The National Science Foundation (NSF) and the Simons Foundation have awarded a grant to Harvard scientists to create a research center aimed at bringing biologists and mathematicians together to answer some of the central questions about living systems.
Dubbed the NSF-Simons Center for Mathematical and Statistical Analysis of Biology at Harvard University, the center will be one of the cornerstones of the recently announced Quantitative Biology Initiative, which is aimed at integrating mathematical, statistical, and engineering approaches with biology.
Led by Andrew Murray, the Herchel Smith Professor of Molecular Genetics and director of the John Harvard Distinguished Science Fellowship Program, the center will focus on three fundamental questions: how molecular networks within cells process information to make decisions; how sophisticated structures, like the outer coating of viruses that segregates chromosomes or the structure of embryos and organs, can assemble without blueprints or explicit instructions; and how organisms change their structure and behavior in response to changes in their environments over time ranging from minutes to millennia.
The other primary investigators on the grant are Cassandra Extavour, professor of organismic and evolutionary biology and of molecular and cellular biology; L. Mahadevan, Lola England de Valpine Professor of Applied Mathematics, of Organismic and Evolutionary Biology, and of Physics; Johan Paulsson, professor of systems biology; and Sharad Ramanathan, Llura and Gordon Gund Professor of Neurosciences and of Molecular and Cellular Biology and professor of applied physics and stem cell and regenerative biology.
“The Quantitative Biology Initiative was founded to tackle the hardest problems in biology,” said Vinothan Manoharan, the Wagner Family Professor of Chemical Engineering and professor of physics, who will lead the initiative along with Ramanathan. “Mathematicians and statisticians are essential to that goal. We’re excited about the new NSF-Simons Center because it will support exactly the kind of collaboration we need, both within Harvard and beyond.”
“These fundamental challenges in life sciences provide ample opportunities not only for multidisciplinary collaborations, but also for advancing methodologies and theory in data science itself,” said Xiao-Li Meng, the Whipple V.N. Jones Professor of Statistics. “Having served as the GSAS [Graduate School of Arts and Sciences] dean, I also particularly appreciate Murray and his colleagues’ strong emphasis on creating a spectacular environment for students and young researchers to do spectacular science. Like my colleagues, I am grateful to NSF and SF [the Simons Foundation] for their trust in Harvard’s ability to lead at both the research and education front. As a statistician, I treasure this exciting opportunity to learn from, and assist, Murray and his colleagues as they bring new life and new science into life sciences.”
“Many people think of math as proofs or calculations, but it’s so much more,” said Juan Meza, director of NSF’s Division of Mathematical Sciences. “It’s really a powerful tool that helps us study nature. By applying mathematics to the complex processes that underlie biology, we may finally be able to understand exactly how something as seemingly simple as a sequence of genes led to the broad diversity of organisms we see today.”
The Harvard center is one of four the NSF and Simons Foundation will fund. Each will receive $10 million to support a five-year project. The NSF will distribute $20 million in federal funds that will be matched by $20 million from the Simons Foundation.
The four centers, selected through NSF’s merit review process and an additional review by the Simons Foundation, are headquartered at Harvard, the Georgia Institute of Technology, Northwestern University, and the University of California, Irvine. Their leaders will come together annually to share progress in meetings at the Simons Foundation in New York City.
“We very much look forward to collaborating with NSF on this major public-private initiative to develop novel mathematical approaches to problems in biology,” said Yuri Tschinkel, director for mathematics and physical sciences at the Simons Foundation.
“This award from the National Science Foundation and the Simons Foundation fits perfectly with the aims of the Quantitative Biology Initiative, and represents a tremendous opportunity to catalyze collaborations between the life and the mathematical sciences at Harvard,” said Jeremy Bloxham, FAS dean of science and Mallinckrodt Professor of Geophysics. “Harvard is at the very forefront of a new wave in the life sciences of mathematically and physically based investigation of deep, fundamental problems.”
Addressing those fundamental questions, Murray said, will involve faculty working at the new center identifying problems that can help shed light on core biological processes.
“What we want to do for each of those areas is to pick not just one problem, but two or three where we think they’re tractable, and where we think what we learn from one problem can inform others,” Murray said.
As an example, he pointed toward questions of how cells self-assemble and self-organize.
“The simplest example is the assembly of viruses,” he said. “That’s a jigsaw puzzle that puts itself together. A more complex problem would be the assembly of the spindle that segregates chromosomes when cells divide. That’s a problem of self-organization, because there are many types of pieces that interact with each other and rules that tell you how they interact with each other … to produce a structure that is much bigger and more dynamic than any of the individual pieces.
“At the next level up there are problems in embryogenesis,” he continued. “Those are questions like how do you mathematically map the shape of a fly wing in different species, or how are intestines coiled differently in different birds? Those questions deal with how cells interact to make tissues and how small differences in the interactions give rise to a variety of shapes, each adapted to a specific function.”
The goal of the center, Murray said, is to bring mathematicians and biologists together and enable each to leverage the other’s expertise in understanding those problems.
“One of the things that’s changed in the last 20 years, but especially quickly over the last five to 10 years, is the scale at which we can generate data,” Murray said. “What we need is some sort of statistical machinery … for doing a better job of looking at large quantities of data and arguing about which parts of the data matter the most, and therefore where to look for clues that reveal the rules that produce complex and subtle biological functions.”
The hope, Murray said, is that researchers will be able to identify the critical variables that play important roles in living systems and use them to develop new mathematical models for predicting the behavior of biological systems.
“Human cells have 20,000 different genes … and in principle, all of those thousands of players have something to do with what cells do,” Murray said. “We cannot in real time measure more than a small fraction of those things, so the question is twofold: What are the things we should measure so we have the best ability to predict and understand these systems, and, given that our measurements are noisy and incomplete, what is the best way of making models that are usefully predictive?
“The ultimate question is how those measu_rements can be usefully coupled with formal mathematical models of how those processes work,” he added. “In particular, the hope is that this interaction will not just help us understand biology, but that it will also call into being new forms or even new branches of mathematics.”
By combining those new mathematical techniques with experiment and statistical inference, Murray said, the hope is that researchers will be able to build and constantly refine new, more accurate models of biological systems.
“For some problems, we may already know enough to start with mathematical models that make predictions that we can test by experiment. For others, the best approach will be to collect a lot of data and use statistical inference to identify the features that best correlate with and explain the biology.” he said. “There are all sorts of new methods for doing a better job of looking at very large quantities of data and applying mathematical and statistical methods that will suggest which parts of the data matter the most, and therefore where you should be looking to create models and design the next generation of experiments.”
The new center is particularly well-situated at Harvard, Murray said, not only because of the strength of research in areas like mathematics, statistics, engineering, and biology, but because of its long history of faculty working across disciplinary boundaries.
“We have strengths in each of those areas, but we also have two sorts of track records,” Murray said. “One is in having people who are already at Harvard and work in those different areas interact and talk with each other, and the other is that we’ve set up programs that have done a good job of recruiting extremely bright young people and giving them the combination of enough independence to do things that are original, and sometimes revolutionary, and enough mentorship to give them help when they need it.”
That training and mentorship will be a key part of the center, Murray said.
Already, plans are being formulated for a program to develop collaborations with scientists outside Harvard through a visiting researcher program. Also in the works is a series of annual workshops, at which at least half the participants will come from outside Harvard, and at least two-thirds will be students and postdoctoral fellows, Murray said.
“We want to train the very best young people and send them forth to improve national and international capacity to solve problems like this,” Murray said. “And while we train and help them, they will do spectacular science.”