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. How can we capture the incredible complexity of brain circuits in quantitative models, and what can such models teach us about mechanisms underlying brain activity? To answer these questions, we set out to build extensive, bio-realistic models of brain circuitry by employing systematic datasets on brain structure and function. Here we report the first modeling results of this project, focusing on the layer 4 of the primary visual cortex (V1) of the mouse. Our simulations reproduced a variety of experimental observations in response to a large battery of visual stimuli. The results elucidated circuit mechanisms determining patters of neuronal activity in layer 4 –in particular, the roles of feedforward thalamic inputs and specific patterns of intracortical connectivity in producing tuning of neuronal responses to the orientation of motion. Simplification of neuronal models led to specific deficiencies in reproducing experimental data, giving insights into how biological details contribute to various aspects of brain activity. To enable future development of more sophisticated models, we make the software code, the model, and simulation results publicly available.

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.

The rise of biomedical foundation models creates new hurdles in model testing and authorization, given their broad capabilities and susceptibility to complex distribution shifts. We suggest tailoring robustness tests according to task-dependent priorities and propose to integrate granular notions of robustness in a predefined specification to guide implementation. Our approach facilitates the standardization of robustness assessments in the model lifecycle and connects abstract AI regulatory frameworks with concrete testing procedures.

In Parkinson's disease, invasive brain recordings show that dopaminergic medication can induce narrowband gamma rhythms in the motor cortex and subthalamic nucleus, which co-fluctuate with dyskinesia scores. Deep brain stimulation can entrain these gamma oscillations to a subharmonic stimulation frequency. However, the incidence of entrainment during chronic therapeutic stimulation, its relationship to the basal ganglia stimulation site, and its effect on dyskinesia remain unknown. Determine whether the behavioral effects and statistical properties of levodopa-induced gamma oscillations are altered when entrained with deep brain stimulation. We used a sensing-enabled deep brain stimulator system, attached to both motor cortex and subthalamic (n = 15) or pallidal (n = 5) leads, to record 993 h of multisite field potentials, with 656 h recorded prior to initiating stimulation. 13 subjects (20 hemispheres) with Parkinson's disease (1/13 female, mean age 59 ± 9 years) streamed data while at home on their usual antiparkinsonian medication. Recordings during stimulation occurred at least five months after initiating stimulation. Cortical entrained gamma oscillations were detected in 4/5 hemispheres undergoing pallidal stimulation and 12/15 hemispheres undergoing subthalamic stimulation. Entraining levodopa-induced gamma oscillations at either site reduced their prodyskinetic effects. Cortical entrained gamma oscillations had reduced variance in peak frequency, increased spectral power, and higher variance in spectral power than levodopa-induced gamma oscillations. Stimulation-entrained gamma oscillations are functionally and physiologically distinct from levodopa-induced gamma oscillations that occur in the absence of deep brain stimulation. Understanding the discrepancies between types of gamma oscillations may improve programming protocols.

Spatial transcriptomics offers unique opportunities to define the spatial organization of tissues and organs, such as the mouse brain. We address a key bottleneck in the analysis of organ-scale spatial transcriptomic data by establishing a workflow for self-supervised spatial domain detection that is scalable to multimillion-cell datasets. This workflow uses a self-supervised framework for learning latent representations of tissue spatial domains or niches. We use an encoder-decoder architecture, which we named CellTransformer, to hierarchically learn higher-order tissue features from lower-level cellular and molecular statistical patterns. Coupling our representation learning workflow with minibatched GPU-accelerated clustering algorithms allows us to scale to multi-million cell MERFISH datasets where other methods cannot. CellTransformer is effective at integrating cells across tissue sections, identifying domains highly similar to ones in existing ontologies such as Allen Mouse Brain Common Coordinate Framework (CCF) while allowing discovery of hundreds of uncataloged areas with minimal loss of domain spatial coherence. CellTransformer domains recapitulate previous neuroanatomical studies of areas in the subiculum and superior colliculus and characterize putatively uncataloged subregions in subcortical areas, which currently lack subregion annotation. CellTransformer is also capable of domain discovery in whole-brain Slide-seqV2 datasets. Our workflows enable complex multi-animal analyses, achieving nearly perfect consistency of up to 100 spatial domains in a dataset of four individual mice with nine million cells across more than 200 tissue sections. CellTransformer advances the state of the art for spatial transcriptomics by providing a performant solution for the detection of fine-grained tissue domains from spatial transcriptomics data.

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.