Graphic Design: Aswin

Aswin Sai Narain Seshasayee has recently returned to his favourite place in the world, south India, to take up a faculty position at NCBS. In the interview below, Aswin tells how his interests in genomics and bacteria were initiated, how his career took him from Chennai to Cambridge, and what he thinks about his current research on global regulators of transcription in the bacterium E. coli.

Hi Aswin. Could you tell us a bit about your early life and how you got into science?

I was born and brought up in Chennai. My parents moved there in the 70s, well before I was born in 1983. I went to the local school, but moved to a bigger school when I was 11. It had a more modern approach - less mugging up!  I had a good time there, and by the time I was ready to start my Plus 1, I had my first important choice: whether to take biology or not. One of my teachers helped me quite a lot. She said: "Unless you want to become a medical doctor: don't take biology! It will mean you can study one subject less." I didn't want to become a doctor, and I wanted to keep my options open. So: I thought, OK, go for computer science instead.

After school came the next big decision: where to go? It came down to two choices. To move away from home, go to the Birla Institute of Technology and Science in Rajasthan, and do IT. Or to stay at home, go to Anna University and do Biotech. Anna was 15 min on a 2-wheeler. I thought: I don't want to go out! Stay here, it will be fine. I enjoyed my undergraduate days in the Biotech Department at Anna. It was very open, very flexible, the profs just let you do what you want. You didn't really need to study very hard to get the marks, and still I could explore various research options. I didn't really study anything I was not interested in, and still somehow managed 9.5 and 10 grades. I could finish some of the 3hr exams in half an hour. The profs really encouraged independence. There were just a few courses I had to mug up, otherwise I could just indulge my interests.

It sounds like you were already a very motivated man, where did that initiative come from?

One thing was in my first year, when the 3^rd year UGs students organised a student conference, BioTechcellence. These are held each year  and are at a national level. Until then I didn't know much about doing research, I thought I'd just be reading stuff, and then I'd get a job somewhere, that was my outlook. But I thought this looks like fun, these were people only a year or two older than me and they were talking about things they had discovered. And then I caught hold of one of the PhD students there. The seniors had always said he was very smart, so I talked to him, and he took me to his professor, Dr Gautam. He is probably the most student friendly prof you can imagine. I did almost all my independent projects with him. He let you do whatever you want....GREAT! It's the best way to learn, because if you're spoon-fed, and you come across a tricky situation, you need your supervisor to solve your problem. He very rarely helped me out of situations, he just said: go find out the answer.

What sort of interests were you developing?

I just started looking at sequences, multiple alignments, very basic things, things you can now do in 10 mins but at that time it was very exciting. I took all these sequences from bacteria. Aha, they all looked so similar! Then I also started to look at protein structures, 3D images. I also got into molecular dynamics, giving the protein structures force-field patterning, and seeing how the protein's atoms will start moving around under certain conditions  You would even get a movie! That really got me going. And I thought this is something I could focus on for a while at least, doing research.

We had a nice computer cluster going that was good enough for the purpose, and Dr Gautam had one of these lipase structures that was interesting. I started simulating that, and that led to a paper. I thought this is what I want to do for my PhD, solving crystal structures and then looking at their dynamics, but I never got to do that.

Because in my final year I met Madan Babu who was 4 years my senior, we never overlapped. He has now got tenure at LMB, Cambridge.  When I met him he had finished his PhD and was on a break, he got me really, really interested in genomics. And he sort of changed my flight path. I had all been set to go to the US, that was sort of traditional, everyone was going there.  But he was the only one who had been to Cambridge, as far as I know, and he somehow convinced me to go. And it was another big decision, it was taking a big risk. At first there was no PhD offer, and I giving up three PhD offers from the US. My dad was not very happy about that! But Nick Luscombe, the supervisor with whom I was going to work, he called and he spoke to my dad. He said, "You don't worry, I'll take care of him."

Aswin

So I went to Cambridge and decided to work on bacteria. That was an interest from my 3^rd year when we had a course on molecular pathogenesis  - that was 60-70% on bacteria. It was given by Dr Sankaran who works on E. coli and Shigella, diarrheal pathogens, and it was an amazing course, one of the best I had ever taken.  But it made it a bit difficult because Nick had never worked on bacteria before, he is a yeast genomics guy. So he was going to be as naive as I was, with my undergraduate background. It was not easy to start with. But there was a point of similarity between Nick and Dr Gautam. They both gave total freedom.

Looking back, I probably would have taken a different approach to what I had taken then. I was basically taking E. coli datasets and looking for patterns that could help answer certain questions. But what I realise now, and it was something in the back of my mind for a long time, but did not come to the front until I was writing my thesis, is this. In the E. coli database, we do not have the regulatory interactions for at least half the transcription factors. And this does not mean we are missing half of those reactions - it means we are missing more, because even for the 50% that are in the database, the information on their interactions is far from complete.

And my outlook also changed when I read a paper by Steve Busby who is one of the leaders in bacterial gene regulation. He was looking at CRP, which is the famous E. coli global transcription factor, and he said it probably does not bind to the DNA as a simple transcription factor.  We thought we knew as much as we possibly could about E. coli, and this protein in particular, but then he shows that it binds all over the genome and 99% of the binding sites do not seem to affect transcription. So I thought, now we have a problem! I started to think we hardly know anything, is the database I am using complete enough and accurate enough? Probably not. So I felt I could not entirely trust patterns I observe with this database anymore.

So this has influenced your current research plans, where you are adopting a deep sequencing approach?

Yes. The way Im looking at it, my interests are such that the bigger problem is that I just do not have the data I need. The lack of comprehensive coverage of regulatory interactions is made worse by the fact that the gaps are not random. If they were, you might be able to get around it somehow. But if there are systematic biases, then there is a real problem. So I think we still need more data for E. coli, and only then can we go back and look for patterns etc. I  want to bring E. coli up to the point where yeast has been for quite a few years now, probably since 2004.  We need a comprehensive dataset for regulatory networks.

So why did E. coli fall behind?

Well the eukaryotic field is a bit more accepted and the younger people do tend to go to the eukaryotes.  Probably, bacteriology remains largely conventional and sees the detailed side of the picture, while not considering the complementary 'large-scale', 'systems' view of things. In Cambridge young people would always be shocked when I told them I was working on bacteria - "But why?" was the typical response. For me I am always tempted to answer that eukaryotes are rather insignificant. Viruses and bacteria are the things that really matter.

And that eukaryotic obsession is worldwide?

Yes but Nick pointed out that, from the applications that pass his desk, there is more interest from the sub-continent. And he told me he realised if he wanted to build up the bacterial side of his lab, he "would have to get Indians"!  So, now back at my old lab Cambridge, there is one person who is  an 'obligate bacterial person', and he is Indian as well.

So India is a good place for you to be now?

Yes, I'm thinking there will be more people with whom I can interact. I've talked to quite few students back in Anna, and the interest in bacteria is still strong, helped no doubt by the biotech emphasis there. But what is not helping is the massive coverage you get in the newspapers for stem cells (which is nice in its own way), they have become trendy, and bacteria are not (unless of course there is some strong message about antibiotic resistance or controversial announcements about bugs that eat arsenic), they are not in the news.

Have you been able to start building up the lab already?

Yes, we have a JRF here already and an undergraduate student, and two more JRFs coming next week. The JRF will actually have a poster next week during the annual talks, on some computational work. We have computers right now, but we are still organising lab equipment. So the JRF has been learning the ropes, getting to know the computational tools. I was not expecting much so soon, but there is a result, it just needs a little fleshing out.

So when will the experimental work start? And will that involve you on the bench at all?

Me on the bench is a disaster!

Looking through your publications, there are many that are purely computational and then others that are heavily experimental, genome sequencing for example. In the latter, are you contributing with your computational expertise?

Mainly but not entirely. In 2008 I went to the lab for 6 months two days a week, to make sure I can do all the lab stuff, RNA extraction, chips etc. If required I can do the experiments but that would not be the most efficient way to do things. In the lab I am too slow, and that is one of the reasons I decided to focus on the computational side originally. But the experimental techniques we will be using in the immediate future are all stock standard stuff, we already have all the protocols. Once standardised they never fail. The real challenge is once you have the data  - that's when you have to make sense of it. That's where I hope to make an important contribution.

In some of your earlier papers there is a strong emphasis on network theory. How do you see that side of things now?

I think my interest in the network stuff is really on the wane. I still read network papers, but the more I read the less I am convinced. In my old lab we used to have a term (proposed by a post-doc Karthi Sivaraman there) - "correlomics" - meaning that for any data you could always find some other data to correlate them with.

It's always been a strength of NCBS that many of its researchers adopt both theoretical and experimental approaches. You would seem to fit in very well.

I think so. And I also plan to extend what I do here with E. coli into the pathogenic area. I'm particularly interested in how these regulatory networks evolve. As an organism acquires new genes, what happens to the existing networks? This is one area where I want to expand. I will set up a collaboration with the molecular pathogenicity team at Anna University. I believe that we could make an important clinical contribution.

In some of your papers I see that you have already suggested how some of the patterns in the regulatory networks could be important clinically.

Yes, well when bacterial develop antibiotic resistance that happens because the drugs are killing off the vast majority of the bacteria, which are not resistant. But another way to approach the problem is to avoid killing the bacteria - just change them so they are not pathogenic anymore. So most will survive (that is an assumption, anyway), and thus you don't select for a population of super bugs. If this is possible we should be able to get around antibiotic resistance. Transcriptional regulators have some promise as targets, because targeting them will not probably kill the bugs. Most promising are the global regulators of transcription, which I'm very interested in. You can get rid of them, at least in standard lab conditions, and the bug is fine. We need to know all the genes they control, and how they in turn are regulated, and interact with each other. And how do they integrate with genes that are horizontally acquired, that we introduce ourselves.

Have you thought about possible collaborators here at NCBS?

The people who are trying to get me out of my bacterial comfort zone are people from Uma Ramakrishnan's lab, who are interested in getting into the genomics of their ecological studies. So that's something that interests me and I will go ahead with. And I've been talking to people at InStem who want to do genomic experiments. And there are points of possible contact with Sandeep Krishna and Mukund Thattai.

How did you decide to you come back to India?

I always knew I wanted to come back. It was just a matter of when. I feel more at home here, more relaxed. I mean Cambridge is great, but I'm never as relaxed there as I am here. It's just so good to be home. If I want to see my family I can just pack up and be in Chennai in 5 hours. That's perfect. I wasn't really homesick, it was not that extreme. It was more like, Cambridge, good, home - better.

Is there anything more you would like to say?

Bacteria are fun to work with! And important!