Publications

Animals can recognize latent structures in their environment and apply this information to efficiently navigate the world. Several works argue that the brain supports these abilities by forming neural representations from which behaviorally relevant variables can be read out across contexts and tasks. However, it is unclear which features of neural activity facilitate downstream readout. Here we analytically determine the geometric properties of neural activity that govern linear readout generalization on a set of tasks sharing a common latent structure. We show that four statistics summarizing the dimensionality, factorization and correlation structures of neural activity determine generalization. Early in learning, optimal neural representations are lower dimensional and exhibit higher correlations between single units and task variables than late in learning. We support these predictions through biological and artificial neural data analysis. Our results tie the linearly decodable information in neural population activity to its geometry.

We introduce a biologically inspired, multilayer neural architecture composed of Rectified Spectral Units (ReSUs). Each ReSU projects a recent window of its input history onto a canonical direction obtained via canonical correlation analysis (CCA) of previously observed past-future input pairs, and then rectifies either its positive or negative component. By encoding canonical directions in synaptic weights and temporal filters, ReSUs implement a local, self-supervised algorithm for progressively constructing increasingly complex features.
To evaluate both computational power and biological fidelity, we trained a two-layer ReSU network in a self-supervised regime on translating natural scenes. First-layer units, each driven by a single pixel, developed temporal filters resembling those of Drosophila post-photoreceptor neurons (L1/L2 and L3), including their empirically observed adaptation to signal-to-noise ratio (SNR). Second-layer units, which pooled spatially over the first layer, became direction-selective -- analogous to T4 motion-detecting cells -- with learned synaptic weight patterns approximating those derived from connectomic reconstructions.
Together, these results suggest that ReSUs offer (i) a principled framework for modeling sensory circuits and (ii) a biologically grounded, backpropagation-free paradigm for constructing deep self-supervised neural networks.

Parkinson’s Disease (PD) is characterized by the loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc), which is associated with changes in microglia function. While age remains the biggest risk factor, the underlying molecular cause of PD onset and its concurrent neuroinflammation are not well understood. Many identified PD risk genes have been directly linked to dopamine neuron impairment, while others are linked to immune cell function. In this study, we found that the PD risk gene FAM49B is critically expressed in microglia of the human SNpc and is downregulated with age and PD. We utilized human and murine microglia cells to demonstrate the role of FAM49B in regulating fundamental microglial functions such as cytoskeletal maintenance, migration, surface adherence, energy homeostasis, autophagy, and, importantly, inflammatory response. Downregulation of microglial FAM49B, as observed in the SNpc of aging individuals, led to significant alterations in these cellular functions, which are associated with increased microglial activation. Thus, our study highlights novel cell-type-specific roles of FAM49B and provides a potential mechanism for susceptibility to neuroinflammation, and reactive gliosis observed in both PD and normal aging.

We present the Score-based Autoencoder for Multiscale Inference (SAMI), a method for unsupervised representation learning that combines the theoretical frameworks of diffusion models and VAEs. By unifying their respective evidence lower bounds, SAMI formulates a principled objective that learns representations through score-based guidance of the underlying diffusion process. The resulting representations automatically capture meaningful structure in the data: it recovers ground truth generative factors in our synthetic dataset, learns factorized, semantic latent dimensions from complex natural images, and encodes video sequences into latent trajectories that are straighter than those of alternative encoders, despite training exclusively on static images. Furthermore, SAMI can extract useful representations from pre-trained diffusion models with minimal additional training. Finally, the explicitly probabilistic formulation provides new ways to identify semantically meaningful axes in the absence of supervised labels, and its mathematical exactness allows us to make formal statements about the nature of the learned representation. Overall, these results indicate that implicit structural information in diffusion models can be made explicit and interpretable through synergistic combination with a variational autoencoder.

Neural activity is usually interpreted by imposing external labels (e.g., stimuli or position during locomotion) and decoding within that space (e.g. replay). While powerful, such supervision can mask structure in the data that do not correspond to the label. Unsupervised methods, in turn, often assume smooth latent dynamics and miss genuine discontinuities. We introduce a conceptually simple, computationally efficient latent variable model that infers both (i) the latent variables organizing population activity and (ii) whether their dynamics are continuous or fragmented in time. Fitting reduces to an expectation-maximization (EM) procedure that alternates two operations familiar to systems neuroscience—tuning-curve estimation and label decoding—without requiring external labels. Applied to rodent hippocampal spike recordings, the model reveals distinct population patterns at the same physical position that supervised spatial decoding fails to detect. While learned latents exhibit place-field-like tuning, their reactivation patterns are better distinguished by behavioral states. The model further identifies a continuity-fragmentation axis that characterizes population activities across sleep-wake brain states that is modulated by cholinergic inputs. By not relying on externally imposed spatial labels, our approach exposes structure that supervised approaches obscure and provides a powerful tool for datasets lacking behavioral tracking.

Pose estimation is a general problem in computer vision with wide applications. The relative orientation of a 3D reference object can be determined from a 3D rotated version of that object, or from a projection of the rotated object to a 2D planar image. This projection can be a perspective projection (the PnP problem) or an orthographic projection (the OnP problem). We restrict our attention here to the OnP problem and the full 3D pose estimation task (the EnP problem). Here we solve the least squares systems for both the error-free EnP and OnP problems in terms of the determinant ratio matrix (DRaM) approach. The noisy-data case can be addressed with a straightforward rotation correction scheme. While the SVD and optimal quaternion eigensystem methods solve the noisy EnP 3D-3D alignment exactly, the noisy 3D-2D orthographic (OnP) task has no known comparable closed form, and can be solved by DRaM-class methods. We note that while previous similar work has been presented in the literature exploiting both the QR decomposition and the Moore-Penrose pseudoinverse transformations, here we place these methods in a larger context that has not previously been fully recognized in the absence of the corresponding DRaM solution. We term this class of solutions as the DRaM family, and conduct comparisons of the behavior of the families of solutions for the EnP and OnP rotation estimation problems. Overall, this work presents both a new solution to the 3D and 2D orthographic pose estimation problems and provides valuable insight into these classes of problems. With hindsight, we are able to show that our DRaM solutions to the exact EnP and OnP problems possess derivations that could have been discovered in the time of Gauss, and in fact generalize to all analogous N-dimensional Euclidean pose estimation problems.

3D chromatin structure is critical for the regulation of gene expression during development. Here we used Micro-C assays at 100-bp resolution to map genome organization in Drosophila melanogaster throughout the first half of embryogenesis. These high-resolution contact maps reveal fine-scale features such as loops and boundaries delineating topologically associating domains. Notably, we observe that 3D chromatin structures form prior to zygotic genome activation and persist during successive mitotic cycles. Integrative analysis with 149 public chromatin immunoprecipitation sequencing (ChIP-seq) datasets identifies four classes of chromatin structuring elements, including a distinct group enriched for GAGA-associated factor (GAF) and Zelda binding, associated with developmental-gene regulation. These elements are mitotically retained and exhibit sequence and structure similarity between D. melanogaster and D. virilis. We propose that 3D chromatin organization in the pre-cellular embryo facilitates deployment of developmentally regulated genes during Drosophila embryogenesis.

The formation and maintenance of the finite mammalian ovarian reserve are critical for fertility and species survival. Genetic and developmental studies have uncovered various mechanisms underlying oocyte development and maturation, revealing two curious features of the ovarian germline: (a) The establishment of the follicle reserve involves an initial massive overproduction of oocyte precursors, and (b) the total number of ovulated oocytes across an animal's fertile lifetime is a very small proportion of the initial ovarian reserve. Many have proposed that this indicates the existence of selective quality control to ensure gamete fitness. Here, we review the findings underlying the hypotheses for germline quality control during prepubertal development, homeostatic fertility, and reproductive aging. We evaluate whether the existing evidence base distinguishes the active selection of specific germ cell subsets from neutral dynamics. Throughout, we discuss strategies for applying statistical frameworks to evaluate selection in oogenesis and the implications of neutrality versus selection at various points in oocyte development.

We model sensory streams as observations from high-dimensional stochastic dynamical systems and conceptualize sensory neurons as self-supervised learners of compact representations of such dynamics. From prior experience, neurons learn coherent sets-regions of stimulus state space whose trajectories evolve cohesively over finite times-and assign membership indices to new stimuli. Coherent sets are identified via spectral clustering of the stochastic Koopman operator (SKO), where the sign pattern of a subdominant singular function partitions the state space into minimally coupled regions. For multivariate Ornstein-Uhlenbeck processes, this singular function reduces to a linear projection onto the dominant singular vector of the whitened state-transition matrix. Encoding this singular vector as a receptive field enables neurons to compute membership indices via the projection sign in a biologically plausible manner. Each neuron detects either a predictive coherent set (stimuli with common futures) or a retrospective coherent set (stimuli with common pasts), suggesting a functional dichotomy among neurons. Since neurons lack access to explicit dynamical equations, the requisite singular vectors must be estimated directly from data, for example, via past-future canonical correlation analysis on lag-vector representations-an approach that naturally extends to nonlinear dynamics. This framework provides a novel account of neuronal temporal filtering, the ubiquity of rectification in neural responses, and known functional dichotomies. Coherent-set clustering thus emerges as a fundamental computation underlying sensory processing and transferable to bio-inspired artificial systems.

Viral infections can induce prolonged changes in innate immunity. Here, we use blood samples from a human influenza H3N2 challenge study (NCT03883113) to perform comprehensive multi-omics analyses. We detect remodeling of immune programs in circulating innate immune cells that persist after resolution of the infection. We find changes associated with suppressed inflammation, including decreased cytokine and AP-1 gene expression as well as decreased accessibility at AP-1 targets and interleukin-related gene promoter regions. We also find decreased histone deacetylase gene expression, increased MAP kinase gene expression, and increased accessibility at interferon-related gene promoter regions. Genes involved in inflammation and methylation remodeling show modulation of gene-chromatin site regulatory circuit activity. These results reveal a coordinated rewiring of the molecular landscape in innate immune cells induced by mild influenza virus infection.