Data Structure Research Articles

Neutron skins: A perspective from dispersive optical models

An overview of neutron skin predictions obtained using an empirical nonlocal dispersive optical model (DOM) is presented. The DOM links both scattering and bound-state experimental data through a subtracted dispersion relation which allows for fully consistent, data-informed predictions for nuclei where such data exist. Large skins were predicted for both 48Ca (Rskin48=0.25±0.023 fm in 2017) and 208Pb (Rskin208=0.25±0.05 fm in 2020). Whereas the DOM prediction in 208Pb is within 1σ of the subsequent PREX-2 measurement, the DOM prediction in 48Ca is over 2σ larger than the thin neutron skin resulting from CREX. From the moment it was revealed, the thin skin in 48Ca has puzzled the nuclear-physics community as no adequate theories simultaneously predict both a large skin in 208Pb and a small skin in 48Ca. The DOM is unique in its ability to treat both structure and reaction data on the same footing, providing a unique perspective on this Rskin puzzle. It appears vital that more neutron data be measured in both the scattering and bound-state domain for 48Ca to clarify the situation.

A longitudinal study of the impacts of area-level social cohesion on multimorbidity in Scotland

Abstract Background Multimorbidity is a growing global public health challenge, commonly referring to the co-occurrence of 2+ conditions but with varying measurements. Prior evidence has linked social cohesion with single health conditions; however, its impact on multimorbidity remains unclear. This study aims to examine the association between area-level social cohesion and multimorbidity and to explore whether this association will vary across multimorbidity measurements and levels of urban-rural classification. Methods Using a dataset that links the Scottish Longitudinal Study (5.3% of the Scottish population) with hospitalisation records (2001-2019) and an index that reflects area-level social cohesion (2012), focused on the cohorts of adults aged 40+, we used 3-level logistic regression to account for the clustered data structure of diagnosed chronic conditions over time within individuals within neighbourhoods. Results Areas with the highest social cohesion were observed to have the lowest risks of multimorbidity (p < 0.001), even after adjusting for area-level income deprivation. Varying by multimorbidity measurements, an inverted U-shaped pattern was observed between reduced area-level social cohesion and increased risks of having 2+ co-occurring chronic conditions; while a linear pattern was found for mental-physical multimorbidity. Moderated by levels of urban-rural classification, as compared to rural residents, urban residents were more sensitive to the impact of social cohesion on the co-occurrence of 2+ conditions; however, for mental-physical multimorbidity, no interaction effects between area-level social cohesion and urban-rural classification were found. Conclusions Although varying patterns were seen for different multimorbidity measurements, people living in areas with the highest area-level social cohesion were less likely to be multimorbid. This study offers new insights for developing effective place-based public health interventions. Key messages • This is the first longitudinal study that analysed the association between area-level social cohesion and multimorbidity using a nationally representative dataset based on secondary care data. • Area-level social cohesion affects multimorbidity, but its impact may vary by multimorbidity measurements and by urban-rural classification.

On a Simplified Approach to Achieve Parallel Performance and Portability Across CPU and GPU Architectures

This paper presents software advances to easily exploit computer architectures consisting of a multi-core CPU and CPU+GPU to accelerate diverse types of high-performance computing (HPC) applications using a single code implementation. The paper describes and demonstrates the performance of the open-source C++ matrix and array (MATAR) library that uniquely offers: (1) a straightforward syntax for programming productivity, (2) usable data structures for data-oriented programming (DOP) for performance, and (3) a simple interface to the open-source C++ Kokkos library for portability and memory management across CPUs and GPUs. The portability across architectures with a single code implementation is achieved by automatically switching between diverse fine-grained parallelism backends (e.g., CUDA, HIP, OpenMP, pthreads, etc.) at compile time. The MATAR library solves many longstanding challenges associated with easily writing software that can run in parallel on any computer architecture. This work benefits projects seeking to write new C++ codes while also addressing the challenges of quickly making existing Fortran codes performant and portable over modern computer architectures with minimal syntactical changes from Fortran to C++. We demonstrate the feasibility of readily writing new C++ codes and modernizing existing codes with MATAR to be performant, parallel, and portable across diverse computer architectures.

What Matters for an occurrenceID and What Is an occurrenceID That Matters?

In the Darwin Core data standard (Darwin Core Maintenance Group 2023), the concept of dwc:Occurrence (Wieczorek et al. 2012) in ecological data presents a data model construct not commonly found in data management practices used by ecological data collectors. We frequently encounter raw data without an Occurrence table. For example, the concept of Occurrence can be represented as a cell, where the rows represent sampling sites, the columns represent species, and the value of each cell indicates the count of the species at a specific site. A value in a cell in such a matrix can be interpreted as x number of individuals of species y occurred at sampling site z. While data providers tend to track data on tangible individual components (e.g., species, location, sample), generating "Occurrence records" typically requires pivoting and/or joining tables associated to these components. Maintaining a stable and persistent occurrenceID for an Occurrence record created through data transformation is not an easy task. This is especially true for long-term monitoring datasets, where the underlying tables used to generate Occurrence records are continuously updated. Additionally, most ecological data collectors are focused on the primary use of the data, not on the long term integration and accessibility of the data. The Occurrence concept is only required in data exchange format but not needed in ecological data management practices. The disconnect between the practical data management needs of data collectors and the abstractions required for data exchange raises challenges, particularly with an increasing expectation for globally unique and persistent occurrenceIDs. This presentation will explore the difficulties of creating and managing occurrenceIDs for data providers and managers, especially those who manage data using basic systems such as spreadsheets and simple relational databases. Maintaining stability and persistence of identifiers for inherently artificial constructs like Occurrences within the original, component-based data structure can pose significant challenges. We will explore why meaningful identifiers for occurrenceIDs are often preferred by data providers. We will unpack different use cases and delve into how and why occurrenceIDs were constructed for each use case. Through this discussion, we hope to spark a conversation that informs future data modeling efforts and addresses the inherent artificiality of Occurrences.

Per Segment Plane Sweep Line Segment Intersection on the GPU

Polygon overlay operations are used for various purposes such as GIS searches and queries, VLSI and basic geometric operations of intersection, union and difference. There have been recent research articles presenting algorithms using the GPU to perform line segment intersection for geometric operations. We present two parallel algorithms implemented on the GPU that focus on the active list portion of the traditional serial plane sweep algorithm. The first algorithm uses a single block of threads to simulate the active list data structure in hardware; this algorithm is slow due to GPU thread block size limitations and synchronization points, but demonstrates favorable time complexity. The second algorithm uses dynamic parallelism to remove synchronization and scales to utilize available GPU hardware (single GPU). We perform experiments on both synthetic and real world data sets. The presented results show improvement in execution time with respect to recent algorithms, and low memory usage compared to recent algorithms. We achieve speedups of up to 38.8 over the serial sweep line algorithm on real world data.

From Data to Decisions: Analyzing PaRIS for Health Policy Insights

Abstract To inform countries on how well they perform from the perspective of people living with chronic conditions in comparison with other countries, we propose an analytical framework for the data collected in the PaRIS survey, detailing how the data will be analysed in a multilevel approach for cross-country comparison. The data structure of the PaRIS survey represents three levels: countries/health systems, primary care practices and patients. Multilevel analysis is used because of its accuracy in estimating country-level outcomes, its flexibility in modelling relationships, and its opportunities in connecting to relevant policy questions. Country-level outcomes will be estimated to facilitate cross-country comparison and (future) within-country comparison over time. A first indication for areas for cross-country learning and policy making is the distribution of variation over patients, care providers and countries. Characteristics of patients that potentially explain variation in patient-reported outcomes and experiences can be linked to primary care practice and country/health system characteristics. The second indication for areas for cross-country learning and policy making is the variation in the slope of the association between patient and practice characteristics and patient-reported outcomes and experiences. This makes it possible to address policy-relevant questions relating, e.g., to whether some health systems or providers are better able to produce positive outcomes and experiences for (e.g.) older patients, and to the impact of chronic care management on patients with a specific chronic condition. To ensure fair cross-country comparisons, all outcomes will be estimated for an OECD (age and sex standardised) reference population, also including case-mix adjustment where necessary. Analyses will be repeated using a country-specific standard population.

Evidence-Based Programs in France: Insights and Strategies for Implementation Support

Abstract Issue In France, both regional and national health institutions increasingly support and fund the implementation of evidence-based programs (EBP), promoted at the national level by the French Public Health agency. Description of the problem practitioners who implement EBP face several operational questions: scaling issues, implementation in specific and varying contexts, integration and adjustment to daily regional practices. Results/effects In 2023, the French Society for Public Health initiated a collaborative effort to create a platform for discussion among practitioners, researchers, and decision-makers involved in EBP implementation. The first step was to assess EPB implementation experiences, using 3 types of data: an analysis of 6 documented EBP (capitalization works), an online questionnaire answered by 29 stakeholders, and four focus groups with funders, researchers, practitioners, and institutions. Data analysis was based on thematic content analysis and structural data analysis. Lessons Different ways to implement EBP were identified through a typology. The results highlight favorable conditions and levers for EPB implementation in France: • Facilitate access and adoption of EBP by concerned stakeholders, including through clear communication and knowledge sharing; • Strengthen and value support for implementation support (i.e., providing necessary resources and continuous monitoring); • A ‘favorable environment’ is also required (i.e. regional context acceptability and information on program themes or topics); • Breadthen perspectives and promote mutual learning through sharing and comparing implementation experiences across regions; • Value transferability and adaptation studies, which are key to enable EBP adaptations to account for context specificities while maintaining effectiveness. The platform’s next steps are to develop deployment tools and structure an advocacy strategy to mobilize stakeholders around the issue. Key messages • Achieving wide-scale implementation of EBP requires to tackle key implementation issues such as scaling, context adaptation, etc. and to give support to implementing stakeholders. • A collaborative assessment helped identify levers and barriers to EBP implementation in France. Involvement in EBP fosters an evidence-based mindset among practitioners.

Harnessing the Power of Machine Learning and Cube Technologies for Transformative Data Analytics

Machine learning, a cornerstone of artificial intelligence, is increasingly being integrated with traditional cube technologies to enhance data analysis and decision-making capabilities. While cube technologies excel at providing structured data exploration and OLAP capabilities, machine learning algorithms can uncover hidden patterns, predict future trends, and automate complex tasks. This technical article explores the synergistic relationship between machine learning and cube technologies, highlighting how their integration is transforming the landscape of data analytics and empowering organizations to gain deeper insights and make more informed decisions.

Optimizing Structured Data Processing Through Robotic Process Automation

Optimizing Structured Data Processing Through Robotic Process Automation

Optimization of Crystal Structures in Polylithionite Concentrate: A Molecular Dynamics Approach to Lithium Extraction Efficiency.

Molecular dynamics (MD) techniques offer significant potential for optimizing mineral extraction processes by simulating economically or physically restrictive conditions at the laboratory level. Lithium, a crucial metal in the electromobility era, exemplifies the need for ongoing re-evaluation of extraction techniques. This research aims to simulate the crystal structures of mineral species present in a polylithionite mineral concentrate [KLi2Al(Si4O10)(F,OH)2] using crystallographic data obtained from X-ray diffraction analysis. This study focuses on optimizing these structures, validating them through density comparisons, and determining the interaction parameter between the identified phases and lithium oxide (Li2O). The X-ray diffraction analysis revealed five predominant mineral phases: quartz (SiO2), calcite [Ca(CO3)], pyrite (FeS2), cassiterite (SiO2), and a compound Pb6O2(BO3)2SO4. Structural data, including lattice parameters, space groups, and atomic coordinates, were used to construct the crystal structures with Materials Studio 8.0, employing the Crystal Builder module. Optimization was performed using the Forcite module with the Smart optimization algorithm and the Universal force field. The interaction parameter (χ) indicated an affinity between lithium oxide and pyrite, as well as between calcite and quartz.

Structure and Magnetic Properties of Iron Oxide Nanoparticles Subjected to Mechanical Treatment

Iron oxide nanoparticles have been fabricated using the electric wire explosion (EWE) technique. The structure and magnetic properties of the nanoparticles have been analyzed before and after mechanical grinding in a ball mill for different time periods, focusing on potential bioapplications. The phase composition of the nanoparticles (70% Fe3O4, 30% Fe2O3) has remained unchanged despite the mechanical effects. The average nanoparticle size has not been affected either. The observation of the Verwey transition in the studied nanoparticles, along with the structural data, provides a better understanding of the physical properties of EWE ensembles of nanoparticles in different states. The analysis of the structure and magnetic properties reveals the development of a material with a high level of internal stress. This finding may be of interest for bioapplications due to its potential impact on the material performance.

Tree sequences as a general-purpose tool for population genetic inference.

As population genetics data increases in size new methods have been developed to store genetic information in efficient ways, such as tree sequences. These data structures are computationally and storage efficient, but are not interchangeable with existing data structures used for many population genetic inference methodologies such as the use of convolutional neural networks (CNNs) applied to population genetic alignments. To better utilize these new data structures we propose and implement a graph convolutional network (GCN) to directly learn from tree sequence topology and node data, allowing for the use of neural network applications without an intermediate step of converting tree sequences to population genetic alignment format. We then compare our approach to standard CNN approaches on a set of previously defined benchmarking tasks including recombination rate estimation, positive selection detection, introgression detection, and demographic model parameter inference. We show that tree sequences can be directly learned from using a GCN approach and can be used to perform well on these common population genetics inference tasks with accuracies roughly matching or even exceeding that of a CNN-based method. As tree sequences become more widely used in population genetics research we foresee developments and optimizations of this work to provide a foundation for population genetics inference moving forward.

Shrinkage estimation of gene interaction networks in single-cell RNA sequencing data

BackgroundGene interaction networks are graphs in which nodes represent genes and edges represent functional interactions between them. These interactions can be at multiple levels, for instance, gene regulation, protein-protein interaction, or metabolic pathways. To analyse gene interaction networks at a large scale, gene co-expression network analysis is often applied on high-throughput gene expression data such as RNA sequencing data. With the advance in sequencing technology, expression of genes can be measured in individual cells. Single-cell RNA sequencing (scRNAseq) provides insights of cellular development, differentiation and characteristics at the transcriptomic level. High sparsity and high-dimensional data structures pose challenges in scRNAseq data analysis.ResultsIn this study, a sparse inverse covariance matrix estimation framework for scRNAseq data is developed to capture direct functional interactions between genes. Comparative analyses highlight high performance and fast computation of Stein-type shrinkage in high-dimensional data using simulated scRNAseq data. Data transformation approaches also show improvement in performance of shrinkage methods in non-Gaussian distributed data. Zero-inflated modelling of scRNAseq data based on a negative binomial distribution enhances shrinkage performance in zero-inflated data without interference on non zero-inflated count data.ConclusionThe proposed framework broadens application of graphical model in scRNAseq analysis with flexibility in sparsity of count data resulting from dropout events, high performance, and fast computational time. Implementation of the framework is in a reproducible Snakemake workflow https://github.com/calathea24/ZINBGraphicalModel and R package ZINBStein https://github.com/calathea24/ZINBStein.

High-pressure synthesis of rhenium carbide Re3C under megabar compression

The rhenium carbide Re3C was predicted to be stable under high pressure and expected to have high hardness and low compressibility. In this study, we realise the synthesis of Re3C at megabar pressures of 105(3) and 140(5) GPa in laser-heated diamond anvil cells and characterise its structure using synchrotron single-crystal X-ray diffraction. The structure of Re3C has the monoclinic space group C2/m and is built of CRe7 capped octahedra. Our combined ab initio calculations and quantitative topological analysis support experimental structural data and further deepen the understanding of the chemical bonding in the newly synthesized compound.

Computing MEMs and Relatives on Repetitive Text Collections

We consider the problem of computing the Maximal Exact Matches (MEMs) of a given pattern \(P[1\mathinner{.. }m]\) on a large repetitive text collection \(T[1\mathinner{.. }n]\) over an alphabet of size \(\sigma\) , which is represented as a (hopefully much smaller) run-length context-free grammar of size \(g_{rl}\) . We show that the problem can be solved in time \(O(m^{2}\log^{\epsilon}n)\) , for any constant \(\epsilon\gt0\) , on a data structure of size \(O(g_{rl})\) . Further, on a locally consistent grammar of size \(O(\delta\log\frac{n\log\sigma}{\delta\log n})\) , the time decreases to \(O(m\log m(\log m+\log^{\epsilon}n))\) . The value \(\delta\) is a function of the substring complexity of \(T\) and \(\Omega(\delta\log\frac{n\log\sigma}{\delta\log n})\) is a tight lower bound on the compressibility of repetitive texts \(T\) , so our structure has optimal size in terms of \(n\) , \(\sigma\) , and \(\delta\) . We extend our results to several related problems, such as finding \(k\) -MEMs, MUMs, rare MEMs, and applications. Categories and Subject Descriptors: E.1 [Data structures] ; E.2 [Data storage representations] ; E.4 [Coding and information theory]: Data compaction and compression; F.2.2 [Analysis of algorithms and problem complexity] : Nonnumerical algorithms and problems— Pattern matching, Computations on discrete structures, Sorting and searching

Towards a method for understanding collaborative learning in makerspaces using dynamic sociotechnical networks

ABSTRACT This study’s objective was to discuss approaches to understanding learning in an educational makerspace by considering complex and dynamic interactions that take place there. Our investigation focusses on technical resources within these interactions – particularly how to gain a structural perspective on these dynamic interactions. We address two primary challenges: (1) conceptualizing makerspace activity as networks of human and nonhuman actors, and (2) capturing the structural dynamics of interaction in sociotechnical makerspace networks, as well as discussing how solving these challenges can help understand technology-mediated collaborative learning practices in the makerspace. We present potential approaches to yield a more granular and nuanced view based on video data on the interplay between collaborators and technical tools and materials, such as how to transform video data into structural data, as well as capture and represent dynamic networks. We discuss how these approaches can inform researchers about collaborative learning in makerspaces by discussing students’ sociotechnical ego-networks, changing roles and the use of sociotechnical structural data as background information in qualitative analysis. Furthermore, the paper examines our study’s broader implications for understanding makerspace networks, suggesting node-grouping strategies and sociotechnograms to simplify and name the display of sociotechnical interdependencies.

Phantom Matter: A Challenging Solution to the Cosmological Tensions

The idea of composite dark energy (DE) is quite natural since on general grounds we expect that the vacuum energy density (associated with the cosmological term Λ) may appear in combination with other effective forms of DE, denoted X. Here we deal with model wXCDM, a simplified version of the old ΛXCDM model, and exploit the possibility that X behaves as “phantom matter” (PM), which appears in stringy versions of the running vacuum model (RVM). Unlike phantom DE, the PM fluid satisfies the strong energy condition like usual matter, hence bringing to bear positive pressure at the expense of negative energy. Bubbles of PM may appear in the manner of a transitory “phantom vacuum” tunneled into the late Universe before it heads toward a new de Sitter era, thereby offering a crop field for the growing of structures earlier than expected. Using Type Ia supernovae, cosmic chronometers, transversal baryon acoustic oscillations (BAO 2D), large-scale structure data, and the full cosmic microwave background likelihood from Planck 2018, we find that the H 0 and growth tensions virtually disappear, provided that BAO 2D are the only source of BAO data used in the fit. In contrast, our preliminary analysis using exclusively anisotropic BAO (BAO 3D) indicates that the ability to ease the H 0 tension is significantly reduced as compared to the scenario with BAO 2D, despite the fact that the overall fit to the cosmological data is still better than in the ΛCDM. Finally, our approach with BAO 2D favors quintessence-like behavior of the DE below z ≃ 1.5 at ≳3σ confidence level, which is compatible with the recent DESI measurements.

Computational Exploration of the Impact of Low‐Spin and High‐Spin Ground State on the Chelating Ability of Dimethylglyoxime Ligand on Dihalo Transition Metal: A QTAIM, EDA, and CDA Analysis

ABSTRACTWe have explored the chelation of dimethylglyoxime ligand to divalent (ndx: x = 6, 7, 8) transition metal (TM) cations in two media (gas phase and water) at the B3LYP//LANL2DZ/6–311+G (d,p) and B3LYP/def2‐TZVP level at lower multiplicity and higher multiplicity states. Majority of the 18 optimized halide (chloride and bromide) complexes prefer square planar configuration. The correlations discerned between the experimental structural data and their estimated counterparts demonstrate a good credibility for complexes at lower multiplicity state. The basis set superposition errors (BSSEs) estimated is very small which reflects the fact that the choice of different basis sets (B3LYP//LANL2DZ/6–311+G (d,p)) introduces a slight bias in the calculation of energies. The ADMP (atom‐centered density matrix propagation) simulations in water on chloride complexes indicate the irreversible nature of these M—N dissociation in trajectory simulation process. This fact explains our exclusive focus on the examination of the [glyoxime ligand]…[MX2] interactions. In addition, the solvation of (3d and 4d) transition metal chloride complexes causes a sensitive augmentation of the metal ion affinity (MIA) with an average of 0.29 and 0.24 kcal/mol. In both multiplicity states, the topological parameters have illustrated that the M—N and M—X bonds are typical metal–ligand in both media. The average ΔEorb/ΔEsteric ratio equal to 0.45 and 0.11 in gas phase and water, respectively, reveals the predominance of the contributions from non‐covalent bonding interactions (NCI) compared to those of covalent bonding. But, the maximal value equal to 6.760 is obtained for bromide rhodium complex in water. NBO analysis in both media highlights the fact that a more pronounced ionic character is observed for the majority of the chloride complexes at both spin multiplicity states because of their higher retained charges on the metal atom. For [dimethylglyoxime]…[MX2] interaction (X = Cl and Br), the charge decomposition analysis demonstrates that the lowest value of the d/b ratio is found for the chloride platinum complex at lower multiplicity state in water. This is a proof of its strong relativistic effects.

Powerful graph neural network for node classification of the IoT network

Internet of Things (IoT) devices are increasingly used in various applications in our daily lives. The network structure for IoT is heterogeneous and can create a complex architecture depending on the application and geographical structure. To efficiently process the information within this diverse and complex relationship, a robust data structure is needed for network operations. Graph neural network (GNN) technology is emerging as a capable tool for predicting complex data structures, such as graphs. Graphs can be employed to mimic the structure of IoT network and process information from IoT nodes using GNN techniques. In this paper, our goal is to explore the effectiveness of GNN in performing the node classification task for a given IoT network. We have generated three different IoT networks with varying network sizes, number of nodes, and feature sizes. We then test 12 different GNN algorithms to evaluate their performance in IoT node classification. Each method is examined in detail to observe its training behavior, testing behavior, and resilience against noise. In addition, time complexity and generalization ability of each model have also been studied. The experimental results show that some methods exhibit high resilience against noisy data for IoT node classification accuracy.

Overcoming artificial structures in resolution-enhanced Hi-C data by signal decomposition and multi-scale attention.

Computational enhancement is an important strategy for inferring high-resolution features from genome-wide chromosome conformation capture (Hi-C) data, which typically have limited resolution. Deep learning has been highly successful in this task but we show that it creates prevalent artificial structures in the enhanced data due to the need to divide the large contact matrix into small patches. In addition, previous deep learning methods largely focus on local patterns, which cannot fully capture the complexity of Hi-C data. Here we propose Smooth, High-resolution, and Accurate Reconstruction of Patterns (SHARP) for enhancing Hi-C data. It uses the novel approach of decomposing the data into three types of signals, due to one-dimensional proximity, contiguous domains, and other fine structures, and applies deep learning only to the third type of signals, such that enhancement of the first two is unaffected by the patches. For the deep learning part, SHARP uses both local and global attention mechanisms to capture multi-scale contextual information. We compare SHARP with state-of-the-art methods extensively, including application to data from new samples and another species, and show that SHARP has superior performance in terms of resolution enhancement accuracy, avoiding creation of artificial structures, identifying significant interactions, and enrichment in chromatin states.