Graphics Processing Unit Card Research Articles

Revolutionizing granular matter simulations by high-performance ray tracing discrete element method for arbitrarily-shaped particles

It remains a challenge for discrete element method to simulate large-scale systems with arbitrarily shaped particles. This paper presents a novel approach called Ray Tracing Discrete Element Method (RTDEM) to offer ultra-high efficiency in simulating particles with arbitrary shapes. In RTDEM, we use a triangular mesh to represent the surface of a particle for greater flexibility in shape representation. Template meshes are further introduced to group particles of different shapes to save computing memory. Key to RTDEM are innovative ray tracing-based algorithms specially designed for accurate and efficient contact detection and resolution between arbitrarily shaped particles. Based on the RT algorithms, discrete potential field functions are further proposed for the discrete mesh vertices and facets to define energy-conservative contact models that include vertex-potential, facet-potential models, and their combination. The integration of these key elements into RTDEM offers a whole new DEM framework that readily empowers any parallel computing architecture for high acceleration on granular media modeling. It matches particularly well with the Graphics Processing Units (GPUs) to fully unleash the computing power of ray tracing cores for large-scale DEM simulations of arbitrarily shaped particles, even on a personal computer with a common gaming GPU card.

GPU accelerated volumetric lattice Boltzmann model for image-based hemodynamics in portal hypertension

We are developing a new technique for monitoring portal hypertension by the pressure gradient between the portal vein and the inferior vena cava (PPG) based on non-invasive measurements (MRI images). Massive parametrization and classification are required to investigate the underlying relationship between the porosity and the stages of liver cirrhosis numerically, and the hepatic-portal venous system is a multi-scale system. Both of them need high computational costs. The suitability of the lattice Boltzmann method for GPU (Graphics Processing Unit) parallel computation provides an opportunity to overcome it. In this paper, we perform GPU parallelization and optimization for the volumetric lattice Boltzmann method with arbitrary geometry based on images. Three application cases, including pipe flow, hemodynamics in the portal venous system, and hemodynamics in a simple hepatic-portal venous system, are employed to prove the method can be applied in the hepatic-portal venous system based on accuracy and efficiency. The reliability of the model is qualitatively validated by the analytical solution of velocity and pressure difference distribution of pipe flow and quantitatively confirmed by the pulsatility of velocity and pressure difference that can be neglected in the portal venous system. The performance of the application cases is examined with Intel Broadwell E5-2683 v3@ 2.30 GHz (CPU) and NVIDIA Tesla V100 16GB (GPU). It shows the GPU algorithm for sparse geometry (SPARSE) has a similar speed to the regular GPU algorithm for dense geometry (DENSE) when the fluid volume fraction (q) is close to 1. And SPARSE speeds up to 2.2 times compared with DENSE when q is in the range of 0.19∼0.27. Meanwhile, the saving ratio of memory cost depends on q. For a numerical case in the hepatic-portal venous system, i.e., a large-scale system, parallel execution can be converged around half an hour with SPARSE, while the memory spills the limitation with DENSE with a single GPU. Hence, multi-GPU implementation is applied to release the limitation, and it can improve performance by increasing the number of GPU cards. In summary, the method presented in the paper is feasibility applied in the hepatic-portal venous system, laying the foundation of the new technique development.

GPU-accelerated artificial neural network potential for molecular dynamics simulation

Artificial neural network potential (ANNP), obtained by training a large database by the first-principles calculations, has become popular in molecular dynamics (MD) simulation since it can capture accurate physical and chemical properties. However, the complex procedure and heavy data dependence during implementation make the performance of CPU-only runs worse, which further limits its application. In this contribution, we report a flexible computation method for ANNP in LAMMPS, in which the simulation box is divided into several parts in accordance with the resource on the accelerator such as the size of global memory and the number of work items (cores). The number of dividing parts has little influence on the performance when the number of calculated atoms per loop is larger than the number of work items on the device. In this approach, the forces of neighbor atoms are updated using hierarchical memory without atomic operation. Typical dynamic and static tests are performed to validate the implementation. The results show that our approach is 12 or 13 times faster when using one graphics processing unit (GPU) compared with 8-MPI tasks CPU-only runs. Additionally, this implementation is supported for CUDA- and OpenCL-enabled GPU cards.

An ST 2110 Network-Attached Display Adapter for Virtualizing Production

Graphics processing unit (GPU) installations in post-production and broadcast facilities require video to be sent remotely conforming to the SMPTE ST 2110 standards for plug-and-play device connectivity. Furthermore, in such facilities, workstations are now being virtualized and hosted in the data center with what would have previously been a physically attached serial digital video (SDI) reference display replaced by an Internet Protocol (IP) network connection. We present a SMPTE ST 2110 virtual network-attached reference display adapter using commercial off-the-shelf (COTS) hardware that presents the GPU and network interface card (NIC) as a display on the computer workstation desktop. Color-accurate rendered video frames are transmitted directly from GPU device memory. Alignment of the GPU display of the rendered frame and the synchronization of the audio, video, and ancillary data streams are maintained by precision time protocol (PTP). Advanced Media Workflow Association Networked Media Open Specifications (AMWA NMOS) integration permits the discovery and registration of the virtual network-attached reference display adapter as a source node on the media network such that it can be remotely managed and controlled by software greatly simplifying the administration compared with a physically connected display and associated cables. Content creation applications send ST 2110-compliant video, audio, and ancillary data flows on the media network from a virtualized computer workstation desktop without the cost of directly integrating ST 2110 transmission capabilities.

Enabling Part-Scale Scanwise process simulation for predicting melt pool variation in LPBF by combining GPU-based Matrix-free FEM and adaptive Remeshing

• A highly efficient GPU-based matrix-free finite element modeling with adaptive remeshing framework is developed to overcome the high computational expense of l -PBF thermal process modeling, making it feasible to do detailed scanwise thermal simulation for part-scale within days. • The effects of the scanning strategy, geometrical features, and proximity to the turning points on the melt pool size variability and lack of fusion porosity are investigated for two different parts. • The predicted variation in melt pool size and lack of fusion porosity has good correlation with experimental measurements. This work proposes to combine matrix-free finite element modeling (FEM), adaptive remeshing, and graphical processing unit (GPU) computing to enable, for the first time, scanwise process simulation of the Laser Powder Bed Fusion (L-PBF) process with temperature-dependent thermophysical properties at the part scale. Compared to the conventional FEM using the global stiffness approach and a uniform mesh running on 10 CPU cores, l -PBF process simulation based on the proposed methodology running on a GPU card with 5,120 Compute Unified Device Architecture (CUDA) cores enables a speedup of over 10,000x. This significant speedup facilitates detailed thermal history and melt pool geometry predictions at high resolution for centimeter-scale parts within days of computation time. Two parts consisting of various geometric features are simulated to reveal the effects of scan strategy and local geometry on melt pool size variation, which correlate well with melt pool and lack-of-fusion porosity measurements obtained via experiment.

Stationary Stokes solver for single-phase flow in porous media: A blastingly fast solution based on Algebraic Multigrid Method using GPU

The paper is focused on high efficiency Stokes solver that is applied to the incompressible flow in porous media. Computational domains are represented by binarized 3D computed tomography voxel models. The solution procedure is constructed around a fast algebraic multigrid solver that utilizes the power of graphics processing units (GPUs). In order to minimize memory footprint and accelerate the solver, a simple MAC-type staggered finite difference discretization is used and a coupled Stokes saddle-point type system is solved directly on a GPU. The MAC discretization is discussed, taking particular care of internal solid boundaries around pores. The method includes topological domain analysis on a GPU, which removes isolated volumes with no flow in the domain (based on the connected component labeling), depending on the boundary conditions. We consider various types of boundary conditions and efficient parallel strategies for the GPUs, including fast matrix assembly and residual regularization. Our method is extensively benchmarked both against analytical solutions and applied problems from digital rock physics in terms of computational wall time, precision, approximation order, and convergence. We demonstrate that it takes up to 5–23 s on a modern GPU card to obtain a solution with 1⋅10−6 error residuals for 3D geometries with 300–4503 voxels and a porosity range of 5–37%.

Computing large 2D convolutions on GPU efficiently with the im2tensor algorithm

Attaining the best possible throughput when computing convolutions is a challenge for signal and image processing systems, be they HPC (High-Performance Computing) machines or embedded real-time targets. This importance is highlighted by the numerous methods and implementations available, often optimized for particular settings: small batched kernels or very large kernels, for example. In the meantime, GPUs (Graphics Processing Units) have become a first-class architecture for real-time and embedded processing. The power offered by those chips stems from their parallel nature, and this advantage has been exploited for convolutions in several libraries. Even more recently, the introduction of tensor cores on NVIDIA GPUs has opened up new limits in terms of attainable FLOPS (Floating-Point Operations per Second). For reaching that performance, GPU applications must use GEMMs (GEneral Matrix Multiplications), that tensor cores accelerate. We then developed an efficient GEMM-based 2D convolution algorithm in a general setting. On relatively large kernels (30–50-pixel wide), im2tensor is, to the best of our knowledge, the fastest method for computing 2D convolutions. We provide detailed performance analysis for different scenarios: small (1024\(\times\)1024) and large (4096\(\times\)4096) images, with convolutions kernels of sizes ranging 1 to 60-pixel wide, on two GPU cards: Jetson AGX Xavier (embedded) and Titan V (server-class). Moreover, the accuracy of im2tensor surpasses non-GEMM based methods, thanks to the larger-precision registers used by tensor cores for intermediate representations.

Fast Monte Carlo-Based Inverse Planning for Prostate Brachytherapy by Using Deep Learning

Inverse planning is an essential tool for optimizing the delivered radiation dose on low-dose-rate (LDR) prostate brachytherapy. Clinical inverse planning systems use the TG-43 dose computation formalism in order to perform a fast optimization. However, this method is an approximation that often leads to a dose overestimation, resulting on suboptimal plans. Alternatively, Monte Carlo simulation (MCS) can be used to obtain an accurate dose distribution, but considerably increasing the estimation time. We propose a fast inverse planning method for LDR prostate brachytherapy that uses a deep convolutional neural network (DCNN) trained on a graphics processing unit (GPU)-based MCS generated database to estimate the dose distribution on the prostate and organs at risk. Segmentations of the organs and seeds’ positions are given as the DCNN input. The mean percent error on the test set was −1.19±0.94% within the prostate. The DCNN was used to estimate the dosimetric parameters in each organ for every configuration of the optimization loop. The dosimetric parameters of the final DCNN-based brachytherapy plans were in good agreement compared to the same plans recalculated with a full MCS. The proposed inverse planning based on DCNN was capable to reach an equivalent level of accuracy with Monte Carlo with a runtime in less than 1 min using conventional GPU card.

Validation of organ dose calculations with PyMCGPU-IR in realistic interventional set-ups.

Interventional radiology procedures are associated with high skin dose exposure. The 2013/59/EURATOM Directive establishes that the equipment used for interventional radiology must have a device or a feature informing the practitioner of relevant parameters for assessing patient dose at the end of the procedure. This work presents and validates PyMCGPU-IR, a patient dose monitoring tool for interventional cardiology and radiology procedures based on MC-GPU. MC-GPU is a freely available Monte Carlo (MC) code of photon transport in a voxelized geometry which uses the computational power of commodity Graphics Processing Unit cards (GPU) to accelerate calculations. PyMCGPU-IR was validated against two different experimental set-ups. The first one consisted of skin dose measurements for different beam angulations on an adult Rando Alderson anthropomorphic phantom. The second consisted of organ dose measurements in three clinical procedures using the Rando Alderson phantom. The results obtained for the skin dose measurements show differences below 6%. For the clinical procedures the differences are within 20% for most cases. PyMCGPU-IR offers both, high performance and accuracy for dose assessment when compared with skin and organ dose measurements. It also allows the calculation of dose values at specific positions and organs, the dose distribution and the location of the maximum doses per organ. In addition, PyMCGPU-IR overcomes the time limitations of CPU-based MC codes.

FSEI-GPU: GPU accelerated simulations of the fluid–structure–electrophysiology interaction in the left heart

The reliability of cardiovascular computational models depends on the accurate solution of the hemodynamics, the realistic characterization of the hyperelastic and electric properties of the tissues along with the correct description of their interaction. The resulting fluid–structure–electrophysiology interaction (FSEI) thus requires an immense computational power, usually available in large supercomputing centers, and requires long time to obtain results even if multi–CPU processors are used (MPI acceleration). In recent years, graphics processing units (GPUs) have emerged as a convenient platform for high performance computing, as they allow for considerable reductions of the time–to–solution.This approach is particularly appealing if the tool has to support medical decisions that require solutions within reduced times and possibly obtained by local computational resources. Accordingly, our multi–physics solver [1] has been ported to GPU architectures using CUDA Fortran to tackle fast and accurate hemodynamics simulations of the human heart without resorting to large–scale supercomputers. This work describes the use of CUDA to accelerate the FSEI on heterogeneous clusters, where both the CPUs and GPUs are used in synergistically with minor modifications of the original source code. The resulting GPU accelerated code solves a single heartbeat within a few hours (from three to ten depending on the grid resolution) running on premises computing facility made of few GPU cards, which can be easily installed in a medical laboratory or in a hospital, thus opening towards a systematic computational fluid dynamics (CFD) aided diagnostic.

Custom Built of Smart Computing Platform for Supporting Optimization Methods and Artificial Intelligence Research

This paper describes a prototype of a computing platform dedicated to artificial intelligence explorations. The platform, dubbed as PakCarik, is essentially a high throughput computing platform with GPU (graphics processing units) acceleration. PakCarik is an Indonesian acronym for Platform Komputasi Cerdas Ramah Industri Kreatif, which can be translated as “Creative Industry friendly Intelligence Computing Platform”. This platform aims to provide complete development and production environment for AI-based projects, especially to those that rely on machine learning and multiobjective optimization paradigms. The method for constructing PakCarik was based on a computer hardware assembling technique that uses commercial off-the-shelf hardware and was tested on several AI-related application scenarios. The testing methods in this experiment include: high-performance lapack (HPL) benchmarking, message passing interface (MPI) benchmarking, and TensorFlow (TF) benchmarking. From the experiment, the authors can observe that PakCarik's performance is quite similar to the commonly used cloud computing services such as Google Compute Engine and Amazon EC2, even though falls a bit behind the dedicated AI platform such as Nvidia DGX-1 used in the benchmarking experiment. Its maximum computing performance was measured at 326 Gflops. The authors conclude that PakCarik is ready to be deployed in real-world applications and it can be made even more powerful by adding more GPU cards in it.

Multi-GPU implementation of a cellular automaton model for dendritic growth of binary alloy

A parallel three-dimensional (3D) cellular automaton (CA) model is developed for dendritic growth simulation during solidification of binary alloy. The CA model is implemented using the graphic processing unit (GPU). Compute Unified Device Architecture (CUDA) is combined with Message Passing Interface (MPI) to perform large-scale simulation on the multi-GPU architecture. A CUDA-Aware OpenMPI is used to reduce the communication cost between different GPU cards. The accuracy and the speedup ratio of the CA model are investigated. First, the simulated steady-state tip velocity under different undercooling is compared with the LGK analytical solution. Then, dendritic growth during directional solidification and equiaxed multi-dendrite growth are simulated. Dendritic growth simulation with a domain size up to 453 million grids is performed. Finally, the speedup ratio of the CA model using four GPUs is compared with that using four CPU processes via MPI, which leads to a maximal speedup ratio of 152.19. The developed CA model achieves high parallel performance compared with the MPI-based parallel model. The developed CA model is capable of predicting the columnar to equiaxed transition (CET) event during directional solidification. The results indicate that increasing cooling rate promotes the occurrence of the equiaxed dendrites in the melt ahead of the solidification front, and equiaxed dendrites are prone to nucleate in the columnar dendrite groove zone when a middle cooling rate is applied.

Accelerating Bayesian microseismic event location with deep learning

Abstract. We present a series of new open-source deep-learning algorithms to accelerate Bayesian full-waveform point source inversion of microseismic events. Inferring the joint posterior probability distribution of moment tensor components and source location is key for rigorous uncertainty quantification. However, the inference process requires forward modelling of microseismic traces for each set of parameters explored by the sampling algorithm, which makes the inference very computationally intensive. In this paper we focus on accelerating this process by training deep-learning models to learn the mapping between source location and seismic traces for a given 3D heterogeneous velocity model and a fixed isotropic moment tensor for the sources. These trained emulators replace the expensive solution of the elastic wave equation in the inference process. We compare our results with a previous study that used emulators based on Gaussian processes to invert microseismic events. For fairness of comparison, we train our emulators on the same microseismic traces and using the same geophysical setting. We show that all of our models provide more accurate predictions, ∼ 100 times faster predictions than the method based on Gaussian processes, and a 𝒪(105) speed-up factor over a pseudo-spectral method for waveform generation. For example, a 2 s long synthetic trace can be generated in ∼ 10 ms on a common laptop processor, instead of ∼ 1 h using a pseudo-spectral method on a high-profile graphics processing unit card. We also show that our inference results are in excellent agreement with those obtained from traditional location methods based on travel time estimates. The speed, accuracy, and scalability of our open-source deep-learning models pave the way for extensions of these emulators to generic source mechanisms and application to joint Bayesian inversion of moment tensor components and source location using full waveforms.

Parallel Large-scale MOSFET Circuit Simulation using Multi-core CPU and Time-saving Techniques

In this paper, we investigate the acceleration properties of fine-grain parallel large-scale MOSFET circuit simulation using multi-core CPU. Several CPU have been tested for understanding maximum possible accelerations. A parallel circuit simulator, MOSTIME, is used to simulate experimental circuits. Performing tests on CPU and MOSTIME, we find that more advanced accelerations on large-scale fine-grain parallel circuit simulation would rely on time-saving software techniques and GPU (Graphic Processing Unit) cards. A newly implemented time-saving technique for MOSTIME has been tested to show promising effects.

Modeling the effect of oxygen on the chemical stage of water radiolysis using GPU-based microscopic Monte Carlo simulations, with an application in FLASH radiotherapy

Oxygen plays a critical role in determining the initial DNA damages induced by ionizing radiation. It is important to mechanistically model the oxygen effect in the water radiolysis process. However, due to the computational costs from the many body interaction problem, oxygen is often ignored or treated as a constant continuum radiolysis-scavenger background in the simulations using common microscopic Monte Carlo tools. In this work, we reported our recent progress on the modeling of the chemical stage of the water radiolysis with an explicit consideration of the oxygen effect, based upon our initial development of an open-source graphical processing unit (GPU)-based MC simulation tool, gMicroMC. The inclusion of oxygen mainly reduces the yields of and chemical radicals, turning them into highly toxic and species. To demonstrate the practical value of gMicroMC in large scale simulation problems, we applied the oxygen-simulation-enabled gMicroMC to compute the yields of chemical radicals under a high instantaneous dose rate to study the oxygen depletion hypothesis in FLASH radiotherapy. A decreased oxygen consumption rate (OCR) was found associated with a reduced initial oxygen concentration level due to reduced probabilities of reactions. With respect to dose rate, for the oxygen concentration of 21% and electron energy of 4.5 , OCR remained approximately constant (∼0.22 ) for ’s of , and reduced to 0.19 at , because the increased dose rate improved the mutual reaction frequencies among radicals, hence reducing their reactions with oxygen. We computed the time evolution of oxygen concentration under the FLASH irradiation setups. At the dose rate of and initial oxygen concentrations from 0.01% to 21%, the oxygen is unlikely to be fully depleted with an accumulative dose of 30 Gy, which is a typical dose used in FLASH experiments. The computational efficiency of gMicroMC when considering oxygen molecules in the chemical stage was evaluated through benchmark work to GEANT4-DNA with simulating an equivalent number of radicals. With an initial oxygen concentration of 3% (∼105 molecules), a speedup factor of 1228 was achieved for gMicroMC on a single GPU card when comparing with GEANT4-DNA on a single CPU.

A Smart Operator Assistance System Using Deep Learning for Angle Measurement

Manual workstations play a critical role in flexible assembly lines by enabling human responses to reconfiguration that is faster than machine responses. As a result, human is more adaptive and sometimes unreplaceable by machines in complex assembly. However, with the increasing complexity of tasks, product quality has become highly susceptible to human error due to increments in operators’ cognitive load. One of the errors that affect assembly quality is the operator’s use of a handheld tool with an unfavorable working angle when handling the workpiece. In this case, an assistive mechanism to remind workers about the wrong working angle is necessary to support the process. To this end, this study proposes an angle monitoring system to inspect the working angle of handheld tools and provide feedback in real time with minimal interruption to the assembly process. The proposed system consists of an angle measurement model and an action recognition model, which are both built using deep-learning-based object detection algorithm. Besides, the system was designed with flexibility that it is applicable to different tools and assembly tasks. A case study on fastening a high-end graphical processing unit card is investigated to evaluate their performance. Results show 95.83% and 99.83% accuracies of the models. In practice, the proposed study is expected to facilitate assembly quality by preventing the failure of angle-related operations in a timely and reliable manner.

GPU-accelerated Monte Carlo simulation of MV-CBCT

Monte Carlo simulation (MCS) is one of the most accurate computation methods for dose calculation and image formation in radiation therapy. However, the high computational complexity and long execution time of MCS limits its broad use. In this paper, we present a novel strategy to accelerate MCS using a graphic processing unit (GPU), and we demonstrate the application in mega-voltage (MV) cone-beam computed tomography (CBCT) simulation. A new framework that generates a series of MV projections from a single simulation run is designed specifically for MV-CBCT acquisition. A Geant4-based GPU code for photon simulation is incorporated into the framework for the simulation of photon transport through a phantom volume. The FastEPID method, which accelerates the simulation of MV images, is modified and integrated into the framework. The proposed GPU-based simulation strategy was tested for its accuracy and efficiency in a Catphan 604 phantom and an anthropomorphic pelvis phantom with beam energies at 2.5 MV, 6 MV, and 6 MV FFF. In all cases, the proposed GPU-based simulation demonstrated great simulation accuracy and excellent agreement with measurement and CPU-based simulation in terms of reconstructed image qualities. The MV-CBCT simulation was accelerated by factors of roughly 900–2300 using an NVIDIA Tesla V100 GPU card against a 2.5 GHz AMD Opteron™ Processor 6380.

Efficient Parallelism of Post-Quantum Signature Scheme SPHINCS

SPHINCS was recently proposed as a stateless, quantum-resilient hash-based signature scheme. However, one possible limitation of SPHINCS is its signing speed, namely, the best known implementation merely produces a few hundred of signatures per second, which is not good enough, e.g., for a social website with a huge amount of users. Aiming at improving the singing throughput, we present highly parallel and optimized implementations of SPHINCS, which can be deployed on various multi-core platforms. As a first step, we give an elementary implementation on x86/64 processors, which proves the effectiveness and correctness of our implementations. To obtain a significantly higher throughput, we implement SPHINCS on Graphics Processing Units (GPUs). Furthermore, we develop a few general and hardware-specific techniques to take full advantage of the computing power of targeted platforms. We instantiate the underlying hash functions with three primitives. Our comprehensive benchmark shows that our work outperforms all the state-of-the-art implementations of SPHINCS regarding throughput with reasonable latency, and has scalability on multiple cores and multiple GPU cards. For instance, for the key generation algorithm instantiated with ChaCha running on a GeForce GTX 1080, we obtain 5152 signatures per second which is $7.88\times$ 7 . 88 × speedup faster than a recent FPGA implementation. When upgrade to TITAN Xp, 6,651 signatures are generated in one second. With four TITAN Xp GPUs, the obtained throughput satisfies vast majority scenarios.

GPU algorithms for density matrix methods on MOPAC: linear scaling electronic structure calculations for large molecular systems.

Purification of the density matrix methods should be employed when dealing with complex chemical systems containing many atoms. The running times for these methods scale linearly with the number of atoms if we consider the sparsity from the density matrix. Since the efficiency expected from those methods is closely tied to the underlying parallel implementations of the linear algebra operations (e.g., P2 = P × P), we proposed a central processing unit (CPU) and graphics processing unit (GPU) parallel matrix-matrix multiplication in SVBR (symmetrical variable block row) format for energy calculations through the SP2 algorithm. This algorithm was inserted in MOPAC's MOZYME method, using the original LMO Fock matrix assembly, and the atomic integral calculation implemented on it. Correctness and performance tests show that the implemented SP2 is accurate and fast, as the GPU is able to achieve speedups up to 40 times for a water cluster system with 42,312 orbitals running in one NVIDIA K40 GPU card compared to the single-threaded version. The GPU-accelerated SP2 algorithm using the MOZYME LMO framework enables the calculations of semiempirical wavefunction with stricter SCF criteria for localized charged molecular systems, as well as the single-point energies of molecules with more than 100.000 LMO orbitals in less than 1h. Graphical abstract Parallel CPU and GPU purification algorithms for electronic structure calculations were implemented in MOPAC's MOZYME method. Some matrices in these calculations, e.g., electron density P, are compressed, and the developed linear algebra operations deal with non-zero entries only. We employed the NVIDIA/CUDA platform to develop GPU algorithms, and accelerations up to 40 times for larger systems were achieved.

Assessment of proton-coupled conformational dynamics of SARS and MERS coronavirus papain-like proteases: Implication for designing broad-spectrum antiviral inhibitors.

Broad-spectrum antiviral drugs are urgently needed to stop the Coronavirus Disease 2019 pandemic and prevent future ones. The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is related to the SARS-CoV and Middle East respiratory syndrome coronavirus (MERS-CoV), which have caused the previous outbreaks. The papain-like protease (PLpro) is an attractive drug target due to its essential roles in the viral life cycle. As a cysteine protease, PLpro is rich in cysteines and histidines, and their protonation/deprotonation modulates catalysis and conformational plasticity. Here, we report the pKa calculations and assessment of the proton-coupled conformational dynamics of SARS-CoV-2 in comparison to SARS-CoV and MERS-CoV PLpros using the recently developed graphical processing unit (GPU)-accelerated implicit-solvent continuous constant pH molecular dynamics method with a new asynchronous replica-exchange scheme, which allows computation on a single GPU card. The calculated pKa’s support the catalytic roles of the Cys–His–Asp triad. We also found that several residues can switch protonation states at physiological pH among which is C270/271 located on the flexible blocking loop 2 (BL2) of SARS-CoV-2/CoV PLpro. Simulations revealed that the BL2 can open and close depending on the protonation state of C271/270, consistent with the most recent crystal structure evidence. Interestingly, despite the lack of an analogous cysteine, BL2 in MERS-CoV PLpro is also very flexible, challenging a current hypothesis. These findings are supported by the all-atom fixed-charge simulations and provide a starting point for more detailed studies to assist the structure-based design of broad-spectrum inhibitors against CoV PLpros.