Speakers
Description
Data collection for biological systems has undergone a revolution over the past few decades with electronic trackers for wildlife populations giving us minute to minute information about their locations and fluorescence data at the cellular level tracking the number and location of proteins. On the theory side, we have well-known population models for the interaction of wildlife and also for intracellular interactions. A difficulty in practice is the divide between the collected data and the idealized models used to represent the relevant biological systems, models which can bring real insight into the biological mechanisms at play, but with the need to be calibrated to the data. The realities of experimental observation can include data on a coarser scale than the one of interest for the model, data collected at uneven time steps, or observational systems that can actually affect the system of interest, such as with laser traps tracking molecular motors. Some of this separation can be handled by improved computational tools, such as improvements in stochastic simulation and statistical tools that rely on stochastic simulation. However, unifying approaches would aid in linking the vagaries of experimental data to theoretical model; this mini-symposium will present some approaches to handling these complexities by bringing together researchers with complimentary perspectives across statistics, stochastic modeling, and machine learning.
