An ability to combine computation and experimentation to discover new insights about the regulation of gene expression and assembly of protein complexes recently won Sarah Teichmann a 2015 European Molecular Biology Organization (EMBO) Gold Medal. The award recognizes outstanding scientific achievements from young researchers in Europe in the field of molecular biology.
At 40, Teichmann holds a joint appointment with the European Bioinformatics Institute (EMBL-EBI) and the Wellcome Trust Sanger Institute on the Wellcome Trust Genome Campus near Cambridge in the United Kingdom. There, she leads a systems biology group of 17 researchers who use computational genomics to shed light on how molecules work together within cells. She is also one of the founders of the new Sanger Institute–EMBL-EBI Single Cell Genomics Centre.
The beauty of a ‘wet–dry’ group is that you have the ability to interpret your own experimental data computationally and, conversely, to test computational predictions experimentally.
“Sarah’s expertise in bioinformatics seems infinitely adaptable and is coupled with skills for directing exquisitely controlled laboratory-based experiments,” Veronica van Heyningen, an EMBO member and geneticist at University College London (UCL) in the United Kingdom, stated in the award announcement. “Such broad-spectrum abilities are rare.”
Science Careers asked Teichmann how she gained her skills and abilities and what doors they opened to her. This interview has been edited for clarity and brevity.
Q: What came first, informatics or biology?
A: Biology came first, but informatics has always been at the heart of my research. I started programing during my bachelor's research project at the University of Cambridge, where I was optimizing the resolution of nuclear magnetic resonance structures of a particular protein, Ras.
But I really felt inspired by computational biology when, still as an undergraduate, I read a 1992 commentary by Cyrus Chothia. In the paper, he made a simple calculation of the likely number of protein families existing in nature. I felt excited about discovering general principles in biology and became convinced that computational biology would provide the necessary tools: By mining large datasets, one can identify overarching trends.
By the time I started looking for a Ph.D., in the mid-1990s, datasets had become large enough to try to gain a global view of the entire universe of protein structures—and the same was true for protein and gene sequences. This realization convinced me that I should take the leap into a computational biology Ph.D. project. So I went to work with Chothia at the Medical Research Council Laboratory of Molecular Biology (MRC-LMB) in Cambridge, exploring protein families and the domain organization of proteins in the first completely sequenced genomes.
Q: Was computational biology a risky career choice?
A: Yes. It was an unusual choice, but I never looked back, even though at one point I came to feel that computational biology and bioinformatics were viewed as eccentric and unorthodox. My Ph.D. mentor exuded such unwavering optimism and confidence, however, that it made his lab a great place to work. Altogether, during my Ph.D., I published 10 papers.
Then, when I moved to UCL to do a postdoc, I found in Janet Thornton’s lab a reassuringly huge bioinformatics group, embedded into a broader university community that embraced computational biology wholeheartedly.
Q: In 2001, less than 2 years after earning your Ph.D., you started your own research group at the MRC-LMB. How did that happen?
A: Toward the end of my Ph.D., based on my dissertation, my Cambridge college awarded me a research fellowship to cover my salary for four out of the next 6 years. Soon after, I received a letter from the MRC-LMB saying that I could use the fellowship to work there independently, provided that I first went away to do a postdoc.
Then, during my postdoc in the Thornton group, I was approached for an assistant professorship in the United States. I mentioned this to the head of the MRC-LMB at that time, whom I met by chance at a bus stop one day when I was back in Cambridge giving a talk. The MRC-LMB went on to upgrade their offer to a programme-leader track position, which is essentially a tenure-track job.
I took the MRC-LMB job because, in combination with my research fellowship, it was an attractive package that included core funding. This position gave me a running start in my life as an independent research group leader. One thing this taught me is that you are automatically in a stronger position if you have several employment options.
Q: What is your research about today?
A: In our work about protein complexes, we ask questions like, “How does the three dimensional architecture of this protein complex affect how its component proteins find their correct place?” More broadly, we are now developing what we call a “periodic table” of all existing protein complexes, providing a framework for topologies of small- and medium-sized complexes.
Regarding gene regulation, having worked on the Wellcome Trust Genome Campus for the past two and a half years, I’ve absorbed approaches that harness human genetic variation and comparative genomics to make functional inferences. These methods now allow us to address how switching genes on and off affects immunity. It’s a lot of fun to apply genomics and bioinformatics to immunological data sets, because it’s relatively unexplored territory. In particular, we are really excited about the huge potential of single-cell transcriptomics for studying immune response dynamics.
Q: What are the pros and cons of combining computational and experimental approaches?
A: The beauty of a “wet–dry” group is that you have the ability to interpret your own experimental data computationally and, conversely, to test computational predictions experimentally. The challenge, however, is maintaining the right balance and harnessing the potential of both sides of the group, because the two ways of working are so different.
Q: How did you learn these two approaches?
A: Computational biology requires both a deep knowledge of biology and high comfort with maths and programming. It helped that, for my first degree, I had enrolled in the natural science degree program at the University of Cambridge, which allowed me to take maths and physics as well as chemistry and biology at a high level. I built on this later on through short courses and learning-by-doing.
Today, I’d however recommend more structured courses in algorithms and programming.
Q: Do you think that informatics has become a must-have skill for all molecular biologists?
A: Yes. It is an invaluable skill, and I believe that it will become even more so as technologies that generate large data sets, such as genomics and imaging, become more and more pervasive in biology and medicine.
Q: Looking back on your career, what do you think have been the key factors in your success?
A: Simple hard work and tenacity, coupled with self-belief. I am grateful for the support many senior people have given me along the way. I also have a wonderful group, and I am very proud that many of our former members now are leading scientists in their own right. They all stay in touch, which is great.
Q: Have you ever found balancing your professional and personal life difficult?
A: Yes. I have two young daughters, and some Ph.D. students and postdocs have had children while working in my group. These experiences have made me sensitive to the fact that our society is a long way from empowering both men and women to balance their professional and family lives. I believe that there should be financial and cultural support for both men and women to work shorter hours for several months after the birth of an infant. Ultimately, fair and equal treatment would decrease unconscious bias in all of us.
Q: Is there anything you wish you had done differently?
A: I made career choices mostly based on the package that was offered and the colleagues I would have. In retrospect, I think it is also important to consider bigger questions such as whether the scientific strategy and values of an institute are aligned with your own.
Over the years, I’ve also learned to shed my introverted nature and appreciate how important it is to talk to people as often and as openly as possible. This makes interactions within the group and with collaborators and managers much smoother.
Q: What advice would you have for young scientists?
A: Research is about having fun and being creative, but it is more of a marathon than a sprint. So first and foremost, pick a research question you care deeply about, even if it’s not the flavor of the month.
Second, choose the right people to work with, both scientifically and personally. I made sure that my Ph.D. and postdoc mentors were people who invested in their group members. As for my group members, I recruit them based on creativity, skill, and commitment. What I am aiming to achieve is a diverse group that works well together, sparks cool ideas, and is highly productive.