neurobeam 0.3.0

Neuroscience Experiments Using Real-time Optogenetics, Behavior, and Microscopy

PROBE (Python Repo)

PROBE is an open-source software-hardware system for behavioral experiments in head-fixed rodents using a low-cost, modular experimental rig. The software is primarily written in python to allow easy distribution, use, and end-user modification with bits of C/C++ glued in for low-latency performance. An optional GUI is largely handled by your GPU to allow real-time streaming of ongoing behavior and data acquisition without disrupting your experiments. Currently PROBE is only supported on windows. Triggers dependent on neural activity require access to a raw-stream of imaging data. PROBE only offers native support for PrairieView at this time, but if you have a raw stream it should work. In constrast, software communication with spatial light modulators or hardware-triggered stimulation for optogenetic actions should be platform-independent.

More information on hardware implementation can be found here.

Integration Tests Make sure to launch prairie_mock_server.exe if conducting integration tests with PrairieView. This is automatically launched if calling prairie_probe_test.exe but NOT any other testing application/suite including google tests

Software Reminders These things aren't listed in TODOs Remember to test on Prairie-1 Workstation-3. It has an old Xeon processer so it will be a good test of performance of single-thread performance on a slow processor. Remember to double-check for potential race conditions. Remember to refactor GUI aesthetic and import logo before pushing here from WRD repo Remember to update badge links before ever making public--just placeholders from WRD repo. Figure out whether I need to do anything special for distribution of C++ lib, executables, & Node.js in a pypi package. Figure out whether anti-viruses will get grumpy from having an executable in a pypi package.

Hardware Reminders The sucrose preference robotic spout-swapper has not been tested. Try to test on a nicer NI-DAQ. Might be able to achieve higher best-case performance since they have hardware-timed digital I/O and I won't have to call those in-between analog-input callbacks. Take a second look at the PCB scheme and see if I can simplify the connection from / to DAQ. I probably won't re-fabricate since I did that out-of-pocket, but some optimization there will make building much simpler/more flexible.

Development Statistics (PyCharm Repo)

~75,000 Lines ~75% Coverage 0 Linting Errors

Developmentn Statistics (CLion Repo)

~25,000 Lines ~20% Coverage 0 Linting Errors

Supported Pre-Configured Behaviors

Module	Task	Description	Recipe	Pilot
Disk or Wheel	Locomotion	Hyperactivity & Motor Function	x	x
	Acoustic Startle	Hearing, Habituation, & Sensitization	x	x
	Pre-Pulse Inhibition	Sensorimotor Gating	x	x
	Go No-Go	Learning & Memory
Linear Track	Navigation	Learning & Memory
Burrow	Shelter Seeking	Anxiety
	Sucrose Preference	Anhedonia
	Attentional Set-Shifting	Behavioral Flexibility
	Go No-Go	Learning & Memory
	Trace Fear Conditioning	Learning & Memory
	Delay Fear Conditioning	Learning & Memory
	Instrumental Learning	Learning & Memory