’t do it alone. I had the fortune of working with Claudia when she happened to go on maternity leave. I thought there couldn’t be a better opportunity than that! I told her don’t go back to teaching, come work with me and I’ll give you a job at the Institute. She accepted and we built the program together.
There were many layers to this, it wasn’t something we got set on and said aha this is what I’m going to do. It was trying things out, taking a step back, and asking what was the bigger picture here and how can I build off of this. It then branched off from there. I said let’s take this program and use that as the foundation of which we build curriculum modules off of. We began building teacher training to then disseminate those modules and that’s how the program really started taking off. In the first module, I had to work with the school district in Bellevue and that was an experience! I didn’t realize how busy the teachers really are. As I said earlier, mutual respect is number one. You have to understand what people’s challenges and motivations are. All the teachers I’ve met always wanted us to do the right thing to provide the best learning experiences they can for their students, but they also have limited time. There’s all these expectations about what they need to do, and if you were doing modules that you wanted them to teach, you had to take all of that into account.
We were not the first to approach the teachers. There were other scientists who did the same thing, but those scientists have not always kept their word and often disappeared after they got their money leaving the teachers to defend on their own. I said that could not happen and that we need to assemble something that is sustained since this is built on trust. We need to keep it going year after year and be there when the teachers need us, so it’s been all about building trust. Over time, we now have teachers from around the world reaching out, 600 students applying for internships, but these things come later. If you do the right thing and you really put your energy into solving something really important, it will come and be sustained. I learned a lot and I still am learning new things every day. If you asked me how I did it, I really began with some small things that I thought were really fun and exciting!
What advice would you give to a high schooler or anyone who may be hoping to pursue a career in healthcare, science, or the STEM field?
Be curious! You should find excitement in observations that try to explain the world around you. There are so many interesting things happening around you all the time. Everyone gets excited by different things, find out what excites you. I would say I actually get excited by things outside of my domain. I get fascinated by people doing wonderful things that I don’t fully understand how they do it.
You always need to be a student, forever. That never changes. Then you have to find out where you can make big contributions and who you can go to to get advice about participating in those contributions.Nitin S. Baliga, MSc, PhD
Keep engaging with professionals and have fun, it shouldn’t be a chore. If you’re having fun it means you’re doing the right thing! Work and education don’t need to be painful, they need to be joyful and then you have to do the hard work. You can’t expect quick success overnight. Things take time! No matter what path you take, if you want to take the leading edge, you will keep running into hurdles, obstacles, and people who don’t believe in you and give you every reason for you to quit. Don’t. Sometimes there are good reasons to change course, but you need to know why you want to change course, not because of a small obstacle. Learning how to overcome an obstacle is the point where you really start to grow. Hard work and failure is required no matter what you do. If you are doing everything and nothing is going wrong, you need to ask yourself, am I pushing myself enough, am I doing only safe things or things that are comfortable. Be the one that is driving the cycle. If you let problems excite you, put energy into solving those problems, and try to get experience with professionals, you’ll figure it out! Also, don’t make a decision too soon. Every step is exciting! Keep your mind open and keep absorbing as much information as you can.
Link to ISB Profile: Nitin S. Baliga, MSc, PhD · Institute for Systems Biology (isbscience.org)
Link to Baliga Lab: Home – Baliga Lab (systemsbiology.net)