AI-driven multiscale simulations illuminate mechanisms of SARS-CoV-2 spike dynamics

Biomolecular simulations at the exascale: From drug design to organelles and beyond

The rapid advancement in computational power available for research offers to bring not only quantitative improvements, but also qualitative changes in the field of biomolecular simulation. Here, we review the state of biomolecular dynamics simulations at the threshold to exascale resources becoming available. Both developments in parallel and distributed computing will be discussed, providing a perspective on the state of the art of both. A main focus will be on obtaining binding and conformational free energies, with an outlook to macromolecular complexes and (sub)cellular assemblies.

Structures, dynamics, complexes, and functions: From classic computation to artificial intelligence

Computational approaches can provide highly detailed insight into the molecular recognition processes that underlie drug binding, the assembly of protein complexes, and the regulation of biological functional processes. Classical simulation methods can bridge a wide range of length- and time-scales typically involved in such processes. Lately, automated learning and artificial intelligence methods have shown the potential to expand the reach of physics-based approaches, ushering in the possibility to model and even design complex protein architectures. The synergy between atomistic simulations and AI methods is an emerging frontier with a huge potential for advances in structural biology. Herein, we explore various examples and frameworks for these approaches, providing select instances and applications that illustrate their impact on fundamental biomolecular problems.

From complex data to clear insights: visualizing molecular dynamics trajectories

Advances in simulations, combined with technological developments in high-performance computing, have made it possible to produce a physically accurate dynamic representation of complex biological systems involving millions to billions of atoms over increasingly long simulation times. The analysis of these computed simulations is crucial, involving the interpretation of structural and dynamic data to gain insights into the underlying biological processes. However, this analysis becomes increasingly challenging due to the complexity of the generated systems with a large number of individual runs, ranging from hundreds to thousands of trajectories. This massive increase in raw simulation data creates additional processing and visualization challenges. Effective visualization techniques play a vital role in facilitating the analysis and interpretation of molecular dynamics simulations. In this paper, we focus mainly on the techniques and tools that can be used for visualization of molecular dynamics simulations, among which we highlight the few approaches used specifically for this purpose, discussing their advantages and limitations, and addressing the future challenges of molecular dynamics visualization.

The structure and physical properties of a packaged bacteriophage particle

A string of nucleotides confined within a protein capsid contains all the instructions necessary to make a functional virus particle, a virion. Although the structure of the protein capsid is known for many virus species1,2, the three-dimensional organization of viral genomes has mostly eluded experimental probes3,4. Here we report all-atom structural models of an HK97 virion5, including its entire 39,732 base pair genome, obtained through multiresolution simulations. Mimicking the action of a packaging motor6, the genome was gradually loaded into the capsid. The structure of the packaged capsid was then refined through simulations of increasing resolution, which produced a 26 million atom model of the complete virion, including water and ions confined within the capsid. DNA packaging occurs through a loop extrusion mechanism7 that produces globally different configurations of the packaged genome and gives each viral particle individual traits. Multiple microsecond-long all-atom simulations characterized the effect of the packaged genome on capsid structure, internal pressure, electrostatics and diffusion of water, ions and DNA, and revealed the structural imprints of the capsid onto the genome. Our approach can be generalized to obtain complete all-atom structural models of other virus species, thereby potentially revealing new drug targets at the genome-capsid interface.

Integrating cellular electron microscopy with multimodal data to explore biology across space and time

Biology spans a continuum of length and time scales. Individual experimental methods only glimpse discrete pieces of this spectrum but can be combined to construct a more holistic view. In this Review, we detail the latest advancements in volume electron microscopy (vEM) and cryo-electron tomography (cryo-ET), which together can visualize biological complexity across scales from the organization of cells in large tissues to the molecular details inside native cellular environments. In addition, we discuss emerging methodologies for integrating three-dimensional electron microscopy (3DEM) imaging with multimodal data, including fluorescence microscopy, mass spectrometry, single-particle analysis, and AI-based structure prediction. This multifaceted approach fills gaps in the biological continuum, providing functional context, spatial organization, molecular identity, and native interactions. We conclude with a perspective on incorporating diverse data into computational simulations that further bridge and extend length scales while integrating the dimension of time.

3D animation as a tool for integrative modeling of dynamic molecular mechanisms

As the scientific community accumulates diverse data describing how molecular mechanisms occur, creating and sharing visual models that integrate the richness of this information has become increasingly important to help us explore, refine, and communicate our hypotheses. Three-dimensional (3D) animation is a powerful tool to capture dynamic hypotheses that are otherwise difficult or impossible to visualize using traditional 2D illustration techniques. This perspective discusses the current and future roles that 3D animation can play in the research sphere.

Integrative approaches for characterizing protein dynamics: NMR, CryoEM, and computer simulations

Proteins are inherently dynamic and their internal motions are essential for biological function. Protein motions cover a broad range of timescales: 10-14-10 s, spanning from sub-picosecond vibrational motions of atoms via microsecond loop conformational rearrangements to millisecond large amplitude domain reorientations. Observing protein dynamics over all timescales and connecting motions and structure to biological mechanisms requires integration of multiple experimental and computational techniques. This review reports on state-of-the-art approaches for assessing dynamics in biological systems using recent examples of virus assemblies, enzymes, and molecular machines. By integrating NMR spectroscopy in solution and the solid state, cryo electron microscopy, and molecular dynamics simulations, atomistic pictures of protein motions are obtained, not accessible from any single method in isolation. This information provides fundamental insights into protein behavior that can guide the development of future therapeutics.

Mechanisms of allostery at the viral surface through the eyes of molecular simulation

The outermost surface layer of any virus is formed by either a capsid shell or envelope. Such layers have traditionally been thought of as immovable structures, but it is becoming apparent that they cannot be viewed exclusively as static architectures protecting the viral genome. A limited number of proteins on the virion surface must perform a multitude of functions in order to orchestrate the viral life cycle, and allostery can regulate their structures at multiple levels of organization, spanning individual molecules, protomers, large oligomeric assemblies, or entire viral surfaces. Here, we review recent contributions from the molecular simulation field to viral surface allostery, with a particular focus on the trimeric spike glycoprotein emerging from the coronavirus surface, and the icosahedral flaviviral envelope complex. As emerging viral pathogens continue to pose a global threat, an improved understanding of viral dynamics and allosteric regulation will prove crucial in developing novel therapeutic strategies.

Understanding the cell: Future views of structural biology

Determining the structure and mechanisms of all individual functional modules of cells at high molecular detail has often been seen as equal to understanding how cells work. Recent technical advances have led to a flush of high-resolution structures of various macromolecular machines, but despite this wealth of detailed information, our understanding of cellular function remains incomplete. Here, we discuss present-day limitations of structural biology and highlight novel technologies that may enable us to analyze molecular functions directly inside cells. We predict that the progression toward structural cell biology will involve a shift toward conceptualizing a 4D virtual reality of cells using digital twins. These will capture cellular segments in a highly enriched molecular detail, include dynamic changes, and facilitate simulations of molecular processes, leading to novel and experimentally testable predictions. Transferring biological questions into algorithms that learn from the existing wealth of data and explore novel solutions may ultimately unveil how cells work.

Two-Stage Recognition Mechanism of the SARS-CoV-2 Receptor-Binding Domain to Angiotensin-Converting Enzyme-2 (ACE2)

The SARS-CoV-2 virus, commonly known as COVID-19, occurred in 2019. It is a highly contagious illness with effects ranging from mild symptoms to severe illness. It is also one of the best-known pathogens since more than 200,000 scientific papers occurred in the last few years. With the publication of the SARS-CoV-2 (SARS-CoV-2-CTD) spike (S) protein in a complex with human ACE2 (hACE2) (PDB (6LZG)), the molecular analysis of one of the most crucial steps on the infection pathway was possible. The aim of this manuscript is to simulate the most widely spread mutants of SARS-CoV-2, namely Alpha, Beta, Gamma, Delta, Omicron, and the first recognized variant (natural wild type). With the wide search of the hypersurface of the potential energy performed using the UNRES force field, the intermediate state of the ACE2-RBD complex was found. R403, K/N/T417, L455, F486, Y489, F495, Y501, and Y505 played a crucial role in the protein recognition mechanism. The intermediate state cannot be very stable since it will prevent the infection cascade.

Dynamics and Function of sRNA/mRNAs Under the Scrutiny of Computational Simulation Methods

Molecular dynamics simulations have proved extremely useful in investigating the functioning of proteins with atomic-scale resolution. Many applications to the study of RNA also exist, and their number increases by the day. However, implementing MD simulations for RNA molecules in solution faces challenges that the MD practitioner must be aware of for the appropriate use of this tool. In this chapter, we present the fundamentals of MD simulations, in general, and the peculiarities of RNA simulations, in particular. We discuss the strengths and limitations of the technique and provide examples of its application to elucidate small RNA's performance.

Introduction to Computational Biomedicine

The domain of computational biomedicine is a new and burgeoning one. Its areas of concern cover all scales of human biology, physiology, and pathology, commonly referred to as medicine, from the genomic to the whole human and beyond, including epidemiology and population health. Computational biomedicine aims to provide high-fidelity descriptions and predictions of the behavior of biomedical systems of both fundamental scientific and clinical importance. Digital twins and virtual humans aim to reproduce the extremely accurate duplicate of real-world human beings in cyberspace, which can be used to make highly accurate predictions that take complicated conditions into account. When that can be done reliably enough for the predictions to be actionable, such an approach will make an impact in the pharmaceutical industry by reducing or even replacing the extremely laboratory-intensive preclinical process of making and testing compounds in laboratories, and in clinical applications by assisting clinicians to make diagnostic and treatment decisions.

Adsorption of pulmonary and exogeneous surfactants on SARS-CoV-2 spike protein

COVID-19 is transmitted by airborne particles containing virions of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Coronavirus virions represent nanoparticles enveloped by a lipid bilayer decorated by a "crown" of Spike protein protrusions. Virus transmission into the cells is induced by binding of Spike proteins with ACE2 receptors of alveolar epithelial cells. Active clinical search is ongoing for exogenous surfactants and biologically active chemicals capable of hindering virion-receptor binding. Here, we explore by using coarse-grained molecular dynamics simulations the physico-chemical mechanisms of adsorption of selected pulmonary surfactants, zwitterionic dipalmitoyl phosphatidyl choline and cholesterol, and exogeneous anionic surfactant, sodium dodecyl sulfate, on the S1-domain of the Spike protein. We show that surfactants form micellar aggregates that selectively adhere to the specific regions of the S1-domain that are responsible for binding with ACE2 receptors. We find distinctly higher cholesterol adsorption and stronger cholesterol-S1 interactions in comparison with other surfactants, that is consistent with the experimental observations of the effects of cholesterol on COVID-19 infection. Distribution of adsorbed surfactant along the protein residue chain is highly specific and inhomogeneous with preferential adsorption around specific amino acid sequences. We observe preferential adsorption of surfactants on cationic arginine and lysine residues in the receptor-binding domain (RBD) that play an important role in ACE2 binding and are present in higher amounts in Delta and Omicron variants, which may lead to blocking direct Spike-ACE2 interactions. Our findings of strong selective adhesion of surfactant aggregates to Spike proteins have important implications for informing clinical search for therapeutic surfactants for curing and preventing COVID-19 caused by SARS-CoV-2 and its variants.

Computational microbiology of bacteria: Advancements in molecular dynamics simulations

Microbiology is traditionally considered within the context of wet laboratory methodologies. Computational techniques have a great potential to contribute to microbiology. Here, we describe our loose definition of "computational microbiology" and provide a short survey focused on molecular dynamics simulations of bacterial systems that fall within this definition. It is our contention that increased compositional complexity and realistic levels of molecular crowding within simulated systems are key for bridging the divide between experimental and computational microbiology.

Efficient Empirical Valence Bond Simulations with GROMACS

We describe a protocol to perform empirical valence bond (EVB) simulations using GROMACS software. EVB is a fast and reliable method that allows one to determine the reaction free-energy profiles in complex systems, such as enzymes, by employing classical force fields to represent a chemical reaction. Therefore, running EVB simulations is basically as fast as any classical molecular dynamics simulation, and the method uses standard free-energy calculations to map the free-energy change along a given reaction path. To exemplify and validate our EVB implementation, we replicated two cases of our earlier enzyme simulations. One of these addresses the decomposition of the activation free energy into its enthalpic and entropic components, and the other is focused on calculating the overall catalytic effect of the enzyme compared to the same reaction in water. These two examples give virtually identical results to those obtained with programs that were specifically designed for EVB simulations and show that the GROMACS implementation is robust and can be used for very large systems.

SARS-CoV-2 spike variants differ in their allosteric responses to linoleic acid

The SARS-CoV-2 spike protein contains a functionally important fatty acid (FA) binding site, which is also found in some other coronaviruses, e.g. SARS-CoV and MERS-CoV. The occupancy of the FA site by linoleic acid (LA) reduces infectivity by 'locking' the spike in a less infectious conformation. Here, we use dynamical-nonequilibrium molecular dynamics (D-NEMD) simulations to compare the allosteric responses of spike variants to LA removal. D-NEMD simulations show that the FA site is coupled to other functional regions of the protein, e.g. the receptor-binding motif (RBM), N-terminal domain (NTD), furin cleavage site, and regions surrounding the fusion peptide. D-NEMD simulations also identify the allosteric networks connecting the FA site to these functional regions. The comparison between the wild-type spike and four variants (Alpha, Delta, Delta plus, and Omicron BA.1) shows that the variants differ significantly in their responses to LA removal. The allosteric connections to the FA site on Alpha are generally similar to those on the wild-type protein, with the exception of the RBM and the S71-R78 region, which show a weaker link to the FA site. In contrast, Omicron is the most different variant, exhibiting significant differences in the RBM, NTD, V622-L629, and furin cleavage site. These differences in the allosteric modulation may be of functional relevance, potentially affecting transmissibility and virulence. Experimental comparison of the effects of LA on SARS-CoV-2 variants, including emerging variants, is warranted.

The confluence of machine learning and multiscale simulations

Multiscale modeling has a long history of use in structural biology, as computational biologists strive to overcome the time- and length-scale limits of atomistic molecular dynamics. Contemporary machine learning techniques, such as deep learning, have promoted advances in virtually every field of science and engineering and are revitalizing the traditional notions of multiscale modeling. Deep learning has found success in various approaches for distilling information from fine-scale models, such as building surrogate models and guiding the development of coarse-grained potentials. However, perhaps its most powerful use in multiscale modeling is in defining latent spaces that enable efficient exploration of conformational space. This confluence of machine learning and multiscale simulation with modern high-performance computing promises a new era of discovery and innovation in structural biology.

Understanding Virus Structure and Dynamics through Molecular Simulations

Viral outbreaks remain a serious threat to human and animal populations and motivate the continued development of antiviral drugs and vaccines, which in turn benefits from a detailed understanding of both viral structure and dynamics. While great strides have been made in characterizing these systems experimentally, molecular simulations have proven to be an essential, complementary approach. In this work, we review the contributions of molecular simulations to the understanding of viral structure, functional dynamics, and processes related to the viral life cycle. Approaches ranging from coarse-grained to all-atom representations are discussed, including current efforts at modeling complete viral systems. Overall, this review demonstrates that computational virology plays an essential role in understanding these systems.

Machine-learning assisted scheduling optimization and its application in quantum chemical calculations

Easy and effective usage of computational resources is crucial for scientific calculations, both from the perspectives of timeliness and economic efficiency. This work proposes a bi-level optimization framework to optimize the computational sequences. Machine-learning (ML) assisted static load-balancing, and different dynamic load-balancing algorithms can be integrated. Consequently, the computational and scheduling engine of the ParaEngine is developed to invoke optimized quantum chemical (QC) calculations. Illustrated benchmark calculations include high-throughput drug suit, solvent model, P38 protein, and SARS-CoV-2 systems. The results show that the usage rate of given computational resources for high throughput and large-scale fragmentation QC calculations can primarily profit, and faster accomplishing computational tasks can be expected when employing high-performance computing (HPC) clusters.

Probing Accuracy-Speedup Tradeoff in Machine Learning Surrogates for Molecular Dynamics Simulations

The performance promise of machine learning surrogates of molecular dynamics simulations of soft materials is significant but generally comes at the cost of acquiring large training datasets to learn the complex relationships between input soft material attributes and output properties. Under the constraint of limited high-performance computing resources, optimizing the size of the training datasets becomes paramount. Using an artificial neural network based surrogate for molecular dynamics simulations of confined electrolytes, we explore the tradeoff between surrogate accuracy and computational gains. Accuracy is assessed by computing the root-mean-square errors between the surrogate predictions and the ground truth results obtained via molecular dynamics simulations. The computational performance is judged by evaluating the speedup which incorporates the training dataset creation time. Improvement in accuracy occurs with a loss of speedup, which scales as the inverse of the training dataset size. The link between surrogate generalizability and the accuracy-speedup tradeoff is assessed by examining the errors incurred in surrogate predictions on unseen, interpolated input variables and developing a net speedup metric to capture the associated gains.

SARS-CoV-2 evolved variants optimize binding to cellular glycocalyx

Viral variants of concern continue to arise for SARS-CoV-2, potentially impacting both methods for detection and mechanisms of action. Here, we investigate the effect of an evolving spike positive charge in SARS-CoV-2 variants and subsequent interactions with heparan sulfate and the angiotensin converting enzyme 2 (ACE2) in the glycocalyx. We show that the positively charged Omicron variant evolved enhanced binding rates to the negatively charged glycocalyx. Moreover, we discover that while the Omicron spike-ACE2 affinity is comparable to that of the Delta variant, the Omicron spike interactions with heparan sulfate are significantly enhanced, giving rise to a ternary complex of spike-heparan sulfate-ACE2 with a large proportion of double-bound and triple-bound ACE2. Our findings suggest that SARS-CoV-2 variants evolve to be more dependent on heparan sulfate in viral attachment and infection. This discovery enables us to engineer a second-generation lateral-flow test strip that harnesses both heparin and ACE2 to reliably detect all variants of concern, including Omicron.

Mesoscale simulations: An indispensable approach to understand biomembranes

Computer simulation techniques form a versatile tool, a computational microscope, for exploring biological processes. This tool has been particularly effective in exploring different features of biological membranes. In recent years, thanks to elegant multiscale simulation schemes, some fundamental limitations of investigations by distinct simulation techniques have been resolved. As a result, we are now capable of exploring processes spanning multiple scales beyond the capacity of any single technique. In this perspective, we argue that mesoscale simulations require more attention and must be further developed to fill evident gaps in a quest toward simulating and modeling living cell membranes.

Molecular architecture and dynamics of SARS-CoV-2 envelope by integrative modeling

Despite tremendous efforts, the exact structure of SARS-CoV-2 and related betacoronaviruses remains elusive. SARS-CoV-2 envelope is a key structural component of the virion that encapsulates viral RNA. It is composed of three structural proteins, spike, membrane (M), and envelope, which interact with each other and with the lipids acquired from the host membranes. Here, we developed and applied an integrative multi-scale computational approach to model the envelope structure of SARS-CoV-2 with near atomistic detail, focusing on studying the dynamic nature and molecular interactions of its most abundant, but largely understudied, M protein. The molecular dynamics simulations allowed us to test the envelope stability under different configurations and revealed that the M dimers agglomerated into large, filament-like, macromolecular assemblies with distinct molecular patterns. These results are in good agreement with current experimental data, demonstrating a generic and versatile approach to model the structure of a virus de novo.

Fostering discoveries in the era of exascale computing: How the next generation of supercomputers empowers computational and experimental biophysics alike

Over a century ago, physicists started broadly relying on theoretical models to guide new experiments. Soon thereafter, chemists began doing the same. Now, biological research enters a new era when experiment and theory walk hand in hand. Novel software and specialized hardware became essential to understand experimental data and propose new models. In fact, current petascale computing resources already allow researchers to reach unprecedented levels of simulation throughput to connect in silico and in vitro experiments. The reduction in cost and improved access allowed a large number of research groups to adopt supercomputing resources and techniques. Here, we outline how large-scale computing has evolved to expand decades-old research, spark new research efforts, and continuously connect simulation and observation. For instance, multiple publicly and privately funded groups have dedicated extensive resources to develop artificial intelligence tools for computational biophysics, from accelerating quantum chemistry calculations to proposing protein structure models. Moreover, advances in computer hardware have accelerated data processing from single-molecule experimental observations and simulations of chemical reactions occurring throughout entire cells. The combination of software and hardware has opened the way for exascale computing and the production of the first public exascale supercomputer, Frontier, inaugurated by the Oak Ridge National Laboratory in 2022. Ultimately, the popularization and development of computational techniques and the training of researchers to use them will only accelerate the diversification of tools and learning resources for future generations.

Use of simplified models for theoretical prediction of the interactions between available antibodies and the receptor-binding domain of SARS-CoV-2 spike protein

The negative impact of infectious diseases like COVID-19 on public health and the global economy is evident. This pandemic represents a significant challenge for the scientific community to develop new practical analytical methods for accurately diagnosing emerging cases. Due to their selectivity and sensitivity, new methodologies based on antigen/antibody interactions to detect COVID-19 biomarkers are necessary. In this context, the theoretical, computational modeling reduces experimental efforts and saves resources for rational biosensor design. This study proposes using molecular dynamics to predict the interactions between the Receptor Binding Domain (RBD) of the SARS-CoV-2 spike protein simplified model and a set of highly characterized antibodies. The binding free energy of the antigen/antibody complexes was calculated for the simplified models and compared against the complete SARS-CoV-2 ectodomain to validate the methodology. The structural data derived from our molecular dynamics and end-point free energy calculations showed a positive correlation between both approximations, with a 0.82 Pearson correlation coefficient; t = 3.661, df = 3, p-value = 0.03522, with a 95% confident interval. Furthermore, we identified the interfacial residues that could generate covalent bonds with a specific chemical surface without perturbing the binding dynamics to develop highly sensitive and specific diagnostic devices. Communicated by Ramaswamy H. Sarma.

Close, but not too close: a mesoscopic description of (a)symmetry and membrane shaping mechanisms

Biomembranes are fundamental to our understanding of the cell, the basic building block of all life. An intriguing aspect of membranes is their ability to assume a variety of shapes, which is crucial for cell function. Here, we review various membrane shaping mechanisms with special focus on the current understanding of how local curvature and local rigidity induced by membrane proteins leads to emerging forces and consequently large-scale membrane deformations. We also argue that describing the interaction of rigid proteins with membranes purely in terms of local membrane curvature is incomplete and that changes in the membrane rigidity moduli must also be considered.

Binding and Unbinding Pathways of Peptide Substrates on the SARS-CoV-2 3CL Protease

Based on many crystal structures of ligand complexes, much study has been devoted to understanding the molecular recognition of SARS-CoV-2 3C-like protease (3CL^pro), a potent drug target for COVID-19. In this research, to extend this present static view, we examined the kinetic process of binding/unbinding of an eight-residue substrate peptide to/from 3CL^pro by evaluating the path ensemble with the weighted ensemble simulation. The path ensemble showed the mechanism of how a highly flexible peptide folded into the bound form. At the early stage, the dominant motion was the diffusion on the protein surface showing a broad distribution, whose center was led into the cleft of the chymotrypsin fold. We observed a definite sequential formation of the hydrogen bonds at the later stage occurring in the cleft, initiated between Glu166 (3CL^pro) and P3_Val (peptide), followed by binding to the oxyanion hole and completed by the sequence-specific recognition at P1_Gln.

Predictive Molecular Models for Charged Materials Systems: From Energy Materials to Biomacromolecules

Electrostatic interactions play a dominant role in charged materials systems. Understanding the complex correlation between macroscopic properties with microscopic structures is of critical importance to develop rational design strategies for advanced materials. But the complexity of this challenging task is augmented by interfaces present in the charged materials systems, such as electrode-electrolyte interfaces or biological membranes. Over the last decades, predictive molecular simulations that are founded in fundamental physics and optimized for charged interfacial systems have proven their value in providing molecular understanding of physicochemical properties and functional mechanisms for diverse materials. Novel design strategies utilizing predictive models have been suggested as promising route for the rational design of materials with tailored properties. Here, an overview of recent advances in the understanding of charged interfacial systems aided by predictive molecular simulations is presented. Focusing on three types of charged interfaces found in energy materials and biomacromolecules, how the molecular models characterize ion structure, charge transport, morphology relation to the environment, and the thermodynamics/kinetics of molecular binding at the interfaces is discussed. The critical analysis brings two prominent field of energy materials and biological science under common perspective, to stimulate crossover in both research field that have been largely separated.

#COVIDisAirborne: AI-enabled multiscale computational microscopy of delta SARS-CoV-2 in a respiratory aerosol

We seek to completely revise current models of airborne transmission of respiratory viruses by providing never-before-seen atomic-level views of the SARS-CoV-2 virus within a respiratory aerosol. Our work dramatically extends the capabilities of multiscale computational microscopy to address the significant gaps that exist in current experimental methods, which are limited in their ability to interrogate aerosols at the atomic/molecular level and thus obscure our understanding of airborne transmission. We demonstrate how our integrated data-driven platform provides a new way of exploring the composition, structure, and dynamics of aerosols and aerosolized viruses, while driving simulation method development along several important axes. We present a series of initial scientific discoveries for the SARS-CoV-2 Delta variant, noting that the full scientific impact of this work has yet to be realized.

Breathing and Tilting: Mesoscale Simulations Illuminate Influenza Glycoprotein Vulnerabilities

Influenza virus has resurfaced recently from inactivity during the early stages of the COVID-19 pandemic, raising serious concerns about the nature and magnitude of future epidemics. The main antigenic targets of influenza virus are two surface glycoproteins, hemagglutinin (HA) and neuraminidase (NA). Whereas the structural and dynamical properties of both glycoproteins have been studied previously, the understanding of their plasticity in the whole-virion context is fragmented. Here, we investigate the dynamics of influenza glycoproteins in a crowded protein environment through mesoscale all-atom molecular dynamics simulations of two evolutionary-linked glycosylated influenza A whole-virion models. Our simulations reveal and kinetically characterize three main molecular motions of influenza glycoproteins: NA head tilting, HA ectodomain tilting, and HA head breathing. The flexibility of HA and NA highlights antigenically relevant conformational states, as well as facilitates the characterization of a novel monoclonal antibody, derived from convalescent human donor, that binds to the underside of the NA head. Our work provides previously unappreciated views on the dynamics of HA and NA, advancing the understanding of their interplay and suggesting possible strategies for the design of future vaccines and antivirals against influenza.

Role of bioaerosol in virus transmission and material-based countermeasures

Respiratory pathogens transmit primarily through particles such as droplets and aerosols. Although often overlooked, the resuspension of settled droplets is also a key facilitator of disease transmission. In this review, we discuss the three main mechanisms of aerosol generation: direct generation such as coughing and sneezing, indirect generation such as medical procedures, and resuspension of settled droplets and aerosols. The size of particles and environmental factors influence their airborne lifetime and ability to cause infection. Specifically, humidity and temperature are key factors controlling the evaporation of suspended droplets, consequently affecting the duration in which particles remain airborne. We also suggest material-based approaches for effective prevention of disease transmission. These approaches include electrostatically charged virucidal agents and surface coatings, which have been shown to be highly effective in deactivating and reducing resuspension of pathogen-laden aerosols.

The dawn of interoperating spatial models in cell biology

Spatial simulations are becoming an increasingly ubiquitous component in the cycle of discovery, experimentation, and communication across the sciences. In cell biology, many researchers share a vision of developing multiscale models that recapitulate observable behaviors spanning from atoms to cells to tissues. For this dream to become a reality, however, simulation technologies must provide a means for integration and interoperability as they advance. Already, the field has developed numerous methods that span scales of length, time, and complexity to create an extensive body of effective simulation approaches, and although these approaches rarely interoperate, they collectively cover a large spectrum of knowledge that future models may handle in a more unified manner. Here, we discuss the importance of making the data, workflows, and outputs of spatial simulations shareable and interoperable; and how democratization could encourage diverse biologists to participate more easily in developing models to advance our understanding of biological systems.

Thirty years of molecular dynamics simulations on posttranslational modifications of proteins

Posttranslational modifications (PTMs) are an integral component to how cells respond to perturbation. While experimental advances have enabled improved PTM identification capabilities, the same throughput for characterizing how structural changes caused by PTMs equate to altered physiological function has not been maintained. In this Perspective, we cover the history of computational modeling and molecular dynamics simulations which have characterized the structural implications of PTMs. We distinguish results from different molecular dynamics studies based upon the timescales simulated and analysis approaches used for PTM characterization. Lastly, we offer insights into how opportunities for modern research efforts on in silico PTM characterization may proceed given current state-of-the-art computing capabilities and methodological advancements.

Harris S, Ramanathan A. Intelligent resolution: Integrating Cryo-EM with AI-driven multi-resolution simulations to observe the severe acute respiratory syndrome coronavirus-2 replication-transcription machinery in action

The severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) replication transcription complex (RTC) is a multi-domain protein responsible for replicating and transcribing the viral mRNA inside a human cell. Attacking RTC function with pharmaceutical compounds is a pathway to treating COVID-19. Conventional tools, e.g., cryo-electron microscopy and all-atom molecular dynamics (AAMD), do not provide sufficiently high resolution or timescale to capture important dynamics of this molecular machine. Consequently, we develop an innovative workflow that bridges the gap between these resolutions, using mesoscale fluctuating finite element analysis (FFEA) continuum simulations and a hierarchy of AI-methods that continually learn and infer features for maintaining consistency between AAMD and FFEA simulations. We leverage a multi-site distributed workflow manager to orchestrate AI, FFEA, and AAMD jobs, providing optimal resource utilization across HPC centers. Our study provides unprecedented access to study the SARS-CoV-2 RTC machinery, while providing general capability for AI-enabled multi-resolution simulations at scale.

AlphaDrug: protein target specific de novo molecular generation

Traditional drug discovery is very laborious, expensive, and time-consuming, due to the huge combinatorial complexity of the discrete molecular search space. Researchers have turned to machine learning methods for help to tackle this difficult problem. However, most existing methods are either virtual screening on the available database of compounds by protein-ligand affinity prediction, or unconditional molecular generation, which does not take into account the information of the protein target. In this paper, we propose a protein target-oriented de novo drug design method, called AlphaDrug. Our method is able to automatically generate molecular drug candidates in an autoregressive way, and the drug candidates can dock into the given target protein well. To fulfill this goal, we devise a modified transformer network for the joint embedding of protein target and the molecule, and a Monte Carlo tree search (MCTS) algorithm for the conditional molecular generation. In the transformer variant, we impose a hierarchy of skip connections from protein encoder to molecule decoder for efficient feature transfer. The transformer variant computes the probabilities of next atoms based on the protein target and the molecule intermediate. We use the probabilities to guide the look-ahead search by MCTS to enhance or correct the next-atom selection. Moreover, MCTS is also guided by a value function implemented by a docking program, such that the paths with many low docking values are seldom chosen. Experiments on diverse protein targets demonstrate the effectiveness of our methods, indicating that AlphaDrug is a potentially promising solution to target-specific de novo drug design.

Deep learning-based molecular dynamics simulation for structure-based drug design against SARS-CoV-2

Coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2), has led to a global pandemic. Deep learning (DL) technology and molecular dynamics (MD) simulation are two mainstream computational approaches to investigate the geometric, chemical and structural features of protein and guide the relevant drug design. Despite a large amount of research papers focusing on drug design for SARS-COV-2 using DL architectures, it remains unclear how the binding energy of the protein-protein/ligand complex dynamically evolves which is also vital for drug development. In addition, traditional deep neural networks usually have obvious deficiencies in predicting the interaction sites as protein conformation changes. In this review, we introduce the latest progresses of the DL and DL-based MD simulation approaches in structure-based drug design (SBDD) for SARS-CoV-2 which could address the problems of protein structure and binding prediction, drug virtual screening, molecular docking and complex evolution. Furthermore, the current challenges and future directions of DL-based MD simulation for SBDD are also discussed.

Probing the mutational landscape of the SARS-CoV-2 spike protein via quantum mechanical modeling of crystallographic structures

We employ a recently developed complexity-reduction quantum mechanical (QM-CR) approach, based on complexity reduction of density functional theory calculations, to characterize the interactions of the SARS-CoV-2 spike receptor binding domain (RBD) with ACE2 host receptors and antibodies. QM-CR operates via ab initio identification of individual amino acid residue's contributions to chemical binding and leads to the identification of the impact of point mutations. Here, we especially focus on the E484K mutation of the viral spike protein. We find that spike residue 484 hinders the spike's binding to the human ACE2 receptor (hACE2). In contrast, the same residue is beneficial in binding to the bat receptor Rhinolophus macrotis ACE2 (macACE2). In agreement with empirical evidence, QM-CR shows that the E484K mutation allows the spike to evade categories of neutralizing antibodies like C121 and C144. The simulation also shows how the Delta variant spike binds more strongly to hACE2 compared to the original Wuhan strain, and predicts that a E484K mutation can further improve its binding. Broad agreement between the QM-CR predictions and experimental evidence supports the notion that ab initio modeling has now reached the maturity required to handle large intermolecular interactions central to biological processes.

Breathing and tilting: mesoscale simulations illuminate influenza glycoprotein vulnerabilities

Influenza virus has resurfaced recently from inactivity during the early stages of the COVID-19 pandemic, raising serious concerns about the nature and magnitude of future epidemics. The main antigenic targets of influenza virus are two surface glycoproteins, hemagglutinin (HA) and neuraminidase (NA). Whereas the structural and dynamical properties of both glycoproteins have been studied previously, the understanding of their plasticity in the whole-virion context is fragmented. Here, we investigate the dynamics of influenza glycoproteins in a crowded protein environment through mesoscale all-atom molecular dynamics simulations of two evolutionary-linked glycosylated influenza A whole-virion models. Our simulations reveal and kinetically characterize three main molecular motions of influenza glycoproteins: NA head tilting, HA ectodomain tilting, and HA head breathing. The flexibility of HA and NA highlights antigenically relevant conformational states, as well as facilitates the characterization of a novel monoclonal antibody, derived from human convalescent plasma, that binds to the underside of the NA head. Our work provides previously unappreciated views on the dynamics of HA and NA, advancing the understanding of their interplay and suggesting possible strategies for the design of future vaccines and antivirals against influenza.

Accurate Molecular-Orbital-Based Machine Learning Energies via Unsupervised Clustering of Chemical Space

We introduce an unsupervised clustering algorithm to improve training efficiency and accuracy in predicting energies using molecular-orbital-based machine learning (MOB-ML). This work determines clusters via the Gaussian mixture model (GMM) in an entirely automatic manner and simplifies an earlier supervised clustering approach [ J. Chem. Theory Comput. 2019, 15, 6668] by eliminating both the necessity for user-specified parameters and the training of an additional classifier. Unsupervised clustering results from GMM have the advantages of accurately reproducing chemically intuitive groupings of frontier molecular orbitals and exhibiting improved performance with an increasing number of training examples. The resulting clusters from supervised or unsupervised clustering are further combined with scalable Gaussian process regression (GPR) or linear regression (LR) to learn molecular energies accurately by generating a local regression model in each cluster. Among all four combinations of regressors and clustering methods, GMM combined with scalable exact GPR (GMM/GPR) is the most efficient training protocol for MOB-ML. The numerical tests of molecular energy learning on thermalized data sets of drug-like molecules demonstrate the improved accuracy, transferability, and learning efficiency of GMM/GPR over other training protocols for MOB-ML, i.e., supervised regression clustering combined with GPR (RC/GPR) and GPR without clustering. GMM/GPR also provides the best molecular energy predictions compared with ones from the literature on the same benchmark data sets. With a lower scaling, GMM/GPR has a 10.4-fold speedup in wall-clock training time compared with scalable exact GPR with a training size of 6500 QM7b-T molecules.

Methodology-Centered Review of Molecular Modeling, Simulation, and Prediction of SARS-CoV-2

Despite tremendous efforts in the past two years, our understanding of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), virus-host interactions, immune response, virulence, transmission, and evolution is still very limited. This limitation calls for further in-depth investigation. Computational studies have become an indispensable component in combating coronavirus disease 2019 (COVID-19) due to their low cost, their efficiency, and the fact that they are free from safety and ethical constraints. Additionally, the mechanism that governs the global evolution and transmission of SARS-CoV-2 cannot be revealed from individual experiments and was discovered by integrating genotyping of massive viral sequences, biophysical modeling of protein-protein interactions, deep mutational data, deep learning, and advanced mathematics. There exists a tsunami of literature on the molecular modeling, simulations, and predictions of SARS-CoV-2 and related developments of drugs, vaccines, antibodies, and diagnostics. To provide readers with a quick update about this literature, we present a comprehensive and systematic methodology-centered review. Aspects such as molecular biophysics, bioinformatics, cheminformatics, machine learning, and mathematics are discussed. This review will be beneficial to researchers who are looking for ways to contribute to SARS-CoV-2 studies and those who are interested in the status of the field.

Quercetin and luteolin are single-digit micromolar inhibitors of the SARS-CoV-2 RNA-dependent RNA polymerase

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has rapidly become a global health pandemic. Among the viral proteins, RNA-dependent RNA polymerase (RdRp) is responsible for viral genome replication and has emerged as one of the most promising targets for pharmacological intervention against SARS-CoV-2. To this end, we experimentally tested luteolin and quercetin for their ability to inhibit the RdRp enzyme. These two compounds are ancestors of flavonoid natural compounds known for a variety of basal pharmacological activities. Luteolin and quercetin returned a single-digit IC₅₀ of 4.6 µM and 6.9 µM, respectively. Then, through dynamic docking simulations, we identified possible binding modes of these compounds to a recently published cryo-EM structure of RdRp. Collectively, these data indicate that these two compounds are a valid starting point for further optimization and development of a new class of RdRp inhibitors to treat SARS-CoV-2 and potentially other viral infections.

Accelerators for Classical Molecular Dynamics Simulations of Biomolecules

Atomistic Molecular Dynamics (MD) simulations provide researchers the ability to model biomolecular structures such as proteins and their interactions with drug-like small molecules with greater spatiotemporal resolution than is otherwise possible using experimental methods. MD simulations are notoriously expensive computational endeavors that have traditionally required massive investment in specialized hardware to access biologically relevant spatiotemporal scales. Our goal is to summarize the fundamental algorithms that are employed in the literature to then highlight the challenges that have affected accelerator implementations in practice. We consider three broad categories of accelerators: Graphics Processing Units (GPUs), Field-Programmable Gate Arrays (FPGAs), and Application Specific Integrated Circuits (ASICs). These categories are comparatively studied to facilitate discussion of their relative trade-offs and to gain context for the current state of the art. We conclude by providing insights into the potential of emerging hardware platforms and algorithms for MD.

Pre-exascale HPC approaches for molecular dynamics simulations. Covid-19 research: A use case

Exascale computing has been a dream for ages and is close to becoming a reality that will impact how molecular simulations are being performed, as well as the quantity and quality of the information derived for them. We review how the biomolecular simulations field is anticipating these new architectures, making emphasis on recent work from groups in the BioExcel Center of Excellence for High Performance Computing. We exemplified the power of these simulation strategies with the work done by the HPC simulation community to fight Covid-19 pandemics. This article is categorized under:Data Science > Computer Algorithms and ProgrammingData Science > Databases and Expert SystemsMolecular and Statistical Mechanics > Molecular Dynamics and Monte-Carlo Methods.

Adsorption of Pulmonary and Exogeneous Surfactants on SARS-CoV-2 Spike Protein

COVID-19 is transmitted by inhaling SARS-CoV-2 virions, which are enveloped by a lipid bilayer decorated by a "crown" of Spike protein protrusions. In the respiratory tract, virions interact with surfactant films composed of phospholipids and cholesterol that coat lung airways. Here, we explore by using coarse-grained molecular dynamics simulations the physico-chemical mechanisms of surfactant adsorption on Spike proteins. With examples of zwitterionic dipalmitoyl phosphatidyl choline, cholesterol, and anionic sodium dodecyl sulphate, we show that surfactants form micellar aggregates that selectively adhere to the specific regions of S1 domain of the Spike protein that are responsible for binding with ACE2 receptors and virus transmission into the cells. We find high cholesterol adsorption and preferential affinity of anionic surfactants to Arginine and Lysine residues within S1 receptor binding motif. These findings have important implications for informing the search for extraneous therapeutic surfactants for curing and preventing COVID-19 by SARS-CoV-2 and its variants.

Cooperative multivalent receptor binding promotes exposure of the SARS-CoV-2 fusion machinery core

The molecular events that permit the spike glycoprotein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) to bind and enter cells are important to understand for both fundamental and therapeutic reasons. Spike proteins consist of S1 and S2 domains, which recognize angiotensin-converting enzyme 2 (ACE2) receptors and contain the viral fusion machinery, respectively. Ostensibly, the binding of spike trimers to ACE2 receptors promotes dissociation of the S1 domains and exposure of the fusion machinery, although the molecular details of this process have yet to be observed. We report the development of bottom-up coarse-grained (CG) models consistent with cryo-electron tomography data, and the use of CG molecular dynamics simulations to investigate viral binding and S2 core exposure. We show that spike trimers cooperatively bind to multiple ACE2 dimers at virion-cell interfaces in a manner distinct from binding between soluble proteins, which processively induces S1 dissociation. We also simulate possible variant behavior using perturbed CG models, and find that ACE2-induced S1 dissociation is primarily sensitive to conformational state populations and the extent of S1/S2 cleavage, rather than ACE2 binding affinity. These simulations reveal an important concerted interaction between spike trimers and ACE2 dimers that primes the virus for membrane fusion and entry.

Assembly and Analysis of Cell-Scale Membrane Envelopes

The march toward exascale computing will enable routine molecular simulation of larger and more complex systems, for example, simulation of entire viral particles, on the scale of approximately billions of atoms─a simulation size commensurate with a small bacterial cell. Anticipating the future hardware capabilities that will enable this type of research and paralleling advances in experimental structural biology, efforts are currently underway to develop software tools, procedures, and workflows for constructing cell-scale structures. Herein, we describe our efforts in developing and implementing an efficient and robust workflow for construction of cell-scale membrane envelopes and embedding membrane proteins into them. A new approach for construction of massive membrane structures that are stable during the simulations is built on implementing a subtractive assembly technique coupled with the development of a structure concatenation tool (fastmerge), which eliminates overlapping elements based on volumetric criteria rather than adding successive molecules to the simulation system. Using this approach, we have constructed two "protocells" consisting of MARTINI coarse-grained beads to represent cellular membranes, one the size of a cellular organelle and another the size of a small bacterial cell. The membrane envelopes constructed here remain whole during the molecular dynamics simulations performed and exhibit water flux only through specific proteins, demonstrating the success of our methodology in creating tight cell-like membrane compartments. Extended simulations of these cell-scale structures highlight the propensity for nonspecific interactions between adjacent membrane proteins leading to the formation of protein microclusters on the cell surface, an insight uniquely enabled by the scale of the simulations. We anticipate that the experiences and best practices presented here will form the basis for the next generation of cell-scale models, which will begin to address the addition of soluble proteins, nucleic acids, and small molecules essential to the function of a cell.

WESTPA 2.0: High-Performance Upgrades for Weighted Ensemble Simulations and Analysis of Longer-Timescale Applications

The weighted ensemble (WE) family of methods is one of several statistical mechanics-based path sampling strategies that can provide estimates of key observables (rate constants and pathways) using a fraction of the time required by direct simulation methods such as molecular dynamics or discrete-state stochastic algorithms. WE methods oversee numerous parallel trajectories using intermittent overhead operations at fixed time intervals, enabling facile interoperability with any dynamics engine. Here, we report on the major upgrades to the WESTPA software package, an open-source, high-performance framework that implements both basic and recently developed WE methods. These upgrades offer substantial improvements over traditional WE methods. The key features of the new WESTPA 2.0 software enhance the efficiency and ease of use: an adaptive binning scheme for more efficient surmounting of large free energy barriers, streamlined handling of large simulation data sets, exponentially improved analysis of kinetics, and developer-friendly tools for creating new WE methods, including a Python API and resampler module for implementing both binned and "binless" WE strategies.

Challenges and frontiers of computational modelling of biomolecular recognition

Biomolecular recognition including binding of small molecules, peptides and proteins to their target receptors plays a key role in cellular function and has been targeted for therapeutic drug design. However, the high flexibility of biomolecules and slow binding and dissociation processes have presented challenges for computational modelling. Here, we review the challenges and computational approaches developed to characterise biomolecular binding, including molecular docking, molecular dynamics simulations (especially enhanced sampling) and machine learning. Further improvements are still needed in order to accurately and efficiently characterise binding structures, mechanisms, thermodynamics and kinetics of biomolecules in the future.