AMS Special Lecture | Assistant Professor Kevin Flores

Location: Hodson 213

When: Tuesday, March 26 at 1:30 pm

Title: Data-driven selection of models for collective cell migration

Abstract: Collective cell migration plays an important role in a diverse range of biological processes, including wound healing and cancer. Mathematical modeling is a powerful tool that can be used to test hypotheses about which subcellular and inter-cellular mechanisms drive collective cell migration, yet the validation of such models with experimental data remains a challenge due to data sparsity, noise, and heterogeneity. I will introduce several mathematical, statistical, and machine learning tools that my research group developed to overcome these challenges. First, I will discuss biologically-informed neural networks, a machine learning-based method that uses spatiotemporal data to discover partial differential equation models that obey conservation laws and biological constraints. Next, I will discuss a statistical framework to estimate inter-cellular heterogeneity as parametric variability in density-dependent partial differential equation models. Lastly, I will discuss our recent efforts to use methods from topological data analysis to infer parameters for agent-based models from microscopy movies of high-density cell populations displaying collective motion patterns.
 
Bio: Dr. Kevin Flores is an Associate Professor in the Department of Mathematics and is a member of the Center for Research in Scientific Computation and the Precision Medicine cluster at North Carolina State University. His research focuses on mathematical and computational modeling in precision medicine and systems biology. Dr. Flores has published 48 manuscripts and edited 3 special issues in mathematical biology journals, and his research has been supported by several grants from the National Science Foundation and the National Institutes of Health.

Zoom link: https://wse.zoom.us/j/91019141436