URofL:EN: Does re-wilding lead to re-wiring of gene expression and species interaction networks?

Evolution is at the root of many socially important problems and solutions. Infectious diseases evolve resistance to drugs or escape vaccines. Tumors evolve to exploit their host’s body and tolerate chemotherapy.

Agricultural pests evolve resistance to pesticides. Threatened species must adapt to environmental variability and land use change or else risk extinction. To solve such problems rooted in evolution, biologists need to be able to forecast future evolutionary change.

Although biologists have a deep understanding of the forces that cause adaptive evolution, evolutionary forecasting remains a major challenge. In the laboratory, if we subject initially similar populations to the same environmental stress, they often evolve the same solutions using the same genes, demonstrating that forecasts are possible. However, sometimes experimental evolution leads to entirely different outcomes.

Why is evolution predictable in some situations, but not others? One hypothesis proposes that evolution is more predictable for traits built by simple genetic networks (few interacting genes) than for complex genetic networks (many genes working synergistically). This project seeks to test this hypothesis.

As part of an ecological restoration, the research team reintroduced native stickleback fish into 8 recently-fishless lakes in Alaska, beginning the largest evolution experiment yet attempted in a natural setting. Tracking evolution in these lakes in coming generations will let the investigators test whether genetic networks evolve, and whether simpler genetic networks evolve more predictably.

The results of this experiment will yield new tools for forecasting evolutionary change using gene network data, which can then be applied to evolutionary problems of public interest. To achieve this aim the team must also develop new tools in computer science and statistics to analyze how networks change through time. These new computational tools will have broad applicability to measure changes in any type of network of concern to impacts national health and security.

In 2019, researchers reintroduced native stickleback fish into 8 lakes where they had been extirpated by invasive species. The team will track the evolution of these experimentally-created populations as they adapt to their new habitats, annually monitoring diet, morphology, parasites and microbiota, immunity, genotypes, and transcriptomes in every source and experimental population in each year.

To analyze these data they will develop new computational and mathematical models in network theory to measure temporally changing network structures. Using these new tools they will test for plastic and evolutionary changes in genetic, transcriptomic, and ecological networks in the focal populations. Comparing network changes between experimental populations they can test the predictability of network structure evolution, and its relation to changing ecological networks.

This convergent research deeply integrates biology (genetics, evolution, ecology, microbiology) with computer science and statistics to promote new advances at the interface of life and data science. To educate the public, the team will work with the Big Biology Podcast to produce six shows about the intersection of network data science, evolution, genetics, and conservation.

Each podcast will be supplemented with ‘virtual field trip” videos, interviews, and lesson plans for K-12 biology, math, and computer science classes.

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.