2015 — 2018 |
Beer, Randall [⬀] Izquierdo, Eduardo |
N/AActivity Code Description: No activity code was retrieved: click on the grant title for more information |
Ri: Small: An Ensemble of Neuromechanical Models of C. Elegans Locomotion
One of the central challenges of neuroscience is understanding how animals operate as integrated wholes, that is, how their neural activity, working in concert with the properties of their bodies and environments, coordinates their behavior. When faced with such a daunting challenge, it makes sense to first study simpler instances of the general problem of interest. In this regard, the nematode worm Caenorhabditis elegans is a uniquely qualified target for investigating the operation of integrated brain-body-environment systems. It is the only animal for which the connectivity of its entire nervous system is known. In fact, the anatomical structure of its entire body has been characterized to the level of individual cells. In addition, its genome has been completely sequenced and its entire developmental lineage has been characterized, from fertilized egg to adult animal. The goal of this project is to draw upon these many resources in order to construct integrated brain-body-environment models of C. elegans locomotion, a behavior that serves as the foundation for all other behaviors that this animal exhibits.
Although the entire connectome of C. elegans is known, detailed knowledge of the electrophysiology of its nervous system is far less complete. For this reason, the approach has two components: constrained stochastic optimization and ensemble analysis. The investigators will construct computational models constrained by the known connectivity of the C. elegans ventral cord circuitry, the known layout of its body wall musculature and the partially-known electrophysiological properties of the neurons involved. They will then apply a stochastic optimization technique (evolutionary algorithms) in order to find values of the unknown electrophysiological parameters of this model that maximize a measure of locomotion performance. In general, different optimizations will result in different values for these parameters, all of which are consistent with the known experimental constraints. Thus, the object of study is not an individual model, but rather the entire ensemble of models that result from repeated optimizations. A detailed study of this ensemble will suggest specific new experiments whose results can then be used to further constrain future optimizations. The particular focus of these modeling efforts is on understanding the relative roles that proprioceptive feedback from body stretch receptors and the intrinsic dynamics of ventral cord circuitry play in the generation and propagation of the locomotion pattern.
|
0.915 |