Strategies for Rapidly Dissecting the Neural and Molecular Correlates of Behavior

Abstract

The complexity of nervous systems can make finding neural and molecular drivers of behavior a difficult task. However, despite this apparent complexity, previous work has suggested that the important neurons that control a behavior might only be a sparse subset of the total number of neurons in the nervous system. This sparsity might offer an opportunity to find the key neurons controlling behavior more efficiently. In this thesis, I will explore how this sparsity in the nervous system might be used in order to greatly accelerate the process of finding neurons. By utilizing a compressed sensing based approach, we have demonstrated that it is possible to rapidly find neurons controlling two aspects of Caenorhabditis elegans behavior – the speed of locomotion and pathogen avoidance. Using a compressed sensing based approach, we identified three key neurons controlling the speed of locomotion of C.elegans, and showed that each of these neurons regulates a different aspect of C.elegans locomotion. In addition, we show that this approach can also be used to find the neurons that serve to encode experiences of pathogen exposure and thus regulate avoidance of pathogenic bacteria. The capacity to rapidly find key neurons controlling a behavior grants us the ability to find the molecular mechanisms underpinning behavior. We demonstrate how this might be approached by finding molecular candidates that are involved in tactile food sensing in mechanosensory dopaminergic neurons in C.elegans. Through this, we have discovered two new ion channels that regulate tactile sensation of food. By using a combination of neural perturbation and imaging, we were able to show how these channels work together to allow these neurons to encode tactile information with high fidelity.