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.

Zebrafish larvae are used to model the pathogenesis of multiple bacteria. This transparent model offers the unique advantage of allowing quantification of fluorescent bacterial burdens (fluorescent pixel counts: FPC) in vivo by facile microscopical methods, replacing enumeration of bacteria using time-intensive plating of lysates on bacteriological media. Accurate FPC measurements require laborious manual image processing to mark the outside borders of the animals so as to delineate the bacteria inside the animals from those in the culture medium that they are in. Here, we have developed an automated ImageJ/Fiji-based macro that accurately detect the outside borders of Mycobacterium marinum-infected larvae.

Assays that measure morphology, proliferation, motility, deformability, and migration are used to study the invasiveness of cancer cells. However, native invasive potential of cells may be hidden from these contextual metrics because they depend on culture conditions. We created a micropatterned chip that mimics the native environmental conditions, quantifies the invasive potential of tumor cells, and improves our understanding of the malignancy signatures. Unlike conventional assays, which rely on indirect measurements of metastatic potential, our method uses three-dimensional microchannels to measure the basal native invasiveness without chemoattractants or microfluidics. No change in cell death or proliferation is observed on our chips. Using six cancer cell lines, we show that our system is more sensitive than other motility-based assays, measures of nuclear deformability, or cell morphometrics. In addition to quantifying metastatic potential, our platform can distinguish between motility and invasiveness, help study molecular mechanisms of invasion, and screen for targeted therapeutics.

Dendritic cells (DCs) are specialized sentinel and APCs coordinating innate and adaptive immunity. Through proteins on their cell surface, DCs sense changes in the environment, internalize pathogens, present processed Ags, and communicate with other immune cells. By combining chemical labeling and quantitative mass spectrometry, we systematically profiled and compared the cell-surface proteomes of human primary conventional DCs (cDCs) in their resting and activated states. TLR activation by a lipopeptide globally reshaped the cell-surface proteome of cDCs, with >100 proteins upregulated or downregulated. By simultaneously elevating positive regulators and reducing inhibitory signals across multiple protein families, the remodeling creates a cell-surface milieu promoting immune responses. Still, cDCs maintain the stimulatory-to-inhibitory balance by leveraging a distinct set of inhibitory molecules. This analysis thus uncovers the molecular complexity and plasticity of the cDC cell surface and provides a roadmap for understanding cDC activation and signaling.

DaCapo is a specialized deep learning library tailored to expedite the training and application of existing machine learning approaches on large, near-isotropic image data. In this correspondence, we introduce DaCapo's unique features optimized for this specific domain, highlighting its modular structure, efficient experiment management tools, and scalable deployment capabilities. We discuss its potential to improve access to large-scale, isotropic image segmentation and invite the community to explore and contribute to this open-source initiative.

Natural physical, chemical, and biological dynamical systems are often complex, with heterogeneous components interacting in diverse ways. We show that graph neural networks can be designed to jointly learn the interaction rules and the structure of the heterogeneity from data alone. The learned latent structure and dynamics can be used to virtually decompose the complex system which is necessary to parameterize and infer the underlying governing equations. We tested the approach with simulation experiments of moving particles and vector fields that interact with each other. While our current aim is to better understand and validate the approach with simulated data, we anticipate it to become a generally applicable tool to uncover the governing rules underlying complex dynamics observed in nature.

The evolutionary expansion of sensory neuron populations detecting important environmental cues is widespread, but functionally enigmatic. We investigated this phenomenon through comparison of homologous olfactory pathways of Drosophila melanogaster and its close relative Drosophila sechellia, an extreme specialist for Morinda citrifolia noni fruit. D. sechellia has evolved species-specific expansions in select, noni-detecting olfactory sensory neuron (OSN) populations, through multigenic changes. Activation and inhibition of defined proportions of neurons demonstrate that OSN number increases contribute to stronger, more persistent, noni-odour tracking behaviour. These expansions result in increased synaptic connections of sensory neurons with their projection neuron (PN) partners, which are conserved in number between species. Surprisingly, having more OSNs does not lead to greater odour-evoked PN sensitivity or reliability. Rather, pathways with increased sensory pooling exhibit reduced PN adaptation, likely through weakened lateral inhibition. Our work reveals an unexpected functional impact of sensory neuron population expansions to explain ecologically-relevant, species-specific behaviour.

AMPA-type receptors (AMPARs) are rapidly inserted into synapses undergoing plasticity to increase synaptic transmission, but it is not fully understood if and how AMPAR-containing vesicles are selectively trafficked to these synapses. Here, we developed a strategy to label AMPAR GluA1 subunits expressed from their endogenous loci in cultured rat hippocampal neurons and characterized the motion of GluA1-containing vesicles using single-particle tracking and mathematical modeling. We find that GluA1-containing vesicles are confined and concentrated near sites of stimulation-induced structural plasticity. We show that confinement is mediated by actin polymerization, which hinders the active transport of GluA1-containing vesicles along the length of the dendritic shaft by modulating the rheological properties of the cytoplasm. Actin polymerization also facilitates myosin-mediated transport of GluA1-containing vesicles to exocytic sites. We conclude that neurons utilize F-actin to increase vesicular GluA1 reservoirs and promote exocytosis proximal to the sites of synaptic activity.

Primary cilia on granule cell neuron progenitors in the developing cerebellum detect sonic hedgehog to facilitate proliferation. Following differentiation, cerebellar granule cells become the most abundant neuronal cell type in the brain. While granule cell cilia are essential during early developmental stages, they become infrequent upon maturation. Here, we provide nanoscopic resolution of cilia in situ using large-scale electron microscopy volumes and immunostaining of mouse cerebella. In many granule cells, we found intracellular cilia, concealed from the external environment. Cilia were disassembled in differentiating granule cell neurons-in a process we call cilia deconstruction-distinct from premitotic cilia resorption in proliferating progenitors. In differentiating granule cells, cilia deconstruction involved unique disassembly intermediates, and, as maturation progressed, mother centriolar docking at the plasma membrane. Unlike ciliated neurons in other brain regions, our results show the deconstruction of concealed cilia in differentiating granule cells, which might prevent mitogenic hedgehog responsiveness. Ciliary deconstruction could be paradigmatic of cilia removal during differentiation in other tissues.

As we move through the world, we see the same visual scenes from different perspectives. Although we experience perspective deformations, our perception of a scene remains stable. This raises the question of which neuronal representations in visual brain areas are perspective-tuned and which are invariant. Focusing on planar rotations, we introduce a mathematical framework based on the principle of equivariance, which asserts that an image rotation results in a corresponding rotation of neuronal representations, to explain how the same representation can range from being fully tuned to fully invariant. We applied this framework to large-scale simultaneous neuronal recordings from four visual cortical areas in mice, where we found that representations are both tuned and invariant but become more invariant across higher-order areas. While common deep convolutional neural networks show similar trends in orientation-invariance across layers, they are not rotation-equivariant. We propose that equivariance is a prevalent computation of populations of biological neurons to gradually achieve invariance through structured tuning.