Search for: All records

Fundamental principles underlying computation in multi-scale brain networks illustrate how multiple brain areas and their coordinated activity give rise to complex cognitive functions. Whereas brain activity has been studied at the micro- to meso-scale to reveal the connections between the dynamical patterns and the behaviors, investigations of neural population dynamics are mainly limited to single-scale analysis. Our goal is to develop a cross-scale dynamical model for the collective activity of neuronal populations. Here we introduce a bio-inspired deep learning approach, termed NeuroBondGraph Network (NBGNet), to capture cross-scale dynamics that can infer and map the neural data from multiple scales. Our model not only exhibits more than an 11-fold improvement in reconstruction accuracy, but also predicts synchronous neural activity and preserves correlated low-dimensional latent dynamics. We also show that the NBGNet robustly predicts held-out data across a long time scale (2 weeks) without retraining. We further validate the effective connectivity defined from our model by demonstrating that neural connectivity during motor behaviour agrees with the established neuroanatomical hierarchy of motor control in the literature. The NBGNet approach opens the door to revealing a comprehensive understanding of brain computation, where network mechanisms of multi-scale activity are critical. 

In this paper, we study ultra-weak discontinuous Galerkin methods with generalized numerical fluxes for multi-dimensional high order partial differential equations on both unstructured simplex and Cartesian meshes. The equations we consider as examples are the nonlinear convection-diffusion equation and the biharmonic equation. Optimal error estimates are obtained for both equations under certain conditions, and the key step is to carefully design global projections to eliminate numerical errors on the cell interface terms of ultra-weak schemes on general dimensions. The well-posedness and approximation capability of these global projections are obtained for arbitrary order polynomial space based on a wide class of generalized numerical fluxes on regular meshes. These projections can serve as general analytical tools to be naturally applied to a wide class of high order equations. Numerical experiments are conducted to demonstrate these theoretical results. 

Aluminum-doped porous silicon was produced by a molten salt reaction. 

Genetically encoded fluorescent protein and fluorogenic RNA sensors are indispensable tools for imaging biomolecules in cells. To expand the toolboxes and improve the generalizability and stability of this type of sensor, we report herein a genetically encoded fluorogenic DNA aptamer (GEFDA) sensor by linking a fluorogenic DNA aptamer for dimethylindole red with an ATP aptamer. The design enhances red fluorescence by 4-fold at 650 nm in the presence of ATP. Additionally, upon dimerization, it improves the signal-to-noise ratio by 2–3 folds. We further integrated the design into a plasmid to create a GEFDA sensor for sensing ATP in live bacterial and mammalian cells. This work expanded genetically encoded sensors by employing fluorogenic DNA aptamers, which offer enhanced stability over fluorogenic proteins and RNAs, providing a novel tool for real-time monitoring of an even broader range of small molecular metabolites in biological systems. 

Amorphous/crystalline high-entropy-alloy (HEA) composites show great promise as structural materials due to their exceptional mechanical properties. However, there is still a lack of understanding of the dynamic nanoindentation response of HEA composites at the atomic scale. Here, the mechanical behavior of amorphous/crystalline HEA composites under nanoindentation is investigated through a large-scale molecular dynamics simulation and a dislocation-based strength model, in terms of the indentation force, microstructural evolution, stress distribution, shear strain distribution, and surface topography. The results show that the uneven distribution of elements within the crystal leads to a strong heterogeneity of the surface tension during elastic deformation. The severe mismatch of the amorphous/crystalline interface combined with the rapid accumulation of elastic deformation energy causes a significant number of dislocation-based plastic deformation behaviors. The presence of surrounding dislocations inhibits the free slip of dislocations below the indenter, while the amorphous layer prevents the movement or disappearance of dislocations towards the substrate. A thin amorphous layer leads to great indentation force, and causes inconsistent stacking and movement patterns of surface atoms, resulting in local bulges and depressions at the macroscopic level. The increasing thickness of the amorphous layer hinders the extension of shear bands towards the lower part of the substrate. These findings shed light on the mechanical properties of amorphous/crystalline HEA composites and offer insights for the design of high-performance materials. 

Neural mechanisms and underlying directionality of signaling among brain regions depend on neural dynamics spanning multiple spatiotemporal scales of population activity. Despite recent advances in multimodal measurements of brain activity, there is no broadly accepted multiscale dynamical models for the collective activity represented in neural signals. Here we introduce a neurobiological-driven deep learning model, termedmultiscale neuraldynamicsneuralordinarydifferentialequation (msDyNODE), to describe multiscale brain communications governing cognition and behavior. We demonstrate that msDyNODE successfully captures multiscale activity using both simulations and electrophysiological experiments. The msDyNODE-derived causal interactions between recording locations and scales not only aligned well with the abstraction of the hierarchical neuroanatomy of the mammalian central nervous system but also exhibited behavioral dependences. This work offers a new approach for mechanistic multiscale studies of neural processes. 

The extracellular matrix (ECM) is a dynamic and complex microenvironment that modulates cell behavior and cell fate. Changes in ECM composition and architecture have been correlated with development, differentiation, and disease progression in various pathologies, including breast cancer [1]. Studies have shown that aligned fibers drive a pro-metastatic microenvironment, promoting the transformation of mammary epithelial cells into invasive ductal carcinomaviathe epithelial-to-mesenchymal transition (EMT) [2]. The impact of ECM orientation on breast cancer metabolism, however, is largely unknown. Here, we employ two non-invasive imaging techniques, fluorescence-lifetime imaging microscopy (FLIM) and intensity-based multiphoton microscopy, to assess the metabolic states of cancer cells cultured on ECM-mimicking nanofibers in a random and aligned orientation. By tracking the changes in the intrinsic fluorescence of nicotinamide adenine dinucleotide and flavin adenine dinucleotide, as well as expression levels of metastatic markers, we reveal how ECM fiber orientation alters cancer metabolism and EMT progression. Our study indicates that aligned cellular microenvironments play a key role in promoting metastatic phenotypes of breast cancer as evidenced by a more glycolytic metabolic signature on nanofiber scaffolds of aligned orientation compared to scaffolds of random orientation. This finding is particularly relevant for subsets of breast cancer marked by high levels of collagen remodeling (e.g. pregnancy associated breast cancer), and may serve as a platform for predicting clinical outcomes within these subsets [3–6].