Committee Members:

Garrett B. Stanley, PhD (Georgia Institute of Technology, Biomedical Engineering)

Todd Sulchek, PhD (Georgia Institute of Technology, Mechanical Engineering)

Hongkui Zeng, PhD (Allen Institute for Brain Science)

Edward S. Boyden, PhD  (Massachusetts Institute of Technology, Biological Engineering and Brain and Cognitive Sciences)

Suhasa B. Kodandaramaiah, PhD (University of Minnesota, Mechanical Engineering)

In vivo serial patch clamp robotics for cell type identification in the mouse visual cortex

Patch-clamping, the gold standard technique for measuring trans-membrane voltages and currents in neurons, involves delicately resting a 1 μm diameter pipette against a cell to create an intimate electrical and mechanical connection between the pipette tip and the cell membrane. From there, it is possible to record essentially interference-free single-neuron “spikes” in membrane voltage. These spikes are the primary method of inter-neuronal communication in the nervous system.

 The experimental procedure to obtain these high-fidelity recordings is considered an art form performed in vivo by a small number of highly trained individuals.  Previous work has introduced mechanical and electrical automation techniques, or "autopatching," to reduce the cognitive load and the required training to obtain these recordings, but the low yield is still a major limitation and requires many attempts to obtain a single recording.  

 This work introduces additional robotic tools to completely automate serial patch clamp recording attempts.  Electrical and mechanical hardware and software algorithms have been developed to automate pipette manipulation, specifically, pneumatic control, electrical control, precise positioning, replacement, filling, wire threading, and storage.  Taken together, these tools enable the first completely autonomous, serial patch clamp recording attempts in the living brain.

 These robotic tools also enable more difficult experiments that combine patch clamp recording with other techniques such as biocytin filling for morphological reconstruction.  Progress towards a survey of 50 cells in the visual cortex will be presented to establish cell type identification schemes based on coupled electrophysiology and morphology.