Motor network gamma oscillations in chronic home recordings predict dyskinesia in Parkinson's disease.

In Parkinson's disease, imbalances between 'antikinetic' and 'prokinetic' patterns of neuronal oscillatory activity are related to motor dysfunction. Invasive brain recordings from the motor network have suggested that medical or surgical therapy can promote a prokinetic state by inducing narrowband gamma rhythms (65-90 Hz). Excessive narrowband gamma in the motor cortex promotes dyskinesia in rodent models, but the relationship between narrowband gamma and dyskinesia in humans has not been well established. To assess this relationship, we used a sensing-enabled deep brain stimulator system, attached to both motor cortex and basal ganglia (subthalamic or pallidal) leads, paired with wearable devices that continuously tracked motor signs in the contralateral upper limbs. We recorded 984 h of multisite field potentials in 30 hemispheres of 16 subjects with Parkinson's disease (2/16 female, mean age 57 ± 12 years) while at home on usual antiparkinsonian medications. Recordings were done 2-4 weeks after implantation, prior to starting therapeutic stimulation. Narrowband gamma was detected in the precentral gyrus, subthalamic nucleus or both structures on at least one side of 92% of subjects with a clinical history of dyskinesia. Narrowband gamma was not detected in the globus pallidus. Narrowband gamma spectral power in both structures co-fluctuated similarly with contralateral wearable dyskinesia scores (mean correlation coefficient of ρ = 0.48 with a range of 0.12-0.82 for cortex, ρ = 0.53 with a range of 0.5-0.77 for subthalamic nucleus). Stratification analysis showed the correlations were not driven by outlier values, and narrowband gamma could distinguish 'on' periods with dyskinesia from 'on' periods without dyskinesia. Time lag comparisons confirmed that gamma oscillations herald dyskinesia onset without a time lag in either structure when using 2-min epochs. A linear model incorporating the three oscillatory bands (beta, theta/alpha and narrowband gamma) increased the predictive power of dyskinesia for several subject hemispheres. We further identified spectrally distinct oscillations in the low gamma range (40-60 Hz) in three subjects, but the relationship of low gamma oscillations to dyskinesia was variable. Our findings support the hypothesis that excessive oscillatory activity at 65-90 Hz in the motor network tracks with dyskinesia similarly across both structures, without a detectable time lag. This rhythm may serve as a promising control signal for closed-loop deep brain stimulation using either cortical or subthalamic detection.

Definitions, methods, and applications in interpretable machine learning.

Machine-learning models have demonstrated great success in learning complex patterns that enable them to make predictions about unobserved data. In addition to using models for prediction, the ability to interpret what a model has learned is receiving an increasing amount of attention. However, this increased focus has led to considerable confusion about the notion of interpretability. In particular, it is unclear how the wide array of proposed interpretation methods are related and what common concepts can be used to evaluate them. We aim to address these concerns by defining interpretability in the context of machine learning and introducing the predictive, descriptive, relevant (PDR) framework for discussing interpretations. The PDR framework provides 3 overarching desiderata for evaluation: predictive accuracy, descriptive accuracy, and relevancy, with relevancy judged relative to a human audience. Moreover, to help manage the deluge of interpretation methods, we introduce a categorization of existing techniques into model-based and post hoc categories, with subgroups including sparsity, modularity, and simulatability. To demonstrate how practitioners can use the PDR framework to evaluate and understand interpretations, we provide numerous real-world examples. These examples highlight the often underappreciated role played by human audiences in discussions of interpretability. Finally, based on our framework, we discuss limitations of existing methods and directions for future work. We hope that this work will provide a common vocabulary that will make it easier for both practitioners and researchers to discuss and choose from the full range of interpretation methods.

Visual physiology of the layer 4 cortical circuit in silico.

Despite advances in experimental techniques and accumulation of large datasets concerning the composition and properties of the cortex, quantitative modeling of cortical circuits under in-vivo-like conditions remains challenging. Here we report and publicly release a biophysically detailed circuit model of layer 4 in the mouse primary visual cortex, receiving thalamo-cortical visual inputs. The 45,000-neuron model was subjected to a battery of visual stimuli, and results were compared to published work and new in vivo experiments. Simulations reproduced a variety of observations, including effects of optogenetic perturbations. Critical to the agreement between responses in silico and in vivo were the rules of functional synaptic connectivity between neurons. Interestingly, after extreme simplification the model still performed satisfactorily on many measurements, although quantitative agreement with experiments suffered. These results emphasize the importance of functional rules of cortical wiring and enable a next generation of data-driven models of in vivo neural activity and computations.

Interpretable video-based tracking and quantification of parkinsonism clinical motor states.

Quantification of motor symptom progression in Parkinson's disease (PD) patients is crucial for assessing disease progression and for optimizing therapeutic interventions, such as dopaminergic medications and deep brain stimulation. Cumulative and heuristic clinical experience has identified various clinical signs associated with PD severity, but these are neither objectively quantifiable nor robustly validated. Video-based objective symptom quantification enabled by machine learning (ML) introduces a potential solution. However, video-based diagnostic tools often have implementation challenges due to expensive and inaccessible technology, and typical "black-box" ML implementations are not tailored to be clinically interpretable. Here, we address these needs by releasing a comprehensive kinematic dataset and developing an interpretable video-based framework that predicts high versus low PD motor symptom severity according to MDS-UPDRS Part III metrics. This data driven approach validated and robustly quantified canonical movement features and identified new clinical insights, not previously appreciated as related to clinical severity, including pinkie finger movements and lower limb and axial features of gait. Our framework is enabled by retrospective, single-view, seconds-long videos recorded on consumer-grade devices such as smartphones, tablets, and digital cameras, thereby eliminating the requirement for specialized equipment. Following interpretable ML principles, our framework enforces robustness and interpretability by integrating (1) automatic, data-driven kinematic metric evaluation guided by pre-defined digital features of movement, (2) combination of bi-domain (body and hand) kinematic features, and (3) sparsity-inducing and stability-driven ML analysis with simple-to-interpret models. These elements ensure that the proposed framework quantifies clinically meaningful motor features useful for both ML predictions and clinical analysis.

Data-driven fine-grained region discovery in the mouse brain with transformers.

Technologies such as spatial transcriptomics offer unique opportunities to define the spatial organization of the mouse brain. We developed an unsupervised training scheme and novel transformer-based deep learning architecture to detect spatial domains across the whole mouse brain using spatial transcriptomics data. Our model learns local representations of molecular and cellular statistical patterns which can be clustered to identify spatial domains within the brain from coarse to fine-grained. Discovered domains are spatially regular, even with several hundreds of spatial clusters. They are also consistent with existing anatomical ontologies such as the Allen Mouse Brain Common Coordinate Framework version 3 (CCFv3) and can be visually interpreted at the cell type or transcript level. We demonstrate our method can be used to identify previously uncatalogued subregions, such as in the midbrain, where we uncover gradients of inhibitory neuron complexity and abundance. Notably, these subregions cannot be discovered using other methods. We apply our method to a separate multi-animal whole-brain spatial transcriptomic dataset and show that our method can also robustly integrate spatial domains across animals.

Machine Learning for Uncovering Biological Insights in Spatial Transcriptomics Data.

Development and homeostasis in multicellular systems both require exquisite control over spatial molecular pattern formation and maintenance. Advances in spatially-resolved and high-throughput molecular imaging methods such as multiplexed immunofluorescence and spatial transcriptomics (ST) provide exciting new opportunities to augment our fundamental understanding of these processes in health and disease. The large and complex datasets resulting from these techniques, particularly ST, have led to rapid development of innovative machine learning (ML) tools primarily based on deep learning techniques. These ML tools are now increasingly featured in integrated experimental and computational workflows to disentangle signals from noise in complex biological systems. However, it can be difficult to understand and balance the different implicit assumptions and methodologies of a rapidly expanding toolbox of analytical tools in ST. To address this, we summarize major ST analysis goals that ML can help address and current analysis trends. We also describe four major data science concepts and related heuristics that can help guide practitioners in their choices of the right tools for the right biological questions.

Structural Compression of Convolutional Neural Networks with Applications in Interpretability.

Deep convolutional neural networks (CNNs) have been successful in many tasks in machine vision, however, millions of weights in the form of thousands of convolutional filters in CNNs make them difficult for human interpretation or understanding in science. In this article, we introduce a greedy structural compression scheme to obtain smaller and more interpretable CNNs, while achieving close to original accuracy. The compression is based on pruning filters with the least contribution to the classification accuracy or the lowest Classification Accuracy Reduction (CAR) importance index. We demonstrate the interpretability of CAR-compressed CNNs by showing that our algorithm prunes filters with visually redundant functionalities such as color filters. These compressed networks are easier to interpret because they retain the filter diversity of uncompressed networks with an order of magnitude fewer filters. Finally, a variant of CAR is introduced to quantify the importance of each image category to each CNN filter. Specifically, the most and the least important class labels are shown to be meaningful interpretations of each filter.

Superficial Bound of the Depth Limit of Two-Photon Imaging in Mouse Brain.

Two-photon fluorescence microscopy has been used extensively to probe the structure and functions of cells in living biological tissue. Two-photon excitation generates fluorescence from the focal plane, but also from outside the focal plane, with out-of-focus fluorescence increasing as the focus is pushed deeper into tissue. It has been postulated that the two-photon depth limit, beyond which results become inaccurate, is where in-focus and out-of-focus fluorescence are equal, which we term the balance depth. Calculations suggest that the balance depth should be at ∼600 µm in mouse cortex. Neither the two-photon depth limit nor the balance depth have been measured in brain tissue. We found the depth limit and balance depth of two-photon excitation in mice with GCaMP6 indicator expression in all layers of visual cortex, by comparing near-simultaneous two-photon and three-photon excitation. Two-photon and three-photon results from superficial locations were almost identical. two-photon results were inaccurate beyond the balance depth, consistent with the depth limit matching the balance depth for two-photon excitation. However, the two-photon depth limit and balance depth were at 450 µm, shallower than predicted by calculations. Our results were from tissue with a largely homogenous distribution of fluorophores. The expected balance depth is deeper in tissue with fewer fluorophores outside the focal plane and our results therefore establish a superficial bound on the two-photon depth limit in mouse visual cortex.

Systematic Integration of Structural and Functional Data into Multi-scale Models of Mouse Primary Visual Cortex.

Structural rules underlying functional properties of cortical circuits are poorly understood. To explore these rules systematically, we integrated information from extensive literature curation and large-scale experimental surveys into a data-driven, biologically realistic simulation of the awake mouse primary visual cortex. The model was constructed at two levels of granularity, using either biophysically detailed or point neurons. Both variants have identical network connectivity and were compared to each other and to experimental recordings of visual-driven neural activity. While tuning these networks to recapitulate experimental data, we identified rules governing cell-class-specific connectivity and synaptic strengths. These structural constraints constitute hypotheses that can be tested experimentally. Despite their distinct single-cell abstraction, both spatially extended and point models perform similarly at the level of firing rate distributions for the questions we investigated. All data and models are freely available as a resource for the community.