High-performance Computing Capabilities Research Articles

Benchmarking quantum computer simulation software packages: State vector simulators

Rapid advances in quantum computing technology lead to an increasing need for software simulators that enable both algorithm design and the validation of results obtained from quantum hardware. This includes calculations that aim at probing regimes of quantum advantage, where a quantum computer outperforms a classical computer in the same task. High performance computing (HPC) platforms play a crucial role as today’s quantum devices already reach beyond the limits of what powerful workstations can model, but a systematic evaluation of the individual performance of the many offered simulation packages is lacking so far. In this Technical Review, we benchmark several software packages capable of simulating quantum dynamics with a special focus on HPC capabilities. We develop a containerized toolchain for benchmarking a large set of simulation packages on a local HPC cluster using different parallelisation capabilities, and compare the performance and system size-scaling for three paradigmatic quantum computing tasks. Our results can help finding the right package for a given simulation task and lay the foundation for a systematic community effort to benchmark and validate upcoming versions of existing and also newly developed simulation packages.

JAX-Fluids 2.0: Towards HPC for differentiable CFD of compressible two-phase flows

In our effort to facilitate machine learning-assisted computational fluid dynamics (CFD), we introduce the second iteration of JAX-Fluids. JAX-Fluids is a Python-based fully-differentiable CFD solver designed for compressible single- and two-phase flows. In this work, the first version is extended to incorporate high-performance computing (HPC) capabilities. We introduce a parallelization strategy utilizing JAX primitive operations that scales efficiently on GPU (up to 512 NVIDIA A100 graphics cards) and TPU (up to 1024 TPU v3 cores) HPC systems. We further demonstrate stable parallel computation of automatic differentiation gradients across extended integration trajectories. The new code version offers enhanced two-phase flow modeling capabilities. In particular, a five-equation diffuse-interface model is incorporated which complements the level-set sharp-interface model. Additional algorithmic improvements include positivity-preserving limiters for increased robustness, support for stretched Cartesian meshes, refactored I/O handling, comprehensive post-processing routines, and an updated list of state-of-the-art high-order numerical discretization schemes. We verify newly added numerical models by showcasing simulation results for single- and two-phase flows, including turbulent boundary layer and channel flows, air-helium shock bubble interactions, and air-water shock drop interactions.

Are 2D shallow-water solvers fast enough for early flood warning? A comparative assessment on the 2021 Ahr valley flood event

Abstract. Flash floods pose a distinct challenge compared to traditional fluvial flooding, with infrastructure-based solutions proving less effective. Effective responses hinge on advanced early warning systems providing actionable information, emphasising the necessity for computational flood forecasting models. However, hydrodynamic models, renowned for accuracy and completeness, face limitations due to computational intensity. This study explores two 2D flood forecasting models, RIM2D and SERGHEI, both with GPU implementations which allow us to maximise the forecast lead time. While RIM2D is less computationally intensive, suitable for operational use, SERGHEI, with higher computational costs, targets large-scale high-performance computing (HPC) systems. The assessment of applicability and trade-offs is carried out on the 2021 Eifel flood event, particularly in the lower Ahr valley. A set of simulations were performed at various resolutions from 1 to 10 m, which reveal similar accuracy among both models at coarser resolutions, yet discrepancies arise at finer resolutions due to the distinct formulations. Both models exhibit a rapid computational cost escalation, but at resolutions equal to or coarser than 5 m, forecasts are remarkably faster than the real-time ideal for operational use, paving the way for their use in early warning systems. However, higher resolutions necessitate multi-GPU and HPC capabilities, underlining the importance of embracing such technology in addressing broader flood domains.

Shaoxia: a web-based interactive analysis platform for single cell RNA sequencing data

BackgroundIn recent years, Single-cell RNA sequencing (scRNA-seq) is increasingly accessible to researchers of many fields. However, interpreting its data demands proficiency in multiple programming languages and bioinformatic skills, which limited researchers, without such expertise, exploring information from scRNA-seq data. Therefore, there is a tremendous need to develop easy-to-use software, covering all the aspects of scRNA-seq data analysis.ResultsWe proposed a clear analysis framework for scRNA-seq data, which emphasized the fundamental and crucial roles of cell identity annotation, abstracting the analysis process into three stages: upstream analysis, cell annotation and downstream analysis. The framework can equip researchers with a comprehensive understanding of the analysis procedure and facilitate effective data interpretation. Leveraging the developed framework, we engineered Shaoxia, an analysis platform designed to democratize scRNA-seq analysis by accelerating processing through high-performance computing capabilities and offering a user-friendly interface accessible even to wet-lab researchers without programming expertise.ConclusionShaoxia stands as a powerful and user-friendly open-source software for automated scRNA-seq analysis, offering comprehensive functionality for streamlined functional genomics studies. Shaoxia is freely accessible at http://www.shaoxia.cloud, and its source code is publicly available at https://github.com/WiedenWei/shaoxia.

Coupled neutronics and thermal-hydraulics using TRUST-NK for high performance computing in molten salt reactor simulation

Molten Salt Reactors display a strong coupling between neutronics and thermal-hydraulics for which various simulation tools have been developed. In this paper we describe TRUST-NK, a neutronics code based on TRUST, CEA’s High Performance Computing (HPC) platform. TRUST-NK provides a multi-group diffusion solver with transport of delayed neutron precursors, for both steady-state and transient simulations. Being based on TRUST allows TRUST-NK to take advantage of the platform’s HPC capabilities and to easily couple to CFD solvers provided by other derivatives of TRUST. After a brief description of the code, we present an example of application to a simplified MSR model.

A general framework for portfolio construction based on generative models of asset returns

In this paper, we present an integrated approach to portfolio construction and optimization, leveraging high-performance computing capabilities. We first explore diverse pairings of generative model forecasts and objective functions used for portfolio optimization, which are evaluated using performance-attribution models based on least absolute shrinkage and selection operator (LASSO). We illustrate our approach using extensive simulations of crypto-currency portfolios, and we show that the portfolios constructed using the vine-copula generative model and the Sharpe-ratio objective function consistently outperform. To accommodate a wide array of investment strategies, we further investigate portfolio blending and propose a general framework for evaluating and combining investment strategies. We employ an extension of the multi-armed bandit framework and use value models and policy models to construct eclectic blended portfolios based on past performance. We consider similarity and optimality measures for value models and employ probability-matching (“blending”) and a greedy algorithm (“switching”) for policy models. The eclectic portfolios are also evaluated using LASSO models. We show that the value model utilizing cosine similarity and logit optimality consistently delivers robust superior performances. The extent of outperformance by eclectic portfolios over their benchmarks significantly surpasses that achieved by individual generative model-based portfolios over their respective benchmarks.

Cloud benchmarking and performance analysis of an HPC application in Amazon EC2

Cloud computing platforms have been continuously evolving. Features such as the Elastic Fabric Adapter (EFA) in the Amazon Web Services (AWS) platform have brought yet another revolution in the High Performance Computing (HPC) world, further accelerating the convergence of HPC and cloud computing. Other public clouds also support similar features further fueling this change. In this paper, we show how and why the performance of a large-scale computational fluid dynamics (CFD) HPC application on AWS competes very closely with the one on Beskow—a Cray XC40 supercomputer at the PDC Center for High-Performance Computing - in terms of cost-efficiency with strong scaling up to 2304 processes. We perform an extensive set of micro and macro benchmarks in both environments and conduct a comparative analysis. Until as recently as 2020 these benchmarks have notoriously yielded unsatisfactory results for the cloud platforms compared with on-premise infrastructures. Our aim is to access the HPC capabilities of the cloud, and in general to demonstrate how researchers can scale and evaluate the performance of their application in the cloud.

Accelerated dynamic data reduction using spatial and temporal properties

Due to improvements in high-performance computing (HPC) capabilities, many of today’s applications produce petabytes worth of data, causing bottlenecks within the system. Importance-based sampling methods, including our spatio-temporal hybrid data sampling method, are capable of resolving these bottlenecks. While our hybrid method has been shown to outperform existing methods, its effectiveness relies heavily on user parameters, such as histogram bins, error threshold, or number of regions. Moreover, the throughput it demonstrates must be higher to avoid becoming a bottleneck itself. In this article, we resolve both of these issues. First, we assess the effects of several user input parameters and detail techniques to help determine optimal parameters. Next, we detail and implement accelerated versions of our method using OpenMP and CUDA. Upon analyzing our implementations, we find 9.8× to 31.5× throughput improvements. Next, we demonstrate how our method can accept different base sampling algorithms and the effects these different algorithms have. Finally, we compare our sampling methods to the lossy compressor cuSZ in terms of data preservation and data movement.

Fully nonlinear phase-resolved wave modelling in the Norwegian fjords for floating bridges along the E39 coastal highway

Coastal wave conditions are highly influenced by bathymetry variations and coastlines. The Norwegian fjords present large gradients of bathymetry changes and irregular coastline geometries that make the understanding of the near-coast wave conditions a challenging task. The proposed floating bridges for permanent fjord-crossings along the Norwegian E39 coastal highway require reliable and efficient numerical simulations of a large domain at the fjords. In this manuscript, a phase-resolving fully non-linear potential flow (FNPF) model is used to investigate wave propagations with the presence of complex bathymetry and coastlines at two fjords along the path of the highway. The FNPF model is equipped with a robust coastline algorithm, flexible wave breaking algorithms, a free-surface and bathymetry-following sigma -grid and parallel high-performance computation capability. Both long-crested and short-crested offshore swell wave conditions are simulated at the two fjords, Sulafjord and Bjørnafjord. Sulafjord is more exposed to the offshore swell while Bjørnafjord is more sheltered. The phase-averaging spectral wave models are also applied for both fjords and compared with the phase-resolving FNPF model. The significance of applying phase-resolving model for strongly nonlinear wave transformations in the fjords is demonstrated. The wave transformations and the wave field inhomogeneity and incoherence are captured using the phase-resolving time domain simulation results, which reveal significant engineering considerations for floating bridge design. The presented numerical approach is seen to be able to simulate the complex transformations of irregular waves and identify key wave field characteristics with the presence of irregular bathymetry and coastlines such as the Norwegian fjords.

The role of spatial cyberinfrastructure in geoinformation systems

The potential future of a spatial cyberinfrastructure is considered, combining advanced computer, information and communication technologies to expand the possibilities of computational and information scientific practice and improve the synthesis and analysis of scientific data through collaboration and sharing. Currently, spatial cyberinfrastructure represents a paradigm shift in scientific research that has facilitated easy access to computational utilities and facilitated collaboration across distances and disciplines, thereby enabling scientific breakthroughs to be achieved faster and more efficiently. Spatial cyberinfrastructure seeks to address the long-standing complex problems of processing and analyzing massive and heterogeneous spatial datasets, as well as the need and benefits of flexible and secure spatial data exchange. The focus is on spatial analysis using distributed and high performance computing, sensor networks and other advanced information technology capabilities to transform massive spatial datasets into information and knowledge. To solve large-scale and multi-scale problems, the spatial cyberinfrastructure, despite the complexity of the geographic space, allows transforming spatial analysis into powerful and affordable computing utilities to provide broad scientific breakthroughs when creating the well-known Digital Earth project. Critical analysis and recommendations are provided on issues of conflicting data that researchers do not want to exclude from various data models and geographic scales. This case study shows the value of spatial cyberinfrastructure in addressing the challenges that arise from spatial as well as multilingual, biographical and temporal ambiguities in various databases, solutions that, again, would not be possible without spatial cyberinfrastructure.

CFD code comparison, verification and validation for decay tests of a FOWT semi-submersible floater

With the advancement of high-performance computation capabilities in recent years, high-fidelity modelling tools such as computational fluid dynamics are becoming increasingly popular in the offshore renewable sector. To justify the credibility of the numerical simulations, thorough verification and validation is essential. In this work, preparatory heave decay tests for a freely floating single cylinder are modelled. Subsequently, the surge and sway decays of a linearly moored floating offshore wind turbine model of the OC4 (Offshore Code Comparison Collaboration Continuation) phase II semi-submersible platform are simulated. Two different viscous-flow CFD codes are used: OpenFOAM (open-source), and ReFRESCO (community-based open-usage). Their results are compared against each other and with water tank experiments. For the single-cylinder decay simulations, it is found that the natural period is accurately modelled compared to the experimental results. Regarding the damping, both CFD codes are overly dissipative. Differences and their potential explanations become apparent in the analysis of the flow field data. Meanwhile, large numerical uncertainties especially in later oscillations make a distinct conclusion difficult. For the OC4 semi-submersible decay simulations, a better agreement in damping can be achieved, however discrepancies in results are observed when restricting the degrees of freedom of the platform. Flow field data again reveals differences between the CFD codes. Meanwhile, through the effort to use similar numerical settings and quantify the numerical uncertainties of the CFD simulations, this work represents a stepping stone towards fairer and more accurate comparison between CFD and experimental results.

Permafrost modelling with OpenFOAM®: New advancements of the permaFoam solver

Permafrost, i.e., soil that is year-round frozen in depth, is covering a quarter of the continents of the northern hemisphere. It currently experiences fast changes due to climate change at global scale and technogenic perturbations at local scale, and the assessment and anticipation of these changes are of primary importance for many environmental and engineering applications in cold regions. To these ends, permafrost modelling is required, while the strong couplings and non-linearities involved in the physics at stake make it highly challenging, especially from a computational point of view. In this work we present a new solver for permafrost hydrology developed in the framework of OpenFOAM®, allowing to benefit from its advanced high-performance computing capabilities. The solver is tested for realistic, field-based cases, and its parallel performances are characterized up to ∼16000 cores on IRENE supercomputer (TGCC, CEA). Program summaryProgram Title: permaFoamCPC Library link to program files:https://doi.org/10.17632/swp88cvpwb.1Developer's repository link:https://develop.openfoam.com/Community/hydrology/Licensing provisions: GPLv3Programming language: C++Nature of problem: This software solves the coupled equations that govern water flow and heat transfer in variably saturated and variably frozen porous media, for transient problems in Three-dimensional, heterogeneous domains. The equation for water flow is Richards equation, which is a very popular model for water transfer in variably saturated porous media (e.g.: soils), and the equation for heat transfer is a Fourier equation including advection and the freeze/thaw of the pore water. The solver is designed to take advantage of the massively parallel computing performance of OpenFOAM®. The goal is to be able to model natural hydrosystems of cold regions on large temporal and spatial scales.Solution method: For each equation a mixed implicit (FVM for Finite Volume Method in the object oriented OpenFOAM framework) and explicit (FVC for Finite Volume Calculus in the object oriented OpenFOAM® framework) discretization with backward time scheme is embedded in an iterative linearization procedure (Picard algorithm). The coupling between the two equations is performed through an operator splitting approach. The implementation has been carried out with a concern for robustness and parallel efficiency.Additional comments including restrictions and unusual features: This version of permaFoam has been tested with OpenFOAM_v1912, v2106, v2112 and v2206, thus everything might not work with other (especially older) versions of OpenFOAM®. When using permaFoam, one should be careful to use fine enough spatial and temporal discretisations where and when steep fronts (freeze/thaw fronts, imbibition/drainage fronts) occur, otherwise numerical stability problems might arise.

Edge Computing for Cyber-physical Systems: A Systematic Mapping Study Emphasizing Trustworthiness

Edge computing is projected to have profound implications in the coming decades, proposed to provide solutions for applications such as augmented reality, predictive functionalities, and collaborative Cyber-Physical Systems (CPS). For such applications, edge computing addresses the new computational needs, as well as privacy, availability, and real-time constraints, by providing local high-performance computing capabilities to deal with the limitations and constraints of cloud and embedded systems. Edge computing is today driven by strong market forces stemming from IT/cloud, telecom, and networking—with corresponding multiple interpretations of “edge computing” (e.g., device edge, network edge, distributed cloud). Considering the strong drivers for edge computing and the relative novelty of the field, it becomes important to understand the specific requirements and characteristics of edge-based CPS, and to ensure that research is guided adequately, e.g., avoiding specific gaps. Our interests lie in the applications of edge computing as part of CPS, where several properties (or attributes) of trustworthiness, including safety, security, and predictability/availability, are of particular concern, each facing challenges for the introduction of edge-based CPS. We present the results of a systematic mapping study, a kind of systematic literature survey, investigating the use of edge computing for CPS with a special emphasis on trustworthiness. The main contributions of this study are a detailed description of the current research efforts in edge-based CPS and the identification and discussion of trends and research gaps. The results show that the main body of research in edge-based CPS only to a very limited extent consider key attributes of system trustworthiness, despite many efforts referring to critical CPS and applications like intelligent transportation. More research and industrial efforts will be needed on aspects of trustworthiness of future edge-based CPS including their experimental evaluation. Such research needs to consider the multiple interrelated attributes of trustworthiness including safety, security, and predictability, and new methodologies and architectures to address them. It is further important to provide bridges and collaboration between edge computing and CPS disciplines.

Interoperable Data Analytics Reference Architectures Empowering Digital-Twin-Aided Manufacturing

The use of mature, reliable, and validated solutions can save significant time and cost when introducing new technologies to companies. Reference Architectures represent such best-practice techniques and have the potential to increase the speed and reliability of the development process in many application domains. One area where Reference Architectures are increasingly utilized is cloud-based systems. Exploiting the high-performance computing capability offered by clouds, while keeping sovereignty and governance of proprietary information assets can be challenging. This paper explores how Reference Architectures can be applied to overcome this challenge when developing cloud-based applications. The presented approach was developed within the DIGITbrain European project, which aims at supporting small and medium-sized enterprises (SMEs) and mid-caps in realizing smart business models called Manufacturing as a Service, via the efficient utilization of Digital Twins. In this paper, an overview of Reference Architecture concepts, as well as their classification, specialization, and particular application possibilities are presented. Various data management and potentially spatially detached data processing configurations are discussed, with special attention to machine learning techniques, which are of high interest within various sectors, including manufacturing. A framework that enables the deployment and orchestration of such overall data analytics Reference Architectures in clouds resources is also presented, followed by a demonstrative application example where the applicability of the introduced techniques and solutions are showcased in practice.

Scaling Scientific Cellular Automata Microstructure Evolution Model of Static Recrystallization toward Practical Industrial Calculations.

An attempt to bridge the gap between capabilities offered by advanced full-field microstructure evolution models based on the cellular automata method and their practical applications to daily industrial technology design was the goal of the work. High-performance parallelization techniques applied to the cellular automata static recrystallization (CA-SRX) model were selected as a case study. Basic assumptions of the CA-SRX model and developed modifications allowing high-performance computing are presented within the paper. Particular attention is placed on the development of the parallel computation scheme allowing numerical simulations even for a large volume of material. The development of new approaches to handle communication within the distributed environment is also addressed in the paper as a means to obtain higher computational efficiency. Evaluation of model limits was based on the scalability analysis. The investigation was carried out for the 3D and 2D case studies. Therefore, the complex static recrystallization cellular automata simulation taking into account the influence of recovery, nucleation based on accumulated energy, and the progress of recrystallization as a function of stored energy and grain boundary mobility with high-performance computing capabilities is now possible. The research highlighted that parallelization is more effective with an increasing number of cellular automata cells processed during the entire simulation. It was also proven that the developed parallelization scheme and communication mechanism provides a possibility of obtaining scaled speedup over 700 times for 2D and over 800 times for 3D computational domains, which is crucial for future applications in industrial practice. Therefore, the presented approach’s main advantage is based on the possibility of running the calculation based on input data obtained directly from high-resolution 3D imaging of the microstructure. With that, the full immersion of the experimental results into the numerical model is possible. The second novelty aspect of this work is related to the identification of the quality of model predictions as a function of model size reductions.

A Review on Detection of Covid Symptoms

Telemedicine may hold the key to limiting the emerging coronavirus disease's global outbreak (COVID-19). The COVID-19 virus directly affects the lungs, causing pneumonia-like symptoms and shortness of breath, both of which can be fatal. Despite the fact that self-quarantine and social isolation are crucial during a pandemic, the technique for identifying COVID-19 contraction via nose swabs, saliva test kits, and blood testing is routinely available at healthcare facilities. As a result, developing customized self-testing kits for the COVID-19 virus and related viruses is a top goal. Many e-health initiatives are now possible thanks to smartphones with embedded software, hardware, high-performance computation, and networking capabilities. COVID-19 contracted users' breathing sounds may reveal certain acoustic signal patterns that are worth researching, according to a thorough assessment of breathing sounds and their implications in recognizing respiratory issues. Obtaining respiratory data solely from breathing noises input into a smartphone's microphone appears to be an intriguing solution in this area. Advanced signal processing and analysis, in conjunction with modern deep/ machine learning and pattern recognition algorithms, can be used to analyze the obtained breathing sounds to separate the breathing phases, estimate lung volume, oxygenation, and further classify the breathing data input into healthy or unhealthy situations. For the ongoing global COVID-19 pandemic, the concepts mentioned have the potential to be employed as self-test breathing monitoring apps, with users being able to check their breathing sound pattern on a frequent basis.

SNF-NN: computational method to predict drug-disease interactions using similarity network fusion and neural networks

BackgroundDrug repositioning is an emerging approach in pharmaceutical research for identifying novel therapeutic potentials for approved drugs and discover therapies for untreated diseases. Due to its time and cost efficiency, drug repositioning plays an instrumental role in optimizing the drug development process compared to the traditional de novo drug discovery process. Advances in the genomics, together with the enormous growth of large-scale publicly available data and the availability of high-performance computing capabilities, have further motivated the development of computational drug repositioning approaches. More recently, the rise of machine learning techniques, together with the availability of powerful computers, has made the area of computational drug repositioning an area of intense activities.ResultsIn this study, a novel framework SNF-NN based on deep learning is presented, where novel drug-disease interactions are predicted using drug-related similarity information, disease-related similarity information, and known drug-disease interactions. Heterogeneous similarity information related to drugs and disease is fed to the proposed framework in order to predict novel drug-disease interactions. SNF-NN uses similarity selection, similarity network fusion, and a highly tuned novel neural network model to predict new drug-disease interactions. The robustness of SNF-NN is evaluated by comparing its performance with nine baseline machine learning methods. The proposed framework outperforms all baseline methods (AUC-ROC = 0.867, and AUC-PR=0.876) using stratified 10-fold cross-validation. To further demonstrate the reliability and robustness of SNF-NN, two datasets are used to fairly validate the proposed framework’s performance against seven recent state-of-the-art methods for drug-disease interaction prediction. SNF-NN achieves remarkable performance in stratified 10-fold cross-validation with AUC-ROC ranging from 0.879 to 0.931 and AUC-PR from 0.856 to 0.903. Moreover, the efficiency of SNF-NN is verified by validating predicted unknown drug-disease interactions against clinical trials and published studies.ConclusionIn conclusion, computational drug repositioning research can significantly benefit from integrating similarity measures in heterogeneous networks and deep learning models for predicting novel drug-disease interactions. The data and implementation of SNF-NN are available at http://pages.cpsc.ucalgary.ca/ tnjarada/snf-nn.php.

A Hybrid Learning Approach to Prognostics and Health Management Applied to Military Ground Vehicles Using Time-Series and Maintenance Event Data

Attempts to leverage operational time-series data in Condition Based Maintenance (CBM) approaches to optimize the life cycle management and Reliability, Availability, and Maintainability (RAM) of military vehicles have encountered several obstacles over decades of data collection. These obstacles have beset similar approaches on civilian ground vehicles, as well as on aircraft and other complex systems. Analysis of operational data is critical because it represents a continuous recording of the state of the system. Applying rudimentary data analytics to operational data can provide insights like fuel usage patterns or observed reliability of one vehicle or even a fleet. Monitoring trends and analyzing patterns in this data over time, however, can provide insight into the health of a vehicle, a complex system, or a fleet, predicting mean time to failure or compiling logistic or life cycle needs. Such High-Performance Data Analytics (HPDA) on operational time-series datasets has been historically difficult due to the large amount of data gathered from vehicle sensors, the lack of association between clusters observed in the data and failures or unscheduled maintenance events, and the deficiency of unsupervised learning techniques for time-series data. We present an HPDA environment and a method of discovering patterns in vehicle operational data that determines models for predicting the likelihood of imminent failure, referred to as Parameter-Based Indicators (PBIs). Our method is a data-driven approach that uses both time-series and relational maintenance data. This hybrid approach combines both supervised and unsupervised machine learning and data analytic techniques to correlate labeled, relational maintenance event data with unlabeled operational time-series data utilizing the DoD High Performance Computing (HPC) capabilities at the U.S. Army Engineer Research and Development Center. In leveraging both time-series and relational data, we demonstrate a means of fast, purely data-driven model creation that is more broadly applicable and requires less a priori information than physics informed, data-driven models. By blending these approaches, this system will be able to relate some lifecycle management goals through the workflow to generate specific PBIs that will predict failures or highlight appropriate areas of concern in individual or collective vehicle histories.

SNF–CVAE: Computational method to predict drug–disease interactions using similarity network fusion and collective variational autoencoder

Drug repositioning is an emerging approach to identify novel therapeutic potentials for approved drugs and discover therapies for previously untreatable diseases. Drug repositioning has also attracted considerable attention in the pharmaceutical industry due to its time and cost efficiency in the drug development process compared to the traditional de novo drug discovery process. Recent advances in genomics, the tremendous growth of large-scale publicly available data, the availability of high-performance computing capabilities, along with the rise of machine learning, have further motivated the development of computational drug repositioning approaches. Investigating the relationship between different biomedical entities (e.g., drugs, diseases, genes) is one vital part of most recent studies in the drug repositioning field. Drug–disease interaction (R-DI) prediction is another main issue in drug repositioning research. Combining these relationships and interactions when introducing computational methods to identify novel drug–disease interactions with high accuracy is very challenging. In this study, we propose a robust approach, SNF–CVAE, for predicting novel drug–disease interactions using drug-related similarity information and known drug–disease interactions. SNF–CVAE integrates similarity measures, similarity selection, similarity network fusion (SNF), and collective variational autoencoder (CVAE) to conduct a non-linear analysis and improve the drug–disease interaction prediction accuracy. We evaluated the robustness of SNF–CVAE using different information models, drug similarity calculation measures, and drug similarity information. Moreover, we compared SNF–CVAE performance with four state-of-the-art machine learning models. SNF–CVAE achieved outstanding performance in stratified 5-fold cross-validation (Prec = 0.902, Rec = 0.883, F1 = 0.893, AUC-ROC = 0.958, and AUC-PR=0.970). Furthermore, we showed the efficiency of SNF–CVAE in predicting novel drug–disease interactions by validating the top-ranked interactions against pharmaceutical indications and clinical trial studies, which resulted in substantial pieces of evidence for almost all of RDIs predicted by our proposed model. To further demonstrate the reliability and robustness of SNF–CVAE, we conducted two case studies on the top predicted drug candidates for potentially treating Alzheimer’s disease and Juvenile rheumatoid arthritis, which were successfully validated against clinical trials and published studies. In conclusion, we strongly believe that computational drug repurposing research could significantly benefit from integrating similarity measures and deep learning models to predict novel drug–disease interactions in heterogeneous networks.

Smartphone-Based Self-Testing of COVID-19 Using Breathing Sounds.

Telemedicine could be a key to control the world-wide disruptive and spreading novel coronavirus disease (COVID-19) pandemic. The COVID-19 virus directly targets the lungs, leading to pneumonia-like symptoms and shortness of breath with life-threatening consequences. Despite the fact that self-quarantine and social distancing are indispensable during the pandemic, the procedure for testing COVID-19 contraction is conventionally available through nasal swabs, saliva test kits, and blood work at healthcare settings. Therefore, devising personalized self-testing kits for COVID-19 virus and other similar viruses is heavily admired. Many e-health initiatives have been made possible by the advent of smartphones with embedded software, hardware, high-performance computing, and connectivity capabilities. A careful review of breathing sounds and their implications in identifying breathing complications suggests that the breathing sounds of COVID-19 contracted users may reveal certain acoustic signal patterns, which is worth investigating. To this end, acquiring respiratory data solely from breathing sounds fed to the smartphone's microphone strikes as a very appealing resolution. The acquired breathing sounds can be analyzed using advanced signal processing and analysis in tandem with new deep/machine learning and pattern recognition techniques to separate the breathing phases, estimate the lung volume, oxygenation, and to further classify the breathing data input into healthy or unhealthy cases. The ideas presented have the potential to be deployed as self-test breathing monitoring apps for the ongoing global COVID-19 pandemic, where users can check their breathing sound pattern frequently through the app.