Afzal, M. F. | Search bioRxiv

bioRxiv2022

CRF Neurons Establish Resilience via Stress-History-Dependent BNST Modulation

IntroductionCumulative stress is a major risk factor for developing major depressive disorder (MDD), yet not everyone experiencing chronic stress develops MDD. In those who do not, it is unclear at what point, or by what mechanism, a trajectory of stable resiliency emerges. MethodsUtilizing a 10-day repeated social defeat stress model (RSDS) for MDD, we observed that a critical period between 7 and 10 daily defeats marks the phenotypical divergence of resilient from susceptible mice. Using cell-type selective electrophysiology, chemogenetics, optogenetics, fiber photometry and RNA quantification was employed to investigate the nature of stress effects on neuroadaptation in the oval nucleus of the bed nucleus of the stria terminalis (BNSTov) required to determine resilience. ResultsIn response to ongoing stress, corticotropin-releasing factor (CRF+, but not CRF-) neurons of the (BNSTov) displayed a sustained increased firing rate in resilient, but not susceptible mice. This neurophysiological adaptation was self-sustaining, but only after 7 critical stress exposures, indicating that the process of developing resilience is dependent on stress history. ConclusionOur study reveals a novel process by which individuals might persist in the face of adversity by way of stress-provoked activation, not inhibition of a key CRF limbic region that establishes a pathway to resilience.

neuroscience↗

bioRxiv2021

TRAKR - A reservoir-based tool for fast and accurate classification of neural time-series patterns

Distinguishing between complex nonlinear neural time-series patterns is a challenging problem in neuroscience. Accurately classifying different patterns could be useful for a wide variety of applications, e.g. detecting seizures in epilepsy and optimizing control spaces for brain-machine interfaces. It remains challenging to correctly distinguish nonlinear time-series patterns because of the high intrinsic dimensionality of such data, making accurate inference of state changes (for intervention or control) difficult. On the one hand, simple distance metrics, which can be computed quickly, often do not yield accurate classifications; on the other hand, ensembles or deep supervised approaches offer high accuracy but are training data intensive. We introduce a reservoir-based tool, state tracker (TRAKR), which provides the high accuracy of ensembles or deep supervised methods while preserving the benefits of simple distance metrics in being applicable to single examples of training data (one-shot classification). We show that TRAKR instantaneously detects deviations in dynamics as they occur through time, and can distinguish between up to 40 patterns from different chaotic data recurrent neural networks (RNNs) with above-chance accuracy. We apply TRAKR to a benchmark time-series dataset - permuted sequential MNIST - and show that it achieves high accuracy, performing on par with deep supervised networks and outperforming other distance-metric based approaches. We also apply TRAKR to electrocorticography (ECoG) data from the macaque orbitofrontal cortex (OFC) and, similarly, find that TRAKR performs on par with deep supervised networks, and more accurately than commonly used approaches such as Dynamic Time Warping (DTW). Altogether, TRAKR allows for high accuracy classification of time-series patterns from a range of different biological and non-biological datasets based on single training examples. These results demonstrate that TRAKR could be a viable alternative in the analysis of time-series data, offering the potential to generate new insights into the information encoded in neural circuits from single-trial data.

neuroscience↗