15064 research outputs found
Sort by
Fast Capture of Cell-Level Provenance in Numpy
Get PDFEffective provenance tracking enhances reproducibility, governance, and data quality in array workflows. However, significant challenges arise in capturing this provenance, including: (1) rapidly evolving APIs, (2) diverse operation types, and (3) large-scale datasets. To address these challenges, this paper presents a prototype annotation system designed for arrays, which captures cell-level provenance specifically within the numpy library. With this prototype, we explore straightforward memory optimizations that substantially reduce annotation latency. We envision this provenance capture approach for arrays as part of a broader governance system for tracking for structured data workflows and diverse data science applications
Teeth outside the jaw: Evolution and development of the toothed head clasper in chimaeras
Get PDFChimaeras (Holocephali) are an understudied group of mostly deep-ocean cartilaginous fishes (Chondrichthyes) with unique characteristics that distinguish them from their distant relatives, sharks, skates, and rays. Unlike sharks, chimaeras lack scales and do not have serially replacing rows of serrated teeth crowned with enameloid. Instead, they possess a fused dentition of dentine tooth plates. Additionally, male chimaeras develop an articulated cartilaginous facial appendage, the tenaculum, which is covered in an arcade of tooth-like structures. These extraoral teeth remain poorly understood, and their evolutionary origin is unclear. We investigate the development of the tenaculum and its teeth throughout the ontogeny of the Spotted Ratfish, Hydrolagus colliei, to assess homology and convergence between this novel craniofacial feature and oral jaws. Our study aims to 1) describe the development of the tenaculum, 2) assess tenaculum tooth development in comparison to oral teeth and denticles, and 3) characterize the genes and tissues responsible for tenaculum tooth emergence. We found that juvenile male chimaeras develop a full tenaculum before tooth development is complete and that only mature males possess a fully toothed tenaculum. These extraoral teeth emerge from within the tenaculum rather than from the surrounding epithelium. We integrate our developmental data with fossil evidence of the tenacular dentition from the Carboniferous holocephalan Helodus simplex. Our findings show that the tenaculum is closely associated with the upper jaw and that tenacular dentition resembles separate shark-like oral tooth whorls more than modified dermal denticles
Sweat-sensitive adaptive warm clothing
Get PDFThermal regulation in warm clothing is essential for enhancing human comfort in cold environments. However, traditional warm clothing lacks the ability to adapt to dynamic changes in the human body’s microenvironment. Here, we present an adaptive warm cloth, featuring a filling made of a natural bacterial cellulose membrane that responds to human sweating. The cloth’s thickness automatically adjusts from 13 millimeters (under low humidity and no sweating conditions) to 2 millimeters (under high humidity and sweating conditions), expanding the thermal regulation capability by 82.8% compared to traditional warm clothing with an unchanged thickness of 13 millimeters. Modeling results further suggest that deploying this adaptive warm clothing across 20 cities in China could extend the duration of the no thermal stress zone by an average of 7.5 hours. Combining exceptional thermal regulation, high stability, and scalability, this clothing represents a notable supplement to existing thermal management technologies
Stochastic noise can be helpful for variational quantum algorithms
Get PDFSaddle points constitute a crucial challenge for first-order gradient descent algorithms. In notions of classical machine learning, they are avoided, for example, by means of stochastic gradient descent methods. In this work, we provide evidence that the saddle-points problem can be naturally avoided in variational quantum algorithms by exploiting the presence of stochasticity. We prove convergence guarantees and present practical examples in numerical simulations and on quantum hardware. We argue that the natural stochasticity of variational algorithms can be beneficial for avoiding strict saddle points, i.e., those saddle points with at least one negative Hessian eigenvalue. This insight that some levels of shot noise could help is expected to add a new perspective to notions of near-term variational quantum algorithms.</p
Development of Coarse-Grained Lipid Force Fields Based on a Graph Neural NetworkClick to copy article link
Get PDFCoarse-grained (CG) lipid models enable efficient simulations of large-scale membrane events. However, achieving both speed and atomic-level accuracy remains challenging. Graph neural networks (GNNs) trained on all-atom (AA) simulations can serve as CG force fields, which have demonstrated success in CG simulations of proteins. Herein, we built data sets of AA simulations of DOPC, DOPS, and mixed DOPC/DOPS lipid bilayers and developed the first GNN-based CG lipid models based on the TorchMD-GN architecture. The CG lipid models reproduce the structural correlations of the AA simulations, accelerate the lipid dynamics by 9.4 times, and exhibit some degree of temperature transferability. Moreover, we demonstrate that training CG models on lipid bicelles enhances the performance of models in the lipid self-assembly and vesicle simulations. Our findings indicate that GNN-based CG lipid force fields show promise as a powerful approach for large-scale membrane simulations
Is AI Our New Addiction? AI and Mental Health
No full textAs Americans have begun to experience unprecedented rates of loneliness and social isolation during the 21st century, a new threat to mental health is emerging: artificial intelligence addiction. The following is a series of posts that will dive deeper into how AI chatbots can co-opt and ultimately replace human connection, leading to greater rates of depression and AI-related mental disorders. Drawing on emerging psychiatric research on Generative Artificial Intelligence Disorder (GAID), case studies of people who have suffered from AI-driven psychosis, and neurological findings regarding AI’s impact on cognitive function, these posts will reveal a mental health crisis in the making. Through exploring the loneliness epidemic, the psychology of AI addiction, corporate policies, and potential solutions, this series argues that intervention is necessary to mitigate the risk of technological dependence before societal isolation becomes the new normal
Biomechanics of the mandibular middle ear of the cynodont <i>Thrinaxodon</i> and the evolution of mammal hearing
No full textThe middle ear of mammals is a major functional innovation, distinctive in that it is detached from the mandible and has a tympanic membrane supported by a ring-like ectotympanic. These novelties of the middle ear have enabled modern mammals to develop more sensitive hearing than all other tetrapods, especially at higher frequencies. Fossils from recent decades have clarified the evolution of the detached middle ear from the jaw bones of Paleozoic therapsids and Mesozoic cynodonts, and the evolution of the tympanum. These discoveries make it possible to answer important questions about the functional significance of these features. Here, we evaluate the relative hearing efficacy of a well-known cynodont precursor to mammals, Thrinaxodon liorhinus. Using finite element analysis (FEA), we calculated the harmonic response of the Thrinaxodon ear to bone-conducted and airborne sound and estimated the sound pressure level (SPL) at the stapedial footplate across a broad range of frequencies. We provide evidence that airborne sound received at the tympanum was the most effective mode of sound reception in Thrinaxodon. In contrast, bone conducted sound through the mandibular bones barely met our estimated hearing threshold. Our findings suggest that, like modern mammals, cynodonts were already reliant on a soft tissue tympanum to receive airborne sound, albeit with limited sensitivity to high frequencies. This is a detailed biomechanical evaluation of tympanum function in the cynodont predecessors of mammals and yields insight into the sequence of functional innovations during the evolution of mammal hearing
Imaging and Control of Mesoscale Structures in Two-Dimensional Materials
Get PDFBoundaries, interfaces, and material heterogeneity are ubiquitous across chemistry and physics, giving rise to emergent phenomena such as localized states and modified dynamics and enabling material applications ranging from digital memory to heterogeneous catalysis. As demand for more compact and efficient devices increases, research efforts have been extended toward realizing these structures within individual nano- and micro-scale components. Intrinsic structures such as ferroic domains and artificial junctions where discontinuities in crystalline phase or composition are engineered through synthetic and lithographic methods have both been explored. Practical implementation of these types of structures in device applications requires innovation in the methodology used for their study and manipulation. Multimodal approaches that can simultaneously capture spatial, energetic, and temporal information to characterize functional interfaces and investigate their emergent properties are particularly desirable. This dissertation addresses each of these goals through the investigation of two-dimensional materials with photoemission electron microscopy (PEEM) and other spectroscopic and microscopy techniques. Chapter 2 details the working principles of PEEM with extra emphasis on polarization dependent experiments. Chapter 3 discusses the ultrahigh vacuum and laser experimental apparatus used in conducting PEEM measurements. Chapter 4 extends these methods to the study of the antiferroelectric domains in β′-In2Se3. Excitation energy dependent measurements and geometric analysis of domain arrangements connect the observed optical responses to the atomic displacements in the material. Chapter 5 presents a laser-induced wrinkle-mediated phase transition pathway for conversion between β′ and α-In2Se3. With thermal annealing steps, this transition enables repeated cycling between phases and strain accumulated throughout the process results in the formation of multiphase heterostructures and the rearrangement of domains. In Chapter 6 the behavior of Cu–Cl Ruddlesden-Popper perovskites under ultraviolet illumination is investigated. These materials undergo a modification where ordered grooves with well defined orientations are etched on their surfaces. The symmetry of the formed pattern is controlled by the organic cation spacer through electron-phonon coupling. Finally, Chapters 7 and 8 explore the impact that these material modifications and intrinsic anisotropy have on coherent light-matter interactions and dynamics in two-dimensional materials
The Roles of the Basement Membrane and Semaphorin Signaling in <i> Drosophila </i> Egg Chamber Elongation
No full textThe morphogenesis of epithelial tissues is a highly dynamic process involving complex interactions between the cells in the tissue and between the cells and their physical environment. Basement membranes are specialized extracellular matrices that line the basal surfaces of epithelia and provide structural and mechanical integrity to the tissue. During morphogenesis, basement membranes can be structurally and mechanically patterned such that they can sculpt epithelia into complex 3D shapes, but the mechanical roles of basement membranes have only recently begun to be understood. The Drosophila melanogaster egg chamber is a genetically, experimentally, and visually tractable system to ask how an epithelium patterns its basement membrane through various mechanisms that allows the matrix to guide tissue shape changes. Egg chambers start as small spherical structures and grow nearly 1000 times in size, through which they become progressively elongated. During egg chamber morphogenesis, follicular epithelial cells synthesize and secrete their own basement membrane proteins and create a complex ultrastructural architecture by synergizing this secretion with a collective migration of the cells that results in a polarized array of fibrils. While these fibrils are known to be required for egg chamber elongation, whether they have a role in mechanically patterning the basement membrane or have some other secondary role to support tissue elongation has been debated. Additionally, much work has gone into uncovering how the collective migration of the follicle cells is governed. A set of planar polarized transmembrane proteins form two signaling axes: Fat2 and Lar, and Sema5c and PlexA. While there is increasing mechanistic insight into Fat2/Lar signaling, relatively little is known about how Sema5c and PlexA function. During my doctoral work, I have investigated the role of these basement membrane fibrils in promoting egg chamber elongation as well as elucidating the function of Sema5c in follicle cell migration. I have organized my work into three data chapters (2-4). In Chapter 2 I show that the elongation defect associated with Sema5c egg chambers is due to a reduction in basement membrane fibrils, and I also identify a genetic condition that suggests that our current understanding of the requirements of specific basement membrane architectures and mechanical properties is not sufficient to explain why an egg chamber that lacks these can properly elongate. In Chapters 3 and 4, I describe candidate-based and unbiased approaches in identifying new proteins that work with Sema5c. In Chapter 3, I describe my work investigating two candidates to act downstream of Sema5c, the F-actin disassembly factor Mical and the small GTPase Rap1, which are known effectors of PlexA signaling in Drosophila. I find evidence that Mical functions as a part of the Sema5c signaling axis but that it may have a second, noncanonical function unrelated to actin disassembly. I also show that Rap1 is involved in follicle cell migration and egg chamber elongation, but it is unclear what its molecular function is and whether it is part of the Sema5c signaling axis. In Chapter 4, I describe the results of a dominant enhancer/suppressor screen of Sema5c that was successful in identifying new genetic interactors. I provide initial characterization of these genes while also providing a guide for future work on mapping unidentified genes in genomic regions known to interact with Sema5c.While the specific molecular function of Sema5c will require further investigation, altogether, my work suggests that the role of the basement membrane in guiding egg chamber elongation is likely more complex than previously understood
Jointly representing long-range genetic similarity and spatially heterogeneous isolation-by-distance
Get PDFIsolation-by-distance patterns in genetic variation are a widespread feature of the geographic structure of genetic variation in many species, and many methods have been developed to illuminate such patterns in genetic data. However, long-range genetic similarities also exist, often as a result of rare or episodic long-range gene flow. Jointly characterizing patterns of isolation-by-distance and long-range genetic similarity in genetic data is an open data analysis challenge that, if resolved, could help produce more complete representations of the geographic structure of genetic data in any given species. Here, we present a computationally tractable method that identifies long-range genetic similarities in a background of spatially heterogeneous isolation-by-distance variation. The method uses a coalescent-based framework, and models long-range genetic similarity in terms of directional events with source fractions describing the fraction of ancestry at a location tracing back to a remote source. The method produces geographic maps annotated with inferred long-range edges, as well as maps of uncertainty in the geographic location of each source of long-range gene flow. We have implemented the method in a package called FEEMSmix (an extension to FEEMS), and validated its implementation using simulations representative of typical data applications. We also apply this method to two empirical data sets. In a data set of over 4,000 humans (Homo sapiens) across Afro-Eurasia, we recover many known signals of long-distance dispersal from recent centuries. Similarly, in a data set of over 100 gray wolves (Canis lupus) across North America, we identify several previously unknown long-range connections, some of which were attributable to recording errors in sampling locations. Therefore, beyond identifying genuine long-range dispersals, our approach also serves as a useful tool for quality control in spatial genetic studies.</p