Dr. Artem Kaznatcheev, University of Pennsylvania
Abstract: We can view the problem faced by evolving populations as a game between the population and the environment with the distribution of different phenotypes as the strategy and the evolutionary dynamics as specifying the strategy update rule. This allows us to make global conclusions about evolution without knowing all the reductive details of population structure. I want to provide two examples of how this is useful for both experiment and theory.
For an experimental example, I will introduce you to the game assay for measuring the ecological interactions of evolving populations. Recently, we applied this assay to measure the games played by sensitive vs resistant non-small cell lung cancer. We saw that the games played by these cancer cells are not only quantitatively different between different environments, but drug and fibroblasts qualitatively switch the type of game from Leader to Deadlock. Focusing on frequency dependent fitness also reveals a surprising absence of a cost of resistance in non-small cell lung cancer.
For a theoretical example, I will examine how the combinatorial structure of some static fitness landscapes can produce a computational constraint that prevents evolution from finding any local fitness optima. On the hardest landscapes, no evolutionary dynamics can find a local fitness optimum quickly, thus allowing for open-ended evolution. Knowing this computational constraint allows us to use the tools of theoretical computer science and combinatorial optimization to understand maladaptation and characterize the fitness landscapes that we expect to see in nature.
Reference:
Kaznatcheev, A., Peacock, J., Basanta, D., Marusyk, A., & Scott, J. G. (2019). Fibroblasts and Alectinib switch the evolutionary games played by non-small cell lung cancer. Nature Ecology and Evolution, 3: 450-456.
Kaznatcheev, A. (2019). Computational complexity as an ultimate constraint on evolution. Genetics, 212(1), 245-265.
Bio: Dr. Artem Kaznatcheev is a James S. McDonnell Foundation independent postdoctoral fellow in dynamic and multi-scale systems. He is currently hosted by the University of Pennsylvania Department of Biology. He received his DPhil in Computer Science from the University of Oxford with the support of the Cleveland Clinic Department of Translational Hematology and Oncology Research. Before this, he was at the Moffitt Cancer Center and McGill University. His current research studies the dynamics of cancer and other empirical systems to uncover the algorithms of evolution, understand the computational complexity of the natural world, and use that knowledge to help treat cancer.