Systems Neuro Browser (SNUB) example datasets

SNUB is a visual interface for systems neuroscience. Using a set of linked data-views, users can explore relationships between raw video, 3D animal pose, behavior annotations, neural activity, or any other relevant time-series data. SNUB includes an I/O module that loads raw data and exports it to a standard format that can be loaded into the SBUB GUI. Here we provide example input data, example code for converting it into a SNUB-compatible format, and copies of SNUB datasets that this code generates.

Overview of files

miniscope_data.zip contains example one-photon calcium imaging data and a corresponding behavioral video, as well as a jupyter notebook for loading these data and exporting them to a SNUB dataset. These data are used in the Calcium Imaging tutorial in the SNUB docs.

miniscope_project.zip contains the SNUB dataset that is produced by the Calcium Imaging tutorial. This dataset can be loaded directly into the SNUB GUI, as described in the docs.

ephys_data.zip contains example spike-sorted electrophysiology data, behavioral video, and annotations of behavioral syllables, as well as a jupyter notebook for loading these data and exporting them to a SNUB dataset. These data are used in the Electrophysiology tutorial in the SNUB docs.

ephys_project.zip contains the SNUB dataset that is produced by the Electrophysiology tutorial. This dataset can be loaded directly into the SNUB GUI, as described in the docs.

Data provenance

This repository contains original data collected in the Datta lab between 2020 and 2022. All experimental procedures were approved by the Harvard Medical School Institutional Animal Care and Use Committee (Protocol Number 04930) and were performed in compliance with the ethical regulations of Harvard University as well as the Guide for Animal Care and Use of Laboratory Animals.

For one-photon calcium imaging, we used viral delivery to express a calcium-dependent fluorescent sensor (GCaMP6f) in prelimbic cortex which was imaged through a gradient index (GRIN) lens using the nVista 3.0 platform from Inscopix. Calcium videos were motion corrected and processed using constrained nonnegative matrix factorization (CNMF-E; implemented in CaImAn), yielding putative single-neuron activity traces for each recording. At the same time, animals were imaged from above using a Kinect Azure camera. This repository contains one recording from a single mouse.

For electrophysiology recordings, a 64-channel silicon probe (NeuroNexus) was implanted in substantia nigra pars reticulata. The probe interfaced with a RHD2164 headstage (Intan Technologies) that was attached to an ultra-thin Intan SPI cable which interfaced with an opto-electric commutator (Doric Lenses). OpenEphys software was used for signal acquisition and spike sorting was performed with the SpikeInterface package using Mountainsort4. This repository contains one recording from a single mouse.

Both experiments used 8-16-week-old C57BL/6J (The Jackson Laboratory stock no. 000664) housed on a 12-hour dark-light cycle (09:00-21:00). Mice were group housed prior to surgery and single-housed afterward. During surgery, mice were anesthetized using 1-2% isofluorane in oxygen, at a flow rate of 1L/min for the duration of the procedure.