bioRxiv Subject Collection: Neuroscience's Journal

Grin1Y647S/+ Mice: A Preclinical Model of GRIN1-Related Neurodevelopmental Disorder
Objective: GRIN1-related neurodevelopmental disorder (GRIN1-NDD) is characterized by clinically significant variation in the GRIN1 gene, which encodes the obligatory GluN1 subunit of N-methyl-D-aspartate receptors (NMDARs). The identified p.Tyr647Ser (Y647S) variant carried by a 33-year-old female with seizures and intellectual disability is located in the M3 helix in the GluN1 transmembrane domain. This study builds upon initial in vitro investigations of the functional impacts of the GRIN1 Y647S variant and examines its in vivo consequences in a mouse model. Methods: To investigate in vitro functional impacts of NMDARs containing GluN1-Y647S variant subunits, GluN1-Y647S was co-expressed with wildtype GluN2A or GluN2B subunits in Xenopus laevis oocytes and HEK cells. Grin1Y647S/+ mice were created by CRISPR-Cas9 endonuclease-mediated transgenesis and the molecular, electrophysiological, and behavioural consequences of the variant were examined. Results: In vitro, NMDARs containing GluN1-Y647S show altered sensitivity to endogenous agonists and negative allosteric modulators, and reduced cell surface trafficking. Grin1Y647S/+ mice displayed a reduction in whole brain GluN1 levels and deficiency in NMDAR-mediated synaptic transmission in the hippocampus. Behaviourally, Grin1Y647S/+ mice exhibited spontaneous seizures, altered vocalizations, muscle strength, sociability, and problem-solving. Interpretation: The Y647S variant confers a complex in vivo phenotype, which reflects largely diminished properties of NMDAR function. As a result, Grin1Y647S/+ mice display atypical behaviour in domains relevant to the clinical characteristics of GRIN1-NDD and the individual carrying the variant. Ultimately, the characterization of Grin1Y647S/+ mice accomplished in the present work expands our understanding of the mechanisms underlying GRIN1-NDD and provides a foundation for the development of novel therapeutics.

Parallel mechanisms signal a hierarchy of sequence structure violations in the auditory cortex
The brain predicts regularities in sensory inputs at multiple complexity levels, with neuronal mechanisms that remain elusive. Here, we monitored auditory cortex activity during the local-global paradigm, a protocol nesting different regularity levels in sound sequences. We observed that mice encode local predictions based on stimulus occurrence and stimulus transition probabilities, because auditory responses are boosted upon prediction violation. This boosting was due to both short-term adaptation and an adaptation-independent surprise mechanism resisting anesthesia. In parallel, and only in wakefulness, VIP interneurons responded to the omission of the locally expected sound repeat at sequence ending, thus providing a chunking signal potentially useful for establishing global sequence structure. When this global structure was violated, by either shortening the sequence or ending it with a locally expected but globally unexpected sound transition, activity slightly increased in VIP and PV neurons respectively. Hence, distinct cellular mechanisms predict different regularity levels in sound sequences.

Anatomically resolved oscillatory bursts orchestrate visual thalamocortical activity during naturalistic stimulus viewing
Natural vision involves encoding of complex visual input, which engages a plethora of interacting circuit mechanisms. In the mammalian forebrain, one signature of such interacting circuit mechanisms is fast oscillatory dynamics, which can be reflected in the local field potential (LFP). We here used data from the Allen Neuropixels Visual Coding project to show that local visual features in naturalistic stimuli induce retinotopically specific V1 oscillations in various frequency bands. These LFP oscillations occurred in bursts, were localized to specific V1 layers, and were associated with phase coupling of V1 translaminar spiking, pointing to feature-specific circuit motifs. Finally, we discovered that these visually-induced circuit motifs occurred across a range of stimuli, suggesting that they might constitute general routes for feature-specific information flow. Together, our analyses demonstrate visually-induced, fast oscillations, which likely reflect the operation of distinct mesoscale circuits for the differential and multiplexed coding of complex visual input and feature-specific thalamo-cortical information propagation.

A Hippocampal-parietal Network for Reference Frame Coordination
Navigating space and forming memories based on spatial experience are crucial for survival, including storing memories in an allocentric (map-like) framework and conversion into body-centered action. The hippocampus and parietal cortex (PC) comprise a network for coordinating these reference frames, though the mechanism remains unclear. We used a task requiring remembering previous spatial locations to make correct future action and observed that hippocampus can encode the allocentric place, while PC encodes upcoming actions and relays this to hippocampus. Transformation from location to action unfolds gradually, with 'Came From' signals diminishing and future action representations strengthening. PC sometimes encodes previous spatial locations in a route-based reference frame and conveys this to hippocampus. The signal for the future location appears first in PC, and then in hippocampus, in the form of an egocentric direction of future goal locations, suggesting egocentric encoding recently observed in hippocampus may originate in PC (or another "upstream" structure). Bidirectional signaling suggests a coordinated mechanism for integrating map-like, route-centered, and person-centered spatial reference frames at the network level during navigation.

Cellular and circuit features distinguish dentate gyrus semilunar granule cells and granule cells activated during contextual memory formation
The dentate gyrus is critical for spatial memory formation and shows task related activation of cellular ensembles considered as memory engrams. Semilunar granule cells (SGCs), a sparse dentate projection neuron subtype distinct from granule cells (GCs), were recently reported to be enriched among behaviorally activated neurons. However, the mechanisms governing SGC recruitment during memory formation and their role in engram refinement remains unresolved. By examining neurons labeled during contextual memory formation in TRAP2 mice, we empirically tested competing hypotheses for GC and SGC recruitment into memory ensembles. In support of the proposal that more excitable neurons are preferentially recruited into memory ensembles, SGCs showed greater sustained firing than GCs. Additionally, SGCs labeled during memory formation showed less adapting firing than unlabeled SGCs. Our recordings did not reveal glutamatergic connections between behaviorally labeled SGCs and GCs, providing evidence against SGCs driving local circuit feedforward excitation in ensemble recruitment. Contrary to a leading hypothesis, there was little evidence for individual SGCs or labeled neuronal ensembles supporting lateral inhibition of unlabeled neurons. Instead, pairs of GCs and SGCs within labeled neuronal cohorts received more temporally correlated spontaneous excitatory synaptic inputs than labeled-unlabeled neuronal pairs, validating a role for correlated afferent inputs in neuronal ensemble selection. These findings challenge the proposal that SGCs drive dentate GC ensemble refinement, while supporting a role for intrinsic active properties and correlated inputs in preferential SGC recruitment to contextual memory engrams.

Learning reshapes the hippocampal representation hierarchy
A key feature of biological and artificial neural networks is the progressive refinement of their neural representations with experience. In neuroscience, this fact has inspired several recent studies in sensory and motor systems. However, less is known about how higher associational cortical areas, such as the hippocampus, modify representations throughout the learning of complex tasks. Here we focus on associative learning, a process that requires forming a connection between the representations of different variables for appropriate behavioral response. We trained rats in a spatial-context associative task and monitored hippocampal neural activity throughout the entire learning period, over several days. This allowed us to assess changes in the representations of context, movement direction and position, as well as their relationship to behavior. We identified a hierarchical representational structure in the encoding of these three task variables that was preserved throughout learning. Nevertheless, we also observed changes at the lower levels of the hierarchy where context was encoded. These changes were local in neural activity space and restricted to physical positions where context identification was necessary for correct decision making, supporting better context decoding and contextual code compression. Our results demonstrate that the hippocampal code not only accommodates hierarchical relationships between different variables but also enables efficient learning through minimal changes in neural activity space. Beyond the hippocampus, our work reveals a representation learning mechanism that might be implemented in other biological and artificial networks performing similar tasks.

Effective excitability captures network dynamics across development and phenotypes
Neuronal cultures in vitro are a versatile system for studying the fundamental properties of individual neurons and neuronal networks. Recently, this approach has gained attention as a precision medicine tool. Mature neuronal cultures in vitro exhibit synchronized collective dynamics called network bursting. If analyzed appropriately, this activity could offer insights into the network's properties, such as its composition, topology, and developmental and pathological processes. A promising method for investigating the collective dynamics of neuronal networks is to map them onto simplified dynamical systems. This approach allows the study of dynamical regimes and the characteristics of the parameters that lead to data-consistent activity. We designed a simple biophysically inspired dynamical system and used Bayesian inference to fit it to a large number of recordings of in vitro population activity. Even with a small number of parameters, the model showed strong inter-parameter dependencies leading to invariant bursting dynamics for many parameter combinations. We further validated this observation in our analytical solution. We found that in vitro bursting can be well characterized by each of three dynamical regimes: oscillatory, bistable, and excitable. The probability of finding a data-consistent match in a particular regime changes with network composition and development. The more informative way to describe the in vitro network bursting is the effective excitability, which we analytically show to be related to the parameter-invariance of the model's dynamics. We establish that the effective excitability can be estimated directly from the experimentally recorded data. Finally, we demonstrate that effective excitability reliably detects the differences between cultures of cortical, hippocampal, and human pluripotent stem cell-derived neurons, allowing us to map their developmental trajectories. Our results open a new avenue for the model-based description of in vitro network phenotypes emerging across different experimental conditions.

An estimate of the longitudinal pace of aging from a single brain scan predicts dementia conversion, morbidity, and mortality
To understand how aging affects functional decline and increases disease risk, it is necessary to develop accurate and reliable measures of how fast a person is aging. Epigenetic clocks measure aging but require DNA methylation data, which many studies lack. Using data from the Dunedin Study, we introduce an accurate and reliable measure for the rate of longitudinal aging derived from cross-sectional brain MRI: the Dunedin Pace of Aging Calculated from NeuroImaging or DunedinPACNI. Exporting this measure to the Alzheimer's Disease Neuroimaging Initiative and UK Biobank neuroimaging datasets revealed that faster DunedinPACNI predicted participants' cognitive impairment, accelerated brain atrophy, and conversion to diagnosed dementia. Underscoring close links between longitudinal aging of the body and brain, faster DunedinPACNI also predicted physical frailty, poor health, future chronic diseases, and mortality in older adults. Furthermore, DunedinPACNI followed the expected socioeconomic health gradient. When compared to brain age gap, an existing MRI aging biomarker, DunedinPACNI was similarly or more strongly related to clinical outcomes. DunedinPACNI is a 'next generation' MRI measure that will be made publicly available to the research community to help accelerate aging research and evaluate the effectiveness of dementia prevention and anti-aging strategies.

The long-range gene regulatory landscape of cerebellar granule neuron progenitors
Neuronal specification, expansion and differentiation are tightly regulated by the concerted actions of transcription and chromatin modifying factors that are recruited to regulatory elements in the genome. Tissue-specific distal regulatory elements are typically located tens to hundreds of kilobases from the gene they regulate. To identify the distal enhancers that directly regulate a gene, information on the localisation of enhancers relative to the gene promoter in the nucleus is crucial. Cerebellar granule cell progenitors (GCps) are important transit amplifying neuronal progenitors, giving rise to the most abundant neuronal cell type in the brain. Many of the key factors that regulate fundamental developmental processes in GCps have been identified. For instance, the proneural transcription factor Atoh1 is essential for GCp specification, proliferation and differentiation and the ATP-dependent chromatin remodeller CHD7 is necessary for normal GCp proliferation and differentiation. However, both these factors are recruited to distal regulatory elements and the direct regulatory relationships between these factors, the enhancers they are recruited to, and the genes they regulate in GCps remain uncharacterised. To identify active, long-range gene regulatory interactions in GCps, we used promoter capture Hi-C (pcHi-C), together with ATAC-seq and ChIP-seq data. We present a rich dataset consisting of 46,428 interactions between 22,797 putative distal regulatory regions and 12,905 protein coding gene promoters in primary mouse GCps. Using VISTA-designated hindbrain enhancers as an example, we show that 80% of these enhancers are incorrectly annotated at present and identify the genes most likely regulated directly by these enhancers. Motif enrichment analyses showed a significant enrichment of proneural transcription factor motifs in CHD7-regulated enhancers. Further analyses revealed co-localisation of Atoh1 and CHD7 at gene enhancers, suggesting a novel regulatory relationship between Atoh1 and CHD7 in controlling the expression of key genes in the GCp lineage. We used our data to identify >1,500 Atoh-regulated enhancers, controlling the expression of 577 genes in GCps, and 197 enhancers of 22 genes that appear to be co-regulated by Atoh1 and CHD7. Co-immunoprecipitation experiments showed that Atoh1 and CHD7 interacted with each other. These findings support the emerging picture of CHD7 as an important gene regulatory co-factor for lineage-specific transcription factors. The pcHi-C data is presented as a useful resource to the community for investigating the function of long-range enhancers in the cerebellar GCp lineage.

Evoked resonant neural activity outperforms spectral markers in decoding sleep from the subthalamic nucleus
Background: Deep brain stimulation is a treatment for advanced Parkinson's disease and currently tuned to target motor symptoms during daytime. Parkinson's disease is associated with multiple nocturnal symptoms such as akinesia, insomnia and sleep fragmentation which may require adjustments of stimulation during sleep for best treatment outcome. Objectives: There is a need for a robust biomarker to guide stimulation titration across sleep stages. This study aimed to investigate whether evoked resonant neural activity (ERNA) is modulated during sleep. Methods: We recorded local field potentials from the subthalamic nucleus of four Parkinson's patients with externalised electrodes while applying single stimulation pulses to investigate the effect of sleep on ERNA. Results: We found that ERNA features change with wakefulness and sleep stages, and are correlated with canonical frequency bands and heart rate. We further evaluated the performance of machine learning models in classifying non-REM sleep versus wakefulness and found that ERNA amplitude outperforms all spectral markers. Conclusions: Given the heterogeneity of spectral features during sleep, their susceptibility to movement artefacts and superior classification accuracy of models using ERNA features, this study paves the way for ERNA as a marker for automatic stimulation titration during sleep and improved patient care.

Inflammatory Signatures of Microglia in the Mouse Model of Corticosterone-Induced Depression
Microglia-mediated inflammation has been recognized as a key feature of major depressive disorder. Although hypercortisolemia commonly occurs in depressed patients and can be predictive of treatment response, how chronic exposure to this stress hormone influences microglia is incompletely characterized. Here, we exploited a standard mouse model of depressive-like behaviors induced by peripheral administration of corticosterone. Microglia in the prefrontal cortex of mice were profiled by bulk RNA sequencing, which exhibited the up-regulation of inflammatory markers. In addition, single-cell RNA sequencing identified distinct molecular patterns of microglial responses. Moreover, we revealed the elevation of Pu.1 and Irf8, the two central transcription factors governing microglia-mediated inflammation, in the prefrontal cortex and hippocampus of corticosterone-treated mice, which was similarly observed in the single-nucleus RNA sequencing dataset of depressed patients' microglia. These results have established inflammatory signatures of microglia in the mouse model recapitulating hypercortisolemia-related depression, providing new insights into diagnostic and therapeutic strategies.

A fluorophore-ligand conjugate for the rapid and reversible staining of native AMPA receptors in living neurons
The subcellular localizations of neurotransmitter receptors are strictly regulated in neurons. Changes in the trafficking of alpha-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA)-type glutamate receptors (AMPARs) play an essential role in synaptic plasticity, which is the cellular basis of learning and memory. To explore receptor trafficking, genetically encoded approaches (e.g., the fusion of fluorescent proteins to receptors) are often used. However, concerns remain that genetic approaches cannot fully reproduce the receptor function that is inherent to neurons. Herein, we report a chemical probe, PFQX1(AF488), for the visualization of cell-surface AMPARs without any genetic manipulation to neurons. The rapid and reversible staining features of this probe allowed the visualization of AMPAR accumulation in dendritic spines during synaptic plasticity in living hippocampal neurons. Moreover, the mechanisms of this synaptic accumulation, for which genetically encoded approaches have given controversial results, were revealed by integrating two chemical methods: PFQX1(AF488) and covalent chemical labeling.

Cortical Excitability is Affected by Light Exposure- Distinct Effects in Adolescents and Young Adults
Light, particularly blue-wavelength light, exerts a broad range of non-image forming (NIF) effects including the stimulation of cognition and alertness and the regulation of mood, sleep and circadian rhythms. However, its underlying brain mechanisms are not fully elucidated. Likewise, whether adolescents show a different NIF sensitivity to light compared to adults is not established. Here, we investigated whether cortical excitability, a basic aspect of brain function that depends on sleep wake regulation, is affected by blue light and whether the effect is similar in young adults and adolescents. To do so, we used transcranial magnetic stimulation coupled to high-density electroencephalography (TMS-EEG) in healthy young adults (N=13, 24.2y+/-3.4) and in adolescents (N=15, 16.9y+/-1.1). Our results showed that, in young adults, blue light affected cortical excitability following an apparent inverted-U relationship, while adolescents' cortical excitability was not significantly impacted by blue light. In addition, although light did not affect performance on a visuomotor vigilance task completed during the TMS-EEG recordings, cortical excitability was positively correlated to task performance in both age groups. This study provides valuable insights into the complex interplay between light, cortical excitability, and behavior. Our findings highlight the role of age in NIF effects of light, suggesting that brain responses to light differ during developmental periods.

Neural Adaptation to the Eye's Optics Through Phase Compensation
How does the brain achieve a seemingly veridical and in-focus perception of the world, knowing how severely corrupted visual information is by the eyes optics? Optical blur degrades retinal image quality by reducing the contrast and disrupting the phase of transmitted signals. Neural adaptation can attenuate the impact of blur on image contrast, yet vision rather relies on perceptually-relevant information contained within the phase structure of natural images. Here we show that neural adaptation can compensate for the impact of optical aberrations on phase congruency. We used adaptive optics to fully control optical factors and test the impact of specific optical aberrations on the perceived phase of compound gratings. We assessed blur-induced changes in perceived phase over three distinct exposure spans. Under brief blur exposure, perceived phase shifts matched optical theory predictions. During short-term (~1h) exposure, we found a reduction in blur-induced phase shifts over time, followed by after-effects in the opposite direction-a hallmark of adaptation. Finally, patients with chronic exposure to poor optical quality showed altered phase perception when tested under fully-corrected optical quality, suggesting long-term neural compensatory adjustments to phase spectra. These findings reveal that neural adaptation to optical aberrations compensates for alterations in phase congruency, helping restore perceptual quality over time.

Social 'envirotyping' the ABCD study contextualizes dissociable brain organization and diverging outcomes
The environment, especially social features, plays a key role in shaping the development of the brain, notably during adolescence. To better understand variation in brain-environment coupling and its associated outcomes, we identified ''envirotypes,'' or different patterns of social environment experience, in the Adolescent Brain Cognitive Development Study by hierarchically clustering subjects. Two focal clusters, which accounted for 89.3% of all participants, differed significantly on eight out of nine youth-report social environment quality measures, representing almost perfect complements. We then applied tools from network neuroscience to show different envirotypes are associated with different patterns of whole brain functional connectivity. Differences were distributed across the brain but were especially prominent in Default and Somatomotor Hand systems for these focal clusters. Finally, we examined how social envirotypes change over development and how these patterns of change are associated with a suite of outcomes. The resulting dynamic envirotypes differed along dimensions of stability and quality, but outcomes diverged based on stability. Specifically, the stable, high quality envirotype was most easily distinguished from the improving envirotype, while the unstable envirotype was associated with the worst outcomes. Altogether, our findings represent significant contributions to both social developmental neuroscience and network neuroscience, emphasizing the variability and dynamicity of brain-environment coupling and its consequences.

Reverse-breaking CFS (rev-bCFS): Disentangling conscious and unconscious effects by measuring suppression and dominance times during continuous flash suppression
Breaking continuous flash suppression (bCFS) is a widely used experimental paradigm that exploits detection tasks to measure the time an initially invisible stimulus requires to escape interocular suppression and access awareness. One pretty contentious and unresolved issue is whether differences in detection times reflect unconscious or conscious processing. To answer this question, here we introduce a novel approach (i.e., reverse-bCFS [rev-bCFS]) that measures the time an initially visible stimulus requires to be suppressed from awareness. Results from two experiments using face stimuli indicate that rev-bCFS can capture conscious effects, which indicates that contrasting standard bCFS with rev-bCFS can isolate unconscious processing occurring specifically during bCFS. For example, while face inversion impacted both bCFS and rev-bCFS, effects were larger in bCFS, suggesting a distinct contribution of unconscious processing to the advantage of upright over inverted faces in accessing awareness. Combining standard bCFS and rev-bCFS may offer a fruitful approach able to disentangle conscious and unconscious effects occurring during interocular suppression.

A molecular brain atlas reveals cellular shifts during the repair phase of stroke
Ischemic stroke triggers a cascade of pathological events that affect multiple cell types and often lead to incomplete functional recovery. Despite advances in single-cell technologies, the molecular and cellular responses that contribute to long-term post-stroke impairment remain poorly understood. To gain better insight into the underlying mechanisms, we generated a single-cell transcriptomic atlas from distinct brain regions using a mouse model of permanent focal ischemia at one month post-injury. Our findings reveal cell- and region-specific changes within the stroke-injured and peri-infarct brain tissue. For instance, GABAergic and glutamatergic neurons exhibited upregulated genes in signaling pathways involved in axon guidance and synaptic plasticity, and downregulated pathways associated with aerobic metabolism. Using cell-cell communication analysis, we identified increased strength in predicted interactions within stroke tissue among both neural and non-neural cells via signaling pathways such as those involving collagen, protein tyrosine phosphatase receptor, neuronal growth regulator, laminin, and several cell adhesion molecules. Furthermore, we found a strong correlation between mouse transcriptome responses after stroke and those observed in human nonfatal brain stroke lesions. Common molecular features were linked to inflammatory responses, extracellular matrix organization, and angiogenesis. Our findings provide a detailed resource for advancing our molecular understanding of stroke pathology and for discovering therapeutic targets in the repair phase of stroke recovery.

Olfactory receptor coexpression and co-option in the dengue mosquito
The olfactory sensory neurons of vinegar flies and mice tend to express a single ligand-specific receptor. While this 'one neuron-one receptor' motif has long been expected to apply broadly across insects, recent evidence suggests it may not extend to mosquitoes. We sequenced and analyzed the transcriptomes of 46,000 neurons from antennae of the dengue mosquito Aedes aegypti to resolve all olfactory, thermosensory, and hygrosensory neuron subtypes and identify the receptors expressed therein. We find that half of all olfactory subtypes coexpress multiple receptors. However, coexpression occurs almost exclusively among genes from the same family--among odorant receptors (ORs) or among ionotropic receptors (IRs). Coexpression of ORs with IRs is exceedingly rare. Many coexpressed receptors are recent duplicates. In other cases, the recruitment or co-option of single receptors by multiple neuron subtypes has placed these genes together in the same cells with distant paralogs. Close examination of data from Drosophila reveal rare cases of both phenomena, indicating that the olfactory systems of these two species are not fundamentally different, but instead fall at different locations along a continuum likely to encompass diverse insects.

Deep inverse modeling reveals dynamic-dependent invariances in neural circuit mechanisms
Neural population dynamics are shaped by many cellular, synaptic, and network properties. Not only is it important to understand how coordinated changes in circuit parameters alter neural activity, but also when dynamics are unaffected by--or invariant to--such changes. Computational modeling has revealed invariances in single neurons and small circuits that are thought to reflect their robustness against variability and perturbations. However, generalizing these insights to larger circuits in cortex and other brain areas remains challenging. A key bottleneck lies in inverse modeling of neural circuits with spiking network models, i.e., identifying parameter configurations that quantitatively match dynamics observed in neural recordings. Here, we present Automated Model Inference from Neural Dynamics (AutoMIND) for efficient discovery of invariant circuit model configurations. AutoMIND leverages a mechanistic model with adaptive spiking neurons and clustered connectivity, which displays a rich variety of spatiotemporal dynamics. Probabilistic deep generative models--trained on network simulations only--then returns many parameter configurations consistent with a given target observation of neural activity. Applied to several datasets, AutoMIND discovers circuit models of synchronous network bursting in human brain organoids across early development, as well as models capturing complex frequency profiles of Neuropixels recordings in mouse hippocampus and cortex. In each case, we obtain hundreds of configurations that compose a (nonlinear) parameter subspace in which population dynamics remain unchanged. Surprisingly, global and local geometries of the invariant subspace are not fixed, but differ for different dynamics. Together, our results shed light on dynamic-dependent invariances of circuit parameters underlying diverse population dynamics, while demonstrating the flexibility of AutoMIND for inverse modeling of neural circuits.

ActSort: An active-learning accelerated cell sorting algorithm for large-scale calcium imaging datasets
Recent advances in calcium imaging enable simultaneous recordings of up to a million neurons in behaving animals, producing datasets of unprecedented scales. Although individual neurons and their activity traces can be extracted from these videos with automated algorithms, the results often require human curation to remove false positives, a laborious process called 'cell sorting'. To address this challenge, we introduce ActSort, an active-learning algorithm for sorting large-scale datasets that integrates features engineered by domain experts together with data formats with minimal memory requirements. By strategically bringing outlier cell candidates near the decision boundary up for annotation, ActSort reduces human labor to about 1-3% of cell candidates and improves curation accuracy by mitigating annotator bias. To facilitate the algorithm's widespread adoption among experimental neuroscientists, we created a user-friendly software and conducted a first-of-its-kind benchmarking study involving about 160,000 annotations. Our tests validated ActSort's performance across different experimental conditions and datasets from multiple animals. Overall, ActSort addresses a crucial bottleneck in processing large-scale calcium videos of neural activity and thereby facilitates systems neuroscience experiments at previously inaccessible scale

An autonomous robotic system for high-throughput phenotyping and behavioral studies of individual fruit flies
The fruit fly, Drosophila melanogaster, is a widely used model species in biomedical research. Despite its importance, conducting manual experiments with individual fruit flies can be challenging and time-consuming, especially for studies of individual fly behaviors. Such studies often involve cumbersome preparatory steps, such as manually tethering a fly and then positioning it within an experimental setup1,2. These procedures commonly require the fly to be anesthetized, and, before behavioral assessments begin, the fly must recover from anesthesia. Hence, the introduction of automated phenotyping and behavioral assays would expedite important aspects of fly research, by minimizing manual handling of flies and decreasing the net time needed for experiments. Here, we introduce FlyMAX (Fly Manipulation and Autonomous eXperimentation), an autonomous robotic system for manipulating adult flies without use of anesthesia. FlyMAX collects individual flies from a standard vial, analyzes them with computer vision, and achieves a throughput of >1,000 flies per day for high-throughput inspection and characterization assays. Robotic handling had no detectable adverse effects on fly longevity or our assessments of fly health. Moreover, the behavioral performance of flies, especially of males, was better and less variable than of flies handled manually. Our system employs deep learning-based machine vision for real-time assessments of picking quality and fly phenotypes. This enables fully pipelined, autonomous experimentation for behavioral assays with individual flies in controlled environments, which was previously infeasible. Overall, FlyMAX constitutes a promising technology to enhance the efficiency and reproducibility of research with flies and other insects in fields such as genetics, neuroscience, and drug screening.

Histone deacetylase inhibition expands cellular proteostasis repertoires to enhance neuronal stress resilience
Neurons are long-lived, terminally differentiated cells with limited regenerative capacity. Cellular stressors such as endoplasmic reticulum (ER) protein folding stress and membrane trafficking stress accumulate as neurons age and accompany age-dependent neurodegeneration. Current strategies to improve neuronal resilience are focused on using factors to reprogram neurons or targeting specific proteostasis pathways. We discovered a different approach. In an unbiased screen for modifiers of neuronal membrane trafficking defects, we unexpectedly identified a role for histone deacetylases (HDACs) in limiting cellular flexibility in choosing cellular pathways to respond to diverse types of stress. We show that genetic or pharmacological inactivation of HDACs results in improved neuronal health in response to ER protein folding stress and endosomal membrane trafficking stress in C. elegans and mammalian neurons. Surprisingly, HDAC inhibition enabled neurons to activate latent proteostasis pathways tailored to the nature of the individual stress, instead of generalized transcriptional upregulation. These findings shape our understanding of neuronal stress responses and suggest new therapeutic strategies to enhance neuronal resilience.

Coordinated cross-brain activity during accumulation of sensory evidence and decision commitment
Cognition is produced by the continuous interactions between many regions across the brain, but has typically been studied one brain region at a time. How signals in different regions coordinate to achieve a single coherent action remains unclear. Here, we address this question by characterizing the simultaneous interactions between up to 20 brain regions across the brain (10 targeted regions per hemisphere), of rats performing the "Poisson Clicks" task, a decision-making task that demands the gradual accumulation of momentary evidence. Using 8 Neuropixels probes in each animal, we recorded simultaneously in prefrontal cortex, striatum, motor cortex, hippocampus, amygdala, and thalamus. To assess decision-related interactions between regions, we quantified correlations of each region's "decision variable": moment-to-moment co-fluctuations along the axis in neural state space that best predicts the upcoming choice. This revealed a network of strongly correlated brain regions that include the dorsomedial frontal cortex (dmFC), anterior dorsal striatum (ADS), and primary motor cortex (M1), whose decision variables also led the rest of the brain. If coordinated activity within this subnetwork reflects an ongoing evidence accumulation process, these correlations should cease at the time of decision commitment. We therefore compared correlations before versus after "nTc", a recently reported estimator of the time of internal decision commitment. We found that correlations in the decision variables between different brain regions decayed to near-zero after nTc. Additionally, we found that choice-predictive activity grew over time before nTc, but abruptly plateaued at nTc, consistent with an evidence accumulation process that has stopped evolving at that time. Assessing nTc from the activity of individual regions revealed that nTc could be reliably detected earlier in M1 than other regions. These results show that evidence accumulation involves coordination within a network of frontal cortical and striatal regions, and suggests that termination of this process may initiate in M1.

Graph models of brain state in deep anaesthesia reveal sink state dynamics of reduced spatiotemporal complexity and integration
Deep anaesthetisation is an important surgical and explorative tool in the study of consciousness. Much work has been done to connect the deeply anaesthetised condition with decreased functional complexity. However, anaesthesia-induced unconsciousness is also a dynamic condition in which functional activity and complexity may fluctuate, being perturbed by internal or external (e.g. noxious) stimuli. We use fMRI data from a cohort undergoing ultra-slow propofol induction and a dynamic graph modelling framework which characterises changes in functional activity and connectivity as changes in brain state. We examine the group-level dynamics of brain state activity, complexity, integration and modularization in deep anaesthesia and wakefulness and propose novel measures for temporal complexity of brain state dynamics. We find that deep anaesthesia states are less temporally complex and less functionally-related to one another than wakeful states. Anaesthesia dynamics are dominated by a handful of sink states that act as low-complexity attractors to which subjects repeatedly return. Our analysis suggests that dynamic functional organisation in anaesthesia can be characterised by temporally stable, modular communities that change little in time, stratifying regional activity and functionally isolating cortical from both subcortical, and in particular, thalamic brain areas. Using data from a large human structural tractography atlas, we show that functional modularisation in deep anaesthesia more closely resembles normative structural organisation than does functional organisation in wakefulness. We find evidence suggesting that brain state trajectories and in particular, the dominance of low complexity attractor-like sink states in anaesthesia, appear to depend on subject-specific age and anaesthesia susceptibility factors.

Keratinocyte-derived exosomes in painful diabetic neuropathy
Painful diabetic neuropathy (PDN) is a challenging complication of diabetes with patients experiencing a painful and burning sensation in their extremities. Existing treatments provide limited relief without addressing the underlying mechanisms of the disease. PDN involves the gradual degeneration of nerve fibers in the skin. Keratinocytes, the most abundant epidermal cell type, are closely positioned to cutaneous nerve terminals, suggesting the possibility of bi-directional communication. Exosomes are small extracellular vesicles released from many cell types that mediate cell to cell communication. The role of keratinocyte-derived exosomes (KDEs) in influencing signaling between the skin and cutaneous nerve terminals and their contribution to the genesis of PDN has not been explored. In this study, we characterized KDEs in a well-established high-fat diet (HFD) mouse model of PDN using primary adult mouse keratinocyte cultures. We obtained highly enriched KDEs through size exclusion chromatography and then analyzed their molecular cargo using proteomic analysis and small RNA sequencing. We found significant differences in the protein and microRNA content of HFD KDEs compared to KDEs obtained from control mice on a regular diet (RD), including pathways involved in axon guidance and synaptic transmission. Additionally, using an in vivo conditional extracellular vesicle (EV) reporter mouse model, we demonstrated that epidermal-originating GFP-tagged KDEs are retrogradely trafficked into the DRG neuron cell body. Overall, our study presents a potential novel mode of communication between keratinocytes and DRG neurons in the skin, revealing a possible role for KDEs in contributing to the axonal degeneration that underlies neuropathic pain in PDN. Moreover, this study presents potential therapeutic targets in the skin for developing more effective, disease-modifying, and better-tolerated topical interventions for patients suffering from PDN, one of the most common and untreatable peripheral neuropathies.