Topological Methods for Uncovering Hidden Structure in Neural Activity and Connectivity
Abstract
The brain is a vast collection of interconnected neural circuits whose computations arc often accomplished by local recurrent or feedforward networks. Understanding the structure of these local circuits, however, is still very much a work in progress. Currently, experimental and computationnl technology in neuroscience are making rapid progress towards amassing large databases of the activity patterns and connection patterns between neurons. How to interpret this data, and what it can tell us about the structure of neural circuits, is still very much unclear. The biggest limitation is perhaps the lack of theoretical frameworks for characterizing meaningful structure in local neural circuits. The central goals of this proposal arc to: (i) develop computational topology tools and mathematical theory for detecting "hidden" structure in neural data, and (ii) to use these tools to test long.standing hypotheses about the organization of local networks in the brain. The scientific basis for this proposal centers on the central hypothesis that there arc robust signatures of underlying network structure that can be detected using novel computational topology tools and mathematical theory when applied to neural activity and connectivity data. The theory will be both tested and guided by the analysis of multi-unit clcctrophysiological recordings of hippocampal and cortical neurons that will be provided by collaborators. The investigators will apply their theory to correlation matrices from ncurnl data and derive connectivity patterns under various conditions, i.e. activity, sleep or pharmacological intervention. Further development of the mathematical methods employed will then be used to test additional theories about network structure within numerous brain regions. The proposed research provides a bridge between distant disciplines: algebra and topology on the one hand, and experimental systems neuroscience on the other. It will introduce new mathcmntical techniques and perspectives to neuroscientists, and could entice pure mathematicians to work on questions relevant to biology.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 12, 2017
- Source ID
- W911NF1510084
Entities
People
- Vladimir Itskov
Organizations
- Army Contracting Command
- Defense Advanced Research Projects Agency
- Pennsylvania State University