Foliation divides the mammalian cerebellum into structurally distinct subdivisions, including the concave sulcus and the convex apex. Purkinje cell (PC) dendritic morphology varies between subdivisions and changes significantly ontogenetically. Since dendritic morphology both enables and limits sensory-motor circuit function, it is important to understand how neuronal architectures differ between brain regions. This study employed quantitative confocal microcopy to reconstruct dendritic arbors of cerebellar PCs expressing green fluorescent protein and compared arbor morphology between PCs of sulcus and apex in young and old mice. Arbors were digitized from high z-resolution (0.25 µm) image stacks using an adaptation of Neurolucida's (MBF Bioscience) continuous contour tracing tool, designed for drawing neuronal somata. Reconstructed morphologies reveal that dendritic arbors of sulcus and apex exhibit profound differences. In sulcus, 72% of the young PC population possesses two primary dendrites, whereas in apex, only 28% do. Spatial constraints in the young sulcus cause significantly more dendritic arbor overlap than in young apex, a distinction that disappears in adulthood. However, adult sulcus PC arbors develop a greater number of branch crossings. These results suggest developmental neuronal plasticity that enables cerebellar PCs to attain correct functional adult architecture under different spatial constraints.

Regional differences in dendritic architecture can influence connectivity and dendritic signal integration, with possible consequences for neuronal computation. In the cerebellum, analyses of Purkinje cells (PCs), which are functionally critical as they provide the sole output of the cerebellar cortex, have suggested that the cerebellar cortex is not uniform in structure as traditionally assumed. However, the limitations of traditional staining methods and microscopy capabilities have presented difficulties in investigating possible local variations in PC morphology. To address this question, we used male mice expressing green fluorescent protein selectively in PCs. Using Neurolucida 360 with confocal image stacks, we reconstructed dendritic arbors of PCs residing in lobule V (anterior) and lobule IX (posterior) of the vermis. We then analyzed morphologies of individual arbors and the structure of the assembled "jungle," comparing these features across anatomical locations and age groups. Strikingly, we found that in lobule IX, half of the reconstructed PCs had two primary dendrites emanating from their soma, whereas fewer than a quarter showed this characteristic in lobule V. Furthermore, PCs in lobule V showed more efficient spatial occupancy compared to lobule IX, as well as greater packing density and increased arbor overlap in the adult. When analyzing complete ensembles of PC arbors, we also observed "hot spots" of increased dendritic density in lobule V, whereas lobule IX showed a more homogeneous spread of dendrites. These differences suggest that input patterns and/or physiology of PCs could likewise differ along the vermis, with possible implications for cerebellar function.

Visual Abstract

The robustness of the visual system lies in its ability to perceive degraded images. This is achieved through interacting bottom-up and top-down pathways that process the visual input in concordance with stored prior information. The interaction mechanism by which they integrate visual input and prior information is still enigmatic. We present a new approach using deep neural network (DNN) representation to reveal the effects of such integration on degraded visual inputs. We transformed measured human brain activity resulting from viewing blurred images to the hierarchical representation space derived from a feedforward DNN. Transformed representations were found to veer towards the original non-blurred image and away from the blurred stimulus image. This indicated deblurring or sharpening in the neural representation, and possibly in our perception, from top-down modulation. We anticipate these results will help unravel the interplay mechanism between bottom-up and top-down pathways, leading to more comprehensive models of vision.

Fig. 1. Observed incidence of mortality vs calibrated predicted probability of mortality amongst patients in the test set (n¼19 394). Predicted probabilities have been calibrated by applying the histogram binning technique in the validation set.

Background The COVID-19 global pandemic has increased the burden of mental illness on Canadian adults. However, the complex combination of demographic, economic, and lifestyle factors and perceived health risks contributing to patterns of anxiety and depression has not been explored. Objective The aim of this study is to harness flexible machine learning methods to identify constellations of factors related to symptoms of mental illness and to understand their changes over time during the COVID-19 pandemic. Methods Cross-sectional samples of Canadian adults (aged ≥18 years) completed web-based surveys in 6 waves from May to December 2020 (N=6021), and quota sampling strategies were used to match the English-speaking Canadian population in age, gender, and region. The surveys measured anxiety and depression symptoms, sociodemographic characteristics, substance use, and perceived COVID-19 risks and worries. First, principal component analysis was used to condense highly comorbid anxiety and depression symptoms into a single data-driven measure of emotional distress. Second, eXtreme Gradient Boosting (XGBoost), a machine learning algorithm that can model nonlinear and interactive relationships, was used to regress this measure on all included explanatory variables. Variable importance and effects across time were explored using SHapley Additive exPlanations (SHAP). Results Principal component analysis of responses to 9 anxiety and depression questions on an ordinal scale revealed a primary latent factor, termed “emotional distress,” that explained 76% of the variation in all 9 measures. Our XGBoost model explained a substantial proportion of variance in emotional distress (r2=0.39). The 3 most important items predicting elevated emotional distress were increased worries about finances (SHAP=0.17), worries about getting COVID-19 (SHAP=0.17), and younger age (SHAP=0.13). Hopefulness was associated with emotional distress and moderated the impacts of several other factors. Predicted emotional distress exhibited a nonlinear pattern over time, with the highest predicted symptoms in May and November and the lowest in June. Conclusions Our results highlight factors that may exacerbate emotional distress during the current pandemic and possible future pandemics, including a role of hopefulness in moderating distressing effects of other factors. The pandemic disproportionately affected emotional distress among younger adults and those economically impacted.

Adaptation in the sensory-mechanical loop during locomotion is a powerful mechanism that allows organisms to survive in different conditions and environments. Motile animals need to alter motion patterns in different environments. For example, crocodiles and other animals can walk on solid ground but switch to swimming in water beds. The nematode Caenorhabditis elegans also shows adaptability by employing thrashing behaviour in low viscosity media and crawling in high viscosity media. The mechanism that enables this adaptability is an active area of research. It has been attributed previously to neuro-modulation by dopamine and serotonin. The aim of this study is to physiologically investigate the neuronal mechanisms of modulation of locomotion by dopamine. The results suggest that the mechanosensory properties of the dopaminergic neurons PDE are not limited to touch sensation, but to surrounding environment resistance as well. The significance of such characterization is improving our understanding of dopamine gait switching which gets impaired in Parkinson's disease.

An applied problem facing all areas of data science is harmonizing data sources. Joining data from multiple origins with unmapped and only partially overlapping features is a prerequisite to developing and testing robust, generalizable algorithms, especially in health care. We approach this issue in the common but difficult case of numeric features such as nearly Gaussian and binary features, where unit changes and variable shift make simple matching of univariate summaries unsuccessful. We develop two novel procedures to address this problem. First, we demonstrate multiple methods of "fingerprinting" a feature based on its associations to other features. In the setting of even modest prior information, this allows most shared features to be accurately identified. Second, we demonstrate a deep learning algorithm for translation between databases. Unlike prior approaches, our algorithm takes advantage of discovered mappings while identifying surrogates for unshared features and learning transformations. In synthetic and real-world experiments using two electronic health record databases, our algorithms outperform existing baselines for matching variable sets, while jointly learning to impute unshared or transformed variables.