Tools to enable brain wide chronic electrophysiology in freely moving rodents

The pending introduction of the 1536 channel probe Neuropixels 2.0 Quad Base in Q2 2024 followed by the much smaller 1536 channel Neuropixels NXT Probe (developed entirely with support from NIH U01 NS115587) will open a path to brain wide electrophysiology recording. However, the challenge of chronically implanting 4-8 such probes on a single animal and recording data at rates of 1.3 to 4.0 Terabytes/hour outside the ability of most groups. To maximize the scientific impact of this new technology, this application will simplify, speed up, and make more reliable the use of multiple 1536 channel probes for brain wide recording in freely moving animals. Three critical elements will be enhanced: multi-probe implantation, large-scale data acquisition, and large-scale data digestion. These will be benchmarked against a challenging scientific application, brain wide recording of spiking activity and local field potentials in a complex freely moving rodent navigation task. Implanting many probes in an animal significantly extends the time and risks of the implantation surgery. We will develop freely available tools for robot assisted surgery in mice and rats. Our goal is 8 probes implanted in less than 8 hours with no more than two attending scientists. In comparison, for those few labs that attempt so difficult a surgery, a 14-18 hour surgery is typical. Reliability of inserting each probe is also critical because failures are prohibitively costly, especially in heavily trained animals. Our surgeries will be reliably planned with a combination of Pinpoint anatomy-based targeting software and software we will develop to model the probe and probe fixtures outside the skull while projecting the paths of probe shanks inside the brain. The probe fixtures will be adapted allow for flexible positioning and angling of the shanks. SpikeGLX, developed in our lab, is currently the highest bandwidth GUI for stable data acquisition from Neuropixels probes. We have successfully recorded from > 12000 channels simultaneously in preliminary testing. We will confirm capacity to > 18,000 channels for the newest probe types, create a new graphical representation of ongoing signals across all probes for assessment and channel selection, and enable selective data saving based on behavioral epoch or external experimental factors (e.g. a trigger). Data analysis of recordings that may exceed 10 terabytes require fundamentally new approaches as manual curation is impractical. We will develop a pipeline for automated spike sorting of the raw data into spikes from individual neurons, including filtering for sorted neuron quality based on region specific properties. We will benchmark this new suite of capabilities against a complex navigation task in rodents. Navigational signals are known to be distributed across many brain regions. Therefore, rather than study one or a few regions at a time as is standard, an 8 probe, 32 shank, 12,000 channel brain wide recording will gather data from 25-40 brain regions simultaneously, allowing estimation of large-scale information flow. Project Number: 1R01NS142455-01A1 | Fiscal Year: 2026 | NIH Institute/Center: National Institute of Neurological Disorders and Stroke (NINDS) | Principal Investigator: TIMOTHY HARRIS (+1 co-PI) | Institution: JOHNS HOPKINS UNIVERSITY, BALTIMORE, MD | Award Amount: $616,585 | Activity Code: R01 | Study Section: Bioengineering and Tissue Engineering for Neuroscience Study Section[BTEN] View on NIH RePORTER: https://reporter.nih.gov/project-details/11298616