Use Of High Performance Computing Research Articles

ESSENCE-Dock: A Consensus-Based Approach to Enhance Virtual Screening Enrichment in Drug Discovery.

Drug development is a complex, costly, and time-consuming endeavor. While high-throughput screening (HTS) plays a critical role in the discovery stage, it is one of many factors contributing to these challenges. In certain contexts, virtual screening can complement the HTS, potentially offering a more streamlined approach in the initial stages of drug discovery. Molecular docking is an example of a popular virtual screening technique that is often used for this purpose; however, its effectiveness can vary greatly. This has led to the use of consensus docking approaches that combine results from different docking methods to improve the identification of active compounds and reduce the occurrence of false positives. However, many of these methods do not fully leverage the latest advancements in molecular docking. In response, we present ESSENCE-Dock (Effective Structural Screening ENrichment ConsEnsus Dock), a new consensus docking workflow aimed at decreasing false positives and increasing the discovery of active compounds. By utilizing a combination of novel docking algorithms, we improve the selection process for potential active compounds. ESSENCE-Dock has been made to be user-friendly, requiring only a few simple commands to perform a complete screening while also being designed for use in high-performance computing (HPC) environments.

Abstract C153: Rare BRAF mutations in menlanoma and beyond: Rationalizing the afficacy of B-raf inhibitors via HPC-based in silico/in vitro investigations

Abstract Melanoma is a highly aggressive type of cancer that affects the integumentary system and for which a family history of the disease, a fair complexion, the presence of a large number of moles, and skin exposure to natural/artificial ultraviolet (UV) radiation constitute established risk factors for malignancy development. The v-RAF murine sarcoma viral oncogene homolog B1 (BRAF) is a serine/threonine protein kinase that plays a critical role in the RAS-RAF-MEK-ERK mitogen–activated protein kinase (MAPK) cell signaling pathway. Despite over 20 years of research into the regulation and function of the RAF proteins, it was only relatively recently that the BRAF isoform is mutated at a high frequency in different human cancers, identifying this kinase as another important oncogene along this pathway. FFocusing on melanoma, activating BRAF mutations were identified in 37%−53% of patients with the malignant form of this pathology, the most prevalent, missense kinase alteration being the replacement of the small, apolar residue valine (V) with the negatively-charged, phosphorylation mimicking glutamic acid (E) at position 600 (BRAFV600E) within the protein catalytic (i.e., ATP binding) site. It is interesting to note that BRAFV600E is not a UV-driven mutation; in fact, it may also be found in the molecular profile of atypical benign nevi; moreover, although BRAFV600E is the most frequent mutation (accounting for nearly 40% of all melanomas and in approximately 90% of BRAF-mutant melanomas), other more or less frequent activating variants in BRAF have also been reported, which include other aminoacidic substitutions at the same position (e.g., V600K/D/R), G469A/V/S, L597Q/R/S/V, A598V, and K601E/N/T, among others. Importantly, these mutations are also found in other cancers, e.g., thyroid cancer, colorectal cancer, and some brain malignancies including glioblastoma, pilocytic astrocytoma, and pediatric low-grade glioma. In this multidisciplinary work, we focused on some of these rare BRAF variants and studied their ability to interact with several, first-line FDA approved BRAF initbitors (BRAFis) from a structural, chemico-phisical and in vitro perspectives. To the purpose, we employed a combitation of in silico techniques, based on the use of high-performance computing (HPC), chemico-physical and spectroscopical techniques, and in vitro tests to verify the interactions between these mutated BRAF isoforms with the inhibitors. As a main point, we were able to determine the binding thermodinamics and kinetics of the selected BRAFis to the mutant BRAF proteins, and each BRAF/drug complex was characterized from a structural standpoint. All these information ultimately revelaed that some of the BRAF isoforms are able to effectively bind the corresponding BRAi, with consequent switch-off of the signaling patwhay, at least in vitro. The presented mutlidisciplinary protocol could be adoped in the future to predict the response of newly discovered BARF variants and/or variants of unknown significance towards current or newly-designed BRAFis. Citation Format: Sabrina Pricl, Erik Laurini, Domenico Marson, Gabriele Cavalieri. Rare BRAF mutations in menlanoma and beyond: Rationalizing the afficacy of B-raf inhibitors via HPC-based in silico/in vitro investigations [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2023 Oct 11-15; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2023;22(12 Suppl):Abstract nr C153.

High-Density Exploration of Activity States in a Multi-Area Brain Model

To simulate whole brain dynamics with only a few equations, biophysical, mesoscopic models of local neuron populations can be connected using empirical tractography data. The development of mesoscopic mean-field models of neural populations, in particular, the Adaptive Exponential (AdEx mean-field model), has successfully summarized neuron-scale phenomena leading to the emergence of global brain dynamics associated with conscious (asynchronous and rapid dynamics) and unconscious (synchronized slow-waves, with Up-and-Down state dynamics) brain states, based on biophysical mechanisms operating at cellular scales (e.g. neuromodulatory regulation of spike-frequency adaptation during sleep-wake cycles or anesthetics). Using the Virtual Brain (TVB) environment to connect mean-field AdEx models, we have previously simulated the general properties of brain states, playing on spike-frequency adaptation, but have not yet performed detailed analyses of other parameters possibly also regulating transitions in brain-scale dynamics between different brain states. We performed a dense grid parameter exploration of the TVB-AdEx model, making use of High Performance Computing. We report a remarkable robustness of the effect of adaptation to induce synchronized slow-wave activity. Moreover, the occurrence of slow waves is often paralleled with a closer relation between functional and structural connectivity. We find that hyperpolarization can also generate unconscious-like synchronized Up and Down states, which may be a mechanism underlying the action of anesthetics. We conclude that the TVB-AdEx model reveals large-scale properties identified experimentally in sleep and anesthesia.

Developing Edge AI Computer Vision for Smart Poultry Farms Using Deep Learning and HPC.

This research describes the use of high-performance computing (HPC) and deep learning to create prediction models that could be deployed on edge AI devices equipped with camera and installed in poultry farms. The main idea is to leverage an existing IoT farming platform and use HPC offline to run deep learning to train the models for object detection and object segmentation, where the objects are chickens in images taken on farm. The models can be ported from HPC to edge AI devices to create a new type of computer vision kit to enhance the existing digital poultry farm platform. Such new sensors enable implementing functions such as counting chickens, detection of dead chickens, and even assessing their weight or detecting uneven growth. These functions combined with the monitoring of environmental parameters, could enable early disease detection and improve the decision-making process. The experiment focused on Faster R-CNN architectures and AutoML was used to identify the most suitable architecture for chicken detection and segmentation for the given dataset. For the selected architectures, further hyperparameter optimization was carried out and we achieved the accuracy of AP = 85%, AP50 = 98%, and AP75 = 96% for object detection and AP = 90%, AP50 = 98%, and AP75 = 96% for instance segmentation. These models were installed on edge AI devices and evaluated in the online mode on actual poultry farms. Initial results are promising, but further development of the dataset and improvements in prediction models is needed.

Cyber Threat Intelligence Discovery using Machine Learning from the Dark Web

Cyber threat intelligence (CTI) is an actionable information or insight an organization uses to understand potential vulnerabilities it does have and threats it is facing. One important CTI for proactive cyber defense is exploit type with possible values system, web, network, website or Mobile. This study compares the performance of machine learning algorithms in predicating exploit types using form posts in the dark web, which is a semi- structured dataset collected from dark web. The study uses the CRISP data science approach. The results of the study show that machine learning algorithms which are function-based including support vector machine and deep-learning using artificial neural network are more accurate than those algorithms which are based on tree including Random Forest and Decision-Tree for CTI discovery from semi-structured dataset. Future research will include the use of high-performance computing and advanced deep-learning algorithms.

Dynamic non-local damage analysis using an octree pattern-based massively parallel explicit solver

This paper presents the development of a massively parallel explicit solver based on the central difference method (CDM) for the dynamic analysis of damage processes. In this context, the material degradation is incorporated using an integral-type non-local isotropic damage model. A fully automatic preprocessing is enabled by following the octree-based mesh generation paradigm. To handle neighbouring elements with different sizes and to avoid hanging nodes in the generated mesh, polyhedral elements are implemented within the framework of the scaled boundary finite element method (SBFEM). A pre-computation approach is advocated to develop a highly efficient solver, where the limited number of master cells in a balanced octree grid are exploited. The parallelisation is carried out using a mesh-partitioning technique and message-passing-interface (MPI) directives, which is highly complementary to the use in high-performance computing (HPC) facilities. In order to showcase the performance of the proposed solver, a problem of practical interest is selected, featuring the analysis of an XCT image-based model of concrete containing about 1 billion voxels. The simulated results have been obtained within minutes by exploiting the computational power of approximately 12k cores.

Interpretable and Explainable Machine Learning for Materials Science and Chemistry

ConspectusMachine learning has become a common and powerful tool in materials research. As more data become available, with the use of high-performance computing and high-throughput experimentation, machine learning has proven potential to accelerate scientific research and technology development. Though the uptake of data-driven approaches for materials science is at an exciting, early stage, to realize the true potential of machine learning models for successful scientific discovery, they must have qualities beyond purely predictive power. The predictions and inner workings of models should provide a certain degree of explainability by human experts, permitting the identification of potential model issues or limitations, building trust in model predictions, and unveiling unexpected correlations that may lead to scientific insights. In this work, we summarize applications of interpretability and explainability techniques for materials science and chemistry and discuss how these techniques can improve the outcome of scientific studies. We start by defining the fundamental concepts of interpretability and explainability in machine learning and making them less abstract by providing examples in the field. We show how interpretability in scientific machine learning has additional constraints compared to general applications. Building upon formal definitions in machine learning, we formulate the basic trade-offs among the explainability, completeness, and scientific validity of model explanations in scientific problems. In the context of these trade-offs, we discuss how interpretable models can be constructed, what insights they provide, and what drawbacks they have. We present numerous examples of the application of interpretable machine learning in a variety of experimental and simulation studies, encompassing first-principles calculations, physicochemical characterization, materials development, and integration into complex systems. We discuss the varied impacts and uses of interpretabiltiy in these cases according to the nature and constraints of the scientific study of interest. We discuss various challenges for interpretable machine learning in materials science and, more broadly, in scientific settings. In particular, we emphasize the risks of inferring causation or reaching generalization by purely interpreting machine learning models and the need for uncertainty estimates for model explanations. Finally, we showcase a number of exciting developments in other fields that could benefit interpretability in material science problems. Adding interpretability to a machine learning model often requires no more technical know-how than building the model itself. By providing concrete examples of studies (many with associated open source code and data), we hope that this Account will encourage all practitioners of machine learning in materials science to look deeper into their models.

Optimised Pre-Processing of Raman Spectra for Colorectal Cancer Detection Using High-Performance Computing.

Spectral pre-processing is an essential step in data analysis for biomedical diagnostic applications of Raman spectroscopy, allowing the removal of undesirable spectral contributions that could mask biological information used for diagnosis. However, due to the specificity of pre-processing for a given sample type and the vast number of potential pre-processing combinations, optimisation of pre-processing via a manual "trial and error" format is often time intensive with no guarantee that the chosen method is optimal for the sample type. Here we present the use of high-performance computing (HPC) to trial over 2.4 million pre-processing permutations to demonstrate the optimisation on the pre-processing of human serum Raman spectra for colorectal cancer detection. The effect of varying pre-processing order, using extended multiplicative scatter correction, spectral smoothing, baseline correction, binning and normalization was considered. Permutations were assessed on their ability to detect patients with disease using a random forest (RF) algorithm trained with 102 patients (510 spectra) and independently tested with a set of 439 patients (1317 spectra) in a primary care patient cohort. Optimising via HPC enables improved performance in diagnostic abilities, with sensitivity increasing by 14.6%, specificity increasing by 6.9%, positive predictive value increasing by 3.4%, and negative predictive value increasing by 2.4% when compared to a standard pre-processing optimisation. Ultimate values of these metrics are very important for diagnostic adoption, and once diagnostics demonstrate good accuracy these types of optimisations can make a significant difference to roll-out of a test and demonstrating advantages over existing tests. We also provide tips/recommendations for pre-processing optimisation without the use of HPC. From the HPC permutations, recommendations for appropriate parameter constraints for conducting a more basic pre-processing optimisation are also detailed, thus helping model development for researchers not having access to HPC.

THE Most Important Book about Climate Change. Unsettled: What Climate Science Tells Us, What It Doesn’t, and Why It Matters by Steven E. Koonin

If you want to know the facts about global warming and climate change, if you want to reach your own informed opinion about what should be done about it, this is the book to read. All relevant facts are presented clearly, very much including the historical records together with scrupulous citing of primary sources, among which the official published reports, international and national, feature prominently. The author’s credentials for this work could hardly be more impressive. Steven Koonin has been a successful physicist, including pioneering work on the use of high-performance computers in simulation and modeling. He gained some insights from the administrative role of Vice President and Provost at Caltech; another highly relevant experience as Chief Scientist for the oil company BP, focusing on renewable energy possibilities; and he was an Undersecretary for Science in the Department of Energy in the Obama Administration, focusing on energy technologies and climate science. He understands the viewpoints of scientists, of government, and of industry; in other words, he knows whereof he speaks. Throughout the book, Koonin takes deliberate, explicit care to write dispassionately and factually. At the outset, he acknowledges that the path to convincing any wide audience calls for arousing emotion, so that scientists are inevitably in a dilemma: whether simply to describe the facts and hope for appropriate attention to them, or whether to find emotionally impressive ways of spinning and hyping the facts. Koonin takes the first course quite effectively. For example, “the science doesn’t support what’s portrayed in most popular discussions” might also have been expressed as, say, “We are being massively misled about so-called climate science, by media, pundits, and official climate scientists.” Thus the book’s careful, factually based wording may easily seem incongruous as it describes unwarranted and unscrupulous mis-stating of facts, spinning, and hyping, which seem to be an integral part of official “consensual” “climate science,” to which the book refers as “The Science” to contrast with actual evidence-based science.

Optimization of the Functional Decomposition of Parallel and Distributed Computations in Graph Coloring With the Use of High-Performance Computing

This article presents methods for correct decomposition for high performance computations related to large sets of graphs. These computations contain large number of calls of sequential, recursive algorithm for NP-complete problem - proper edge coloring of graph. Decomposition of this computational problem is not trivial, since the number of recursions in various parts of the computation is different and causes a high load and time imbalance. We designed, implemented and experimentally verified a new decomposition method that significantly reduces the computational time for a large set of graphs (up to 404 million graphs). This method ensures the same duration of computational time for partial subtasks and thus eliminates the need to wait for synchronization of parallel computations.

Using HPC infrastructures for deep learning applications in fusion research

In the fusion community, the use of high performance computing (HPC) has been mostly dominated by heavy-duty plasma simulations, such as those based on particle-in-cell and gyrokinetic codes. However, there has been a growing interest in applying machine learning for knowledge discovery on top of large amounts of experimental data collected from fusion devices. In particular, deep learning models are especially hungry for accelerated hardware, such as graphics processing units (GPUs), and it is becoming more common to find those models competing for the same resources that are used by simulation codes, which can be either CPU- or GPU-bound. In this paper, we give examples of deep learning models—such as convolutional neural networks, recurrent neural networks, and variational autoencoders—hat can be used for a variety of tasks, including image processing, disruption prediction, and anomaly detection on diagnostics data. In this context, we discuss how deep learning can go from using a single GPU on a single node to using multiple GPUs across multiple nodes in a large-scale HPC infrastructure.

The ethical use of high-performance computing and artificial intelligence: fighting COVID-19 at Barcelona Supercomputing Center

The COVID-19 pandemic has created an extraordinary medical, economic and humanitarian emergency. Artificial intelligence, in combination with other digital technologies, is being used as a tool to support the fight against the viral pandemic that has affected the entire world since the beginning of 2020. Barcelona Supercomputing Center collaborates in the battle against the coronavirus in different areas: the application of bioinformatics for the research on the virus and its possible treatments, the use of artificial intelligence, natural language processing and big data techniques to analyse the spread and impact of the pandemic, and the use of the MareNostrum 4 supercomputer to enable massive analysis on COVID-19 data. Many of these activities have included the use of personal and sensitive data of citizens, which, even during a pandemic, should be treated and handled with care. In this work we discuss our approach based on an ethical, transparent and fair use of this information, an approach aligned with the guidelines proposed by the European Union.

Review of Soil-Structure Interaction Based on Continuum Mechanics Theory and Use of High Performance Computing

Recent achievement of research on soil-structure interaction (SSI) is reviewed, with a main focus on the numerical analysis. The review is based on the continuum mechanics theory and the use of high-performance computing (HPC) and clarifies the characteristics of a wide range of treatment of SSI from a simplified model to a high fidelity model. Emphasized is that all the treatment can be regarded as the result of the mathematical approximations in solving a physical continuum mechanics problem of a soil-structure system. The use of HPC is inevitable if we need to obtain a solution of higher accuracy and finer resolution. An example of using HPC for the analysis of SSI is presented.

Mitigating Uncertainty in Developing and Applying Scientific Applications in an Integrated Computing Environment

Effective solving complex mathematical modeling problems is based on the use of high-performance computing. Clouds, grids, and public access supercomputer centers are commonly used platforms. Their integration into a unified environment provides possibilities for carrying out mass large-scale scientific experiments and efficient scalable resource allocation at different stages of the application design and execution. However, end-users have to carefully select optimization criteria such as completion time, deadlines, reliability, cost, etc. It is a complicated problem due to integrated resources differ significantly in their computing capabilities, hardware and software platforms, system architectures, user interfaces, etc. The paper presents new features of the Orlando Tools framework for the development of distributed applied software packages (scalable scientific applications) that mitigates various types of uncertainties arising from the job distribution in the integrated computing environment. It provides continuous integration, delivery, and deployment of applied and system software to significantly mitigate the negative impact of uncertainty on problem-solving time, computation reliability, and resource efficiency. An experimental analysis of the sustainable design and development of the real energy sector clearly demonstrates the advantages of the tools.

A Computational Architecture for Inference of a Quantized-CNN for Detecting Atrial Fibrillation

Atrial Fibrillation is a common cardiac arrhythmia, which is characterized by an abnormal heartbeat rhythm that can be life-threatening. Recently, researchers have proposed several Convolutional Neural Networks (CNNs) to detect Atrial Fibrillation. CNNs have high requirements on computing and memory resources, which usually demand the use of High Performance Computing (eg, GPUs). This high energy demand is a challenge for portable devices. Therefore, efficient hardware implementations are required. We propose a computational architecture for the inference of a Quantized Convolutional Neural Network (Q-CNN) that allows the detection of the Atrial Fibrillation (AF). The architecture exploits data-level parallelism by incorporating SIMD-based vector units, which is optimized in terms of computation and storage and also optimized to perform both the convolutional and fully connected layers. The computational architecture was implemented and tested in a Xilinx Artix-7 FPGA. We present the experimental results regarding the quantization process in a different number of bits, hardware resources, and precision. The results show an accuracy of 94% accuracy for 22-bits. This work aims to be the basis for the future implementation of a portable, low-cost, and high-reliability device for the diagnosis of Atrial Fibrillation.

Computational biomedicine. Part 1: molecular medicine

Computational biomedicine. Part 1: molecular medicine

Analítica Prescriptiva en Sistemas con Grandes Flujos de Eventos

Los procesos de negocio exigen tomar decisiones rápidas para lograr la adaptación constante a los cambios en búsqueda de mejorar el desempeño y aprovechar las oportunidades. Una posibilidad es resolver el proceso computacionalmente. Para ello es necesario producir analíticas que transformen los datos en conocimiento para la toma de decisiones. Existen varios tipos de analíticas, en este trabajo se introduce una línea de investigación enfocada en la analítica prescriptiva, como nivel más avanzado, capaz de calcular acciones a ser ejecutadas en el momento (decisiones operativas) o en el futuro (decisiones tácticas para corto y mediano plazo, decisiones estratégicas para largo plazo) para lograr un objetivo deseado. El cálculo de las acciones involucra el procesamiento del flujo de eventos del negocio en forma de datastreams, la aplicación de técnicas y algoritmos de Soft Computing e Inteligencia Computacional y, derivado de la necesidad de bajos tiempos de respuesta, el empleo de Computación de Alto Desempeño.

The Reference Model: An Initial Use Case for COVID-19.

The outbreak of the coronavirus disease-19 (COVID-19) pandemic has created much speculation on the behavior of the disease. Some of the questions that have been asked can be addressed by computational modeling based on the use of high-performance computing (HPC) and machine learning techniques. The Reference Model previously used such techniques to model diabetes. The Reference Model is now used to answer a few questions on COVID-19, while changing the traditional susceptible-infected-recovered (SIR) model approach. This adaptation allows us to answer questions such as the probability of transmission per encounter, disease duration, and mortality rate. The Reference Model uses data on US infection and mortality from 52 states and territories combining multiple assumptions of human interactions to compute the best fitting parameters that explain the disease behavior for given assumptions and accumulated data from April 2020 to June 2020.This is a preliminary report aimed at demonstrating the possible use of computational models based on computing power to aid comprehension of disease characteristics. This infrastructure can accumulate models and assumptions from multiple contributors.

PROGNOZOWANIE PROCESÓW FRONTOGENEZY I FRONTOLIZY METODĄ Q-WEKTORÓW Z ZASTOSOWANIEM PRZETWARZANIARÓWNOLEGŁEGO

Artykuł przedstawia autorską metodykę prognozowania procesów frontogenezy i frontolizy metodą Q-wektorów. Do jej realizacji wykorzystano zdjęcia satelitarne i dane numeryczne oraz zastosowano ideę przetwarzania równoległego. Metoda Q-wektorów służy do wyznaczania obszarów występowania prądów pionowych, określania ich kierunku i intensywności. Obszary Q-wektorów wskazują miejsca zbieżności i rozbieżności pola przepływu atmosfery, a tym samym wyznaczają rejony występowania prądów wstępujących i zstępujących. Obliczona na podstawie metody Q-wektorów funkcja frontogenetyczna wyznacza obszary frontogenezy i frontolizy. Przetwarzanie danych numerycznych wymaga zastosowania wydajnych systemów obliczeniowych. Zwiększanie wydajności współczesnych komputerów poprzez podwyższanie częstotliwości ich zegarów wewnętrznych w zasadzie dobiega końca. Właściwym kierunkiem rozwoju wydaje się być budowa komputerów o strukturach równoległych.

Portability Study of an OpenCL Algorithm for Automatic Target Detection in Hyperspectral Images

In the last decades, the problem of target detection has received considerable attention in remote sensing applications. When this problem is tackled using hyperspectral images with hundreds of bands, the use of high-performance computing (HPC) is essential. One of the most popular algorithms in the hyperspectral image analysis community for this purpose is the automatic target detection and classification algorithm (ATDCA). Previous research has already investigated the mapping of ATDCA on HPC platforms such as multicore processors, graphics processing units (GPUs), and field-programmable gate arrays (FPGAs), showing impressive speedup factors (after careful fine-tuning) that allow for its exploitation in time-critical scenarios. However, the lack of standardization resulted in most implementations being too specific to a given architecture, eliminating (or at least making extremely difficult) code reusability across different platforms. In order to address this issue, we present a portability study of an implementation of ATDCA developed using the open computing language (OpenCL). We focus on cross-platform parameters such as performance, energy consumption, and code design complexity, as compared to previously developed (hand-tuned) implementations. Our portability study analyzes different strategies to expose data parallelism as well as enable the efficient exploitation of complex memory hierarchies in heterogeneous devices. We also conduct an assessment of energy consumption and discuss metrics to analyze the quality of our code. The conducted experiments—using synthetic and real hyperspectral data sets collected by the Hyperspectral Digital Imagery Collection Experiment (HYDICE) and NASA’s Airborne Visible Infra-Red Imaging Spectrometer (AVIRIS)—demonstrate, for the first time in the literature, that portability across different HPC platforms can be achieved for real-time target detection in hyperspectral missions.