Speaker
Description
Signaling pathways are an important part of cellular machinery responsible for processing biochemical signals and translating them into fate decisions such as proliferation, migration, or programmed cell death. Aberrant signal processing due to mutations leads to errors in fate determination, such as an increased rate of proliferation which is one of the hallmarks of cancer. The process of cell fate determination depends on the dynamics rather than the steady state of signaling components, thus to dissect network wiring that orchestrates fates, it is crucial to monitor signaling over time and subject the network to genetic and chemical perturbations.
In the Cellular Dynamics Lab at the University of Bern, we have established a number of experimental techniques based on time-lapse microscopy to measure and perturb the dynamics of signaling pathways in individual cells over time. Such experiments generate vast amount of images, a problem exacerbated by a recent shift towards the study of 3D organoids that mimic physiological conditions better than conventional 2D cell cultures. Storage and the analysis of images, extraction of relevant features from time-series data, visualization of multidimensional feature spaces, and fitting of mechanistic models to data all pose a significant challenge to the field.
Our overarching aim is to understand oncogenic signaling and sources of drug resistance, and pave the way for personalized approaches to cancer treatment. In my talk, I will present a pipeline realized in our group and roadblocks we have stumbled upon during the implementation. I will demonstrate how efficient data analytic approaches coupled to modern visualization techniques allow us to quickly gain insight into the measurements and plan the next-best experiment. I will focus on a suite of homegrown tools for spectral analysis of time series data as well as approaches based on machine learning.