It’s clear that microbes play a crucial role in practically every aspect of ecosystems globally. From the deepest, most remote and unexplored regions of the ocean, to the human oral cavity, there are diverse microbial assemblages driving Earth’s biogeochemical cycles.
Back in March, I was fortunate enough to catch a presentation Dr. A. Murat Eren (Meren) gave at Susan Holmes’ lab at Stanford. Meren is an assistant professor at the University of Chicago, and was just a few weeks ago was named an MBL fellow. His lab focuses on a range of interesting topics that generally fall under the broader umbrella of microbial metagenomics.
I was intrigued by his research, which has included projects concerning the microbiota associated with bacterial vaginosis, oligotyping to differentiate microbial taxa, and even an assessment of the contamination in the recently published tardigrade genome. The talk was great, and I was particularly struck by one point he emphasized: without the appropriate approach to analyzing and visualizing data, you can’t fully understand what your genomes (or metagenomes) are telling you.
We currently have affordable (getting cheaper daily) as well as accessible tools to study microbial communities, allowing us to (begin to) understand their role in the environment. A plethora of sequence data is now constantly accumulating. The challenge we now face has begun to shift from obtaining genome / metagenome sequences (although that can still be tricky) to figuring out how to analyze and interpret the data properly. This is especially difficult for scientists who have plenty of sequence data but lack the appropriate bioinformatics skills.
Meren has also recognized this obstacle, and in response created anvi’o (an analysis & visualization platform for ‘omics data), which is a fantastic software for the analysis of metagenomic data. The most recent release of anvi’o promises to continue to provide the means to analyze and visualize large data sets, with the goal of allowing for a clear interpretation of the results. I recently attended ECOGEO, a workshop on interpreting ‘omics data, and Meren walked us through the tutorial, which you can attempt yourself, if you’re interested.
I recently had a chance to talk with Meren about his research and work habits, he has an interesting eclectic background, and I wondered what led him eventually down the metagenomic path. Some questions were borrowed from our “How a Molecular Ecologist Works” series.
If you could pick one word that describes how you work what would it be?
Well, I’m going back and forth between ‘passionately’ or ‘obsessively’, because I kind of spend my life in the lab…uh, you know. I guess I should say ‘a bit obsessively’, since I feel that’s how I work. I really, really enjoy what I’m doing. I wake up with excitement every morning, take a quick shower and get back to the lab as soon as possible. It’s a luxury to enjoy what you’re doing, and I think being obsessive is a small price to pay for it.
Other than computer and phone, what other electronic thing do you need to have around you?
Headphones. Noise canceling headphones. Almost constantly. Especially when I’m in an airport, airplane, bus, or train. I sometimes use them even when I’m walking home.
Where do you work best?
There are two places where I work best: one is my office, and the other is an airport. Because I’m a bit paranoid, I usually get to airports way earlier than the actual departure time, sometimes even 5 – 6 hours earlier, and then I just sit in a corner and work. I usually try not to have an internet connection. No one knows me, no one stays long enough to judge me or ask questions, because they have a plane to catch and all. I do a lot of work very efficiently, and I write a lot of code in airports without any distractions.
Do you have a favorite airport?
Well, every airport is special, I really like small ones, big ones, ones that I get lost in, each one of them is great. I learned to enjoy the airport in Honolulu, because once my flight was delayed more than 18 hours.
Do you have a personal philosophy that you kept in mind when you were setting up your own lab?
That is difficult to answer, but yes. I just put together my research interests page (because I thought I needed one at this point), and it starts with a quote from Richard Feynman, that says “study hard what interests you the most, in the most undisciplined, irreverent, and original manner possible,” which really resonates with how I think things should be.
Apart from that, I am not a good manager. I never intended to be one. I imagine myself telling individuals who might want to work with me in the near future that managing their time will never be my concern, and that they should only come into the lab when they really want to work. If there’s some fun to be had outside, they should not even bother coming in, in fact. I would like to provide complete freedom to people I work with, and know that they are not doing anything they don’t want to. Because that is what science is all about, in my opinion.
“Study hard what interests you the most, in the most undisciplined, irreverent, and original manner possible” – Richard Feynman
Labs are not companies, students and postdocs are not workers, and PIs can get away with not being managers. So I want to make sure that intellectual freedom is retained at every step. Of course, I am just an assistant professor, and who knows what will happen in ten years if I can stick around. Maybe I would become a micromanger.
What are you reading?
Reading? Embarrassingly nothing, I don’t read much, I read blogs and papers, and I usually have a hard time reading other things. I enjoy a very small number of authors, which is not something I’m proud of. Whenever I start reading a novel, it puts me in a mindset that makes me remember what I could do in the code base or in an article I’m working on, or whatever. So I’ll read like 10 sentences and then want to go back to work. And I blame the authors for that. Maybe in ten years I will become a micromanager, AND a good reader.
What advice would you give somebody who’s just embarking on graduate school?
I wrote a very long block post in Turkish about my journey in graduate school. I had written in that post that the only advice that I really keep giving people with great confidence is “don’t take any advice from anyone”. Listen, sure, but don’t care about what they are suggesting. Every journey is extremely unique, and no on can know what one needs.
The fact that there are too many people on this planet forces us to create bins and boilerplate recipes: “ways to become an engineer”, “ways to become a scientist”, “ways to become this and that”, etc. Even within the domain of science, we need to make sense of all the scientists that are out there – to quantify their output, in order to give them positions, and promotions. Everything is being standardized in one way or another, even the advice that is being given to students. But maybe students should not really care too much about the advice they are given, and should instead feel completely free to follow their gut feelings, they should only do things they want to do.
I constantly try to keep the portion of things I have to do overlapping with the things I want to do.
Of course, when I say this, it sounds a bit superficial. But, I think there is a constant struggle between things you want to do, and things you have to do. Not everything is black and white, but I constantly try to keep the portion of things I have to do overlapping with the things I want to do.
I don’t see why anyone would take anyone’s advice, really, people give terrible advice! They don’t even know you, they don’t know what is going to happen. Life is a stretch of serendipitous events stitched together. If you could go back in time, and ask me where would I like to see myself in 10 years, and what would I need to do to achieve that, my responses would have been very embarrassing today. Fifteen years ago, I was on my way to becoming a musician without any intention of becoming a scientist. So, maybe some people need advice, but I really don’t see why would anyone take anyone’s advice. At least I’m not interested in giving any advice beyond the things I mentioned here.
What inspired you back in Turkey to do computer science, was there a moment?
Yes, there was one moment, during which I decided to become a computer engineer. It was right before university. At the time, I was planning on becoming an industrial engineer. I wanted to build stuff that would streamline people’s lives and make it possible for them to do things more efficiently. I was getting ready to take the nationwide general exam get into a department for industrial engineering. I met this guy who knew a lot about computers, and he installed this game on our little network. This is the first time I am seeing a computer in my life, and in this game, I play with other people over the network – it was a first person shooter game, the first Quake – and it just blew my mind. Looking at the computer screen, I could see him moving around, and he’s just sitting right there – this was such a magical moment. I asked him, in fact, I asked him literally, “what should I do in life so that it would be okay if I spent my entire time in front of a computer playing this game?” He said, “become a computer engineer, no one will judge you if you spend your entire life in front of a computer.”
So I decided to go for computer engineering…nothing much inspirational here. Just a game. I don’t tell people this, because, you know, I don’t want them to think that I, well I don’t want them to know the truth I guess.
>In a profile you have online at the MBL, you mentioned there are many stories one image can inspire. Do you think it’s necessary to have an eye for design to communicate science?
I don’t know how much of my experience in photography or how it contributes to the way I look at things, or if there was a reason for me to be interested in photography in the first place. But I know that it’s extremely critical to be able to visualize things. I believe that if you can’t visualize something, you can’t claim you truly understand it.
I try to set visualization as a priority in my studies. Even with the oligotyping pipeline, I had tried to make sure everything is visualized so that people could get their hands dirty with the data by seeing what they are doing. Some people don’t put as much emphasis on visualization, but I think it’s quite essential.
Ernst Haeckel. He was drawing things at a time when there were no TVs, when people couldn’t take photographs. He was drawing living organisms from diatoms and bats to sponges and plants. He was visualizing life in a sense to communicate the diversity and beauty of it to people. We evolved with this visual curiosity about life itself. We understood things mostly by seeing them for a very long time, and there has been a huge transformation in the way we understand things now.
Most of the knowledge we generated about living organisms came from simple observations. The number of legs, the color and shape of fur, scales, flowers etc. Now we generate sequencing data, and run into trouble because we don’t have great ways to visualize it. It’s something new, and we’re struggling with it. That’s why I think exploring different ways to visualize data is really important. The data visualization language that is emerging is essential to communicate many things to everyone.
It is hard to understand molecular evolution. Why? We see evolution happening very clearly, but we haven’t been able to communicate it as clearly to people who do not have time to study the concepts that allow us to understand it. When you can’t visualize things, they become very hard to communicate.
When the way we acquire information changes, it should be followed by changes in the way we communicate it.
For instance, in our recent publication on the tardigrade genome, we used anvi’o to visualize the contamination everyone has been talking about: we created one holistic display that shows the RNA data, multiple sequencing libraries from multiple labs, and where the horizontally transferred genes were. One single display that shows every aspect of data. It’s still not perfect, as it requires an understanding of things such as mapping, clustering, etc., but nevertheless, most would agree that it was better than tables with numbers and 2D ordinations that reduce our ability to see while also reducing complexity.
We like to spend time talking about visualization in our lab. When the way we acquire information changes, it should be followed by changes in the way we communicate it.
Papers referenced in this article:
Delmont, Tom O., and A. Murat Eren. “Identifying contamination with advanced visualization and analysis practices: metagenomic approaches for eukaryotic genome assemblies.” PeerJ 4 (2016): e1839.
Eren, A. Murat, et al. “Anvi’o: an advanced analysis and visualization platform for ‘omics data.” PeerJ 3 (2015): e1319.
Eren, A. Murat, et al. “Exploring the diversity of Gardnerella vaginalis in the genitourinary tract microbiota of monogamous couples through subtle nucleotide variation.” PloS one 6.10 (2011): e26732.