A Graph Inference Case Study: The C. Elegans Neural Network

• Activity Number: 569
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Li Chen*+ and Carey Priebe and Joshua Vogelstein
• Companies: Johns Hopkins University and Johns Hopkins University and Duke University
• Keywords: Graph inference ; The C. elegans neural network ; Random graph models

Abstract

We investigate graph inference for the pair of neural networks of the Caenorhabditis elegans. The C. elegans network consists of 302 neurons with known functional attributes. There are two networks, one for each of the two types of connections: synaptic and gap junctional. We formulate our inference in the setting of random graph models and study the structures of the worm neural networks, the correspondence between the two types of neuron connections, and the prediction of neuron types.