Bridging Function and Anatomy in Wide-field Calcium Imaging Data - Masters Project

alternate text
Multiple experiments carried out in the Neuronal Circuit Dynamics Lab acquire in vivo full brain wide-field optical recordings of mouse cortex activity during behaviour. The measurements are mesoscopic, i.e. cover many cortex regions but without cellular resolution. Each recorded pixel carries signal averaged over multiple neurons. Such recordings can uncover context- and task-specific activation patterns in mouse brains. However, the spatiotemporal characteristics of wide-field signals are not fully understood. We seek a talented and motivated student to help analyze existing data sets and study the following questions: (1) She or he will use unsupervised machine learning tools to identify effective functional areas, as well as construct mappings between functional data and anatomical areas. (2) The student will use information-theoretic metrics to study the information content in wide-field recordings, as well as identify the information overlap between pixels, functional and anatomical areas. (3) It has been hypothesized that wide-field signals in the cortex may loosely follow local propagation laws similar to those of water waves. The student would propose a few model propagation laws, fit them to the data, and then assess and report on the plausibility of this hypothesis.

Research Group:
Laboratory of Neural Circuit Dynamics (Prof. Fritjof Helmchen) at the Brain Research Institute, University of Zurich (Irchel Campus, building 55).

Requirements

Experience in programming (Python/Matlab), statistics and/or machine learning.

Contact

Philipp Bethge (bethge@hifo.uzh.ch)