Stefano Allesina - US grants

Most living systems are comprised of complex networks of interactions; describing the structure of these networks and their dynamic properties is a major challenge for theoretical and empirical biologists. Food webs are paradigmatic of complex natural networks. Formed by species and their feeding relationships, they underlie the flow of energy through ecosystems and their responses to human-induced and environmental perturbations, including species' extinctions. Several simple models of food web structure exist, but these do not accurately represent the complexities present in real networks. The goals of this project are to develop new theory to extend simple models of food web structure, investigate biological mechanisms underlying this structure, examine the modularity of food webs, and consider networks quantitatively by incorporating interaction strengths among components. In particular, the researchers will address the effects of species extinctions on food web robustness and examine how stability at the level of an entire network results from instability of individual components. The resulting theory will provide a general framework to advance our understanding of food webs and diverse other complex biological networks.

The proposed research will contribute directly to education and outreach programs for grade school students, undergraduate and graduate students, and postdoctoral researchers. These activities will be coordinated through ongoing programs at the National Center for Ecological Analysis and Synthesis' 'Kids Do Ecology' program and through the University of Michigan. Web-based teaching modules and classroom presentations will introduce grade school students to ecology, complex systems, and mathematics by teaching them about many different types of biological networks. Seminars presented at Historically Minority Institutions and Minority Serving Institutions will introduce undergraduate students to careers in ecology and mathematics. Results from the project will be incorporated into graduate and undergraduate courses at University of Michigan.

Peer review is the main tool for scrutinizing scholarly publications, grant proposals and career advancements in science. However, the current peer review system is under severe strain, with consequences for the quality of science and the rapidity of dissemination of scientific results. Several studies have found that the current way of performing peer review can be inefficient, slow, and even biased. Almost every scientist has ideas on how to improve the system, but it is very difficult, if not impossible, to perform experiments to determine which measures are most effective. The project implements a simulation framework in which many ideas of how to improve the review process can be quantitatively tested.

Intellectual Merit The framework is built using agent-based modeling. Scientists, manuscripts and journals are digital agents and a peer review system emerges from their interaction. Multiple experiments can be run: for example, one proof-of-concept application shows how changing the way peer review is performed can dramatically alter the pace at which science is disseminated.

The research develops a full-fledged and open-source simulation software that allows to study alternatives to the current system.

Broader Impacts The proposed work is potentially transformative of the way science is carried out. This framework can be used to identify better and more efficient models for peer review, leading to profound changes on scientific publishing and funding. Also, if this exploratory research is successful, a new computational branch of sociology of science could emerge. Changing the way peer review is performed to favor faster and more efficient solutions could potentially have broad effects on the daily work of scientists, including more time for academic training and research, and less time spent revising and reformatting manuscripts and grant proposals. Favoring unbiased practices could enlarge the representation of minorities in science.

Ecology is about to face the data deluge that other biological disciplines have already experienced. With ecological data increasing rapidly in quality and size, new methods are needed to extract the most relevant biological information from massive data-sets. The objective of this project is to develop new mathematical, computational and statistical tools for the analysis of three ecological problems. First, when a species goes extinct, the impact reverberates through the ecological network, possibly causing the extinction of other species. A new method to predict such "secondary extinctions" will be developed. Second, the number and size of published ecological networks is increasing rapidly, making it possible to answer one of the oldest questions in ecology: how many species traits (e.g., body size, swimming speed, metabolic rate) does one need to measure to predict whether two species will interact? A new computational method, coupled with a large dataset will attempt to answer this question. Knowing which are the critical traits determining the possibility of interactions could find application in the study of invasive species. Third, the spatial structure of ecosystems mediates many ecological processes. A new method will be developed to measure the impact of spatial heterogeneity on the structure of ecological networks.

The development of these new tools require sophisticated methods, which are not typically included in the curriculum of biologists. The educational goal of the project is to train ecologists in the computational methods that will be needed to advance the discipline in the decades to come. Graduate students will learn how to automate the analysis of biological data, distribute computation over large computer clusters, organize data into relational databases, program in different languages, collaborate on data, code and manuscripts, automatically managing versions and conflicts, and pick the right tool for each task. Outreach activities will be provided through lectures and media interviews and with activities carried out in collaboration with local elementary schools and the Museum of Science and Industry.

Current research in biology is producing increasingly large and complex sets of data. These data could represent, for example: DNA sequences, images of the brain, or number of species in an ecosystem. In each case, unlocking the information within these big data sets requires sophisticated mathematical and computational approaches. The standard curriculum for graduate students in biological sciences was designed well before this data deluge. As a consequence, today's graduate students are not being adequately trained for their future careers. At the same time, there are growing concerns that scientists are sometimes unable to reproduce published findings. This inability often results from poor data analysis strategies. The future success of the US biological research mission hinges on training students to use data analysis approaches that are both rigorous and reproducible. This National Science Foundation Research Traineeship (NRT) award in the Innovations in Graduate Education (IGE) Track to the University of Chicago seeks to meet this need by developing a new and effective approach to the training of early stage graduate students in the quantitative analysis of biological data.

The overarching goal of this program is to teach students to critically evaluate quantitative analysis methods in the scientific literature, and to acquire good programming habits that support reproducibility and rigor in their own research. An interdisciplinary team of quantitative biologists will direct and lead the program, exposing students to the faculty that can advise them in future work. The training program begins with an intensive residential week-long boot camp that brings together students across diverse sub-fields of biology to promote teamwork and prepare them for interdisciplinary research. The boot camp includes introductory tutorials in computer programming, statistics, and modeling in modern biology, as well as more advanced tutorials in statistical approaches to large data sets and practical lessons in organizing and sharing code and data. The boot camp is capped off with a series of workshops in which students apply what they have learned to real biological data spanning a wide range of fields. A subsequent on-campus course builds on and reviews these concepts, and integrates training in rigor and reproducibility with concepts of responsible research. We hypothesize that this program will produce trainees who are well-prepared for the future scientific workforce. We will evaluate the impact of this intervention through quizzes, surveys, and targeted interviews. All teaching materials and data sets used in the workshops will be shared online so that any university can implement a similar training module on their own campus.

The NSF Research Traineeship (NRT) Program is designed to encourage the development and implementation of bold, new, potentially transformative models for STEM graduate education training. The Innovations in Graduate Education Track is dedicated solely to piloting, testing, and evaluating novel, innovative, and potentially transformative approaches to graduate education.

The relationship between species diversity (i.e., the number of species) of an ecological community and its productivity (e.g., the amount of biomass produced in a year) has been studied intensively in ecology, both from a theoretical point of view, and through some of the largest experiments ever conducted in the field. Understanding how biodiversity and productivity are related is essential for applications, such as the development of biofuels, as well as to predict the effects of species loss. The project aims to leverage the large amount of experimental data available to develop new statistical, mathematical and computational tools that can be used to predict the productivity of untested ecological communities, and to determine the contribution of shared evolutionary history to the functioning of ecosystems. Importantly, the same mathematical machinery developed for the study of ecological communities can be used to answer distant questions, such as the effect of combinations of antibiotics on the growth of bacteria. The work is complemented by the publication of software packages, lecture notes and teaching material, to facilitate community adoption and extension of the newly developed tools.

The project aims to develop a statistical framework to fit data stemming from Biodiversity-Ecosystem Functioning experiments that is consistent with the Generalized Lotka-Volterra model, one of the best-studied models for population dynamics. To keep the number of free parameters sufficiently small, interactions between any two species are modeled as a function of their shared evolutionary history. In turn, this makes it possible to test the effects of phylogeny on productivity, by including the whole structure of the evolutionary tree connecting all members of an ecological community, rather than summary statistics as in current practice. By moving beyond the linear regression approaches traditionally used to model these data, the project will bridge the gap between theory of population dynamics and experimental approaches. To further connect population dynamics with phylogenetic information, the project will develop models with explicit accounting for phylogeny, allowing research to investigate the effects of shared evolutionary history on the coexistence and stability of ecological communities.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.