Search NASA⌕ Search

SEARCH · Search NASA

Results for “dynamic sampling”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,765 records · Page 98

Ultrashort XUV pulse absorption spectroscopy of partially oxidized cobalt nanoparticles

High-order harmonic generation (HHG) based transient extreme ultraviolet (XUV) absorption spectroscopy is an emerging technique to trace photoinduced charge carrier dynamics in condensed phase materials with femtosecond and even attosecond temporal resolution and elemental specificity. However, its application to nanoparticulate samples that are relevant, for example, for novel photocatalytic light harvesting concepts, has been limited. This is in part due to the challenge to produce residual-free samples on ultrathin, XUV-transparent substrates as well as a widespread understanding that sparsely distributed nanoparticles do not provide sufficient contrast for XUV absorption measurements. Here, we present static XUV absorption spectra of partially oxidized Co nanowire-structures with diameters of approximately 4.5 nm and lengths between 10 and 40 nm, recorded with an ultrashort pulse HHG light source. Nanoparticles are synthesized by the agglomeration of Co atoms inside superfluid helium droplets, followed by surface deposition and oxidation in ambient air. The method is uniquely suited for residual-free synthesis of transition metal nanowires and their deposition on ultrathin substrates. Analysis by high-resolution transmission electron microscopy reveals the formation of CoO nanowires with regions of unoxidized Co in their interior. The nanoparticle samples are investigated in an HHG-driven ultrafast XUV absorption setup. Despite the low surface coverage of only 23%, the recorded spectrum exhibits a distinct absorption feature at the Co M 2,3 (2p) edge near 60 eV with a peak height of about 40 mOD. Finally, the results support the feasibility of table-top ultrafast transient XUV absorption studies of photoinduced dynamics in transition metal oxide nanoparticles with sub-monolayer surface coverage.

74 ATOMIC AND MOLECULAR PHYSICS↗

Molecular structure of vapor-deposited amorphous selenium

The structure of amorphous selenium is clouded with much uncertainty and contradictory results regarding the dominance of polymeric chains versus monomer rings. The analysis of the diffraction radial distribution functions are inconclusive because of the similarities between the crystalline allotropes of selenium in terms of the coordination number, bond length, bond angle, and dihedral angle. Here, we took a much different approach and probed the molecular symmetry of the thermodynamically unstable amorphous state via analysis of structural phase transformations. We verified the structure of the converted metastable and stable crystalline structures using scanning transmission electron microscopy. In addition, given that no experimental technique can tell us the exact three-dimensional atomic arrangements in glassy semiconductors, we performed molecular-dynamic simulations using a well-established empirical three-body interatomic potential. In this work, we developed a true vapor-deposited process for the deposition of selenium molecules onto a substrate using empirical molecular vapor compositions and densities. We prepared both vapor-deposited and melt-quenched samples and showed that the simulated radial distribution functions match very well to experiment. The combination of our experimental and molecular-dynamic analyses shows that the structures of vapor- and melt-quenched glassy/amorphous selenium are quite different, based primarily on rings and chains, respectively, reflecting the predominant structure of the parent phase in its thermodynamic equilibrium.

36 MATERIALS SCIENCE↗

Data Analysis Approaches for the Risk-Informed Safety Margins Characterization Toolkit

In the past decades, several numerical simulation codes have been employed to simulate accident dynamics (e.g., RELAP5-3D, RELAP-7, MELCOR, MAAP). In order to evaluate the impact of uncertainties into accident dynamics, several stochastic methodologies have been coupled with these codes. These stochastic methods range from classical Monte-Carlo and Latin Hypercube sampling to stochastic polynomial methods. Similar approaches have been introduced into the risk and safety community where stochastic methods (such as RAVEN, ADAPT, MCDET, ADS) have been coupled with safety analysis codes in order to evaluate the safety impact of timing and sequencing of events. These approaches are usually called Dynamic PRA or simulation-based PRA methods. These uncertainties and safety methods usually generate a large number of simulation runs (database storage may be on the order of gigabytes or higher). The scope of this paper is to present a broad overview of methods and algorithms that can be used to analyze and extract information from large data sets containing time dependent data. In this context, “extracting information” means constructing input-output correlations, finding commonalities, and identifying outliers. Some of the algorithms presented here have been developed or are under development within the RAVEN statistical framework.

97 MATHEMATICS AND COMPUTING↗

PROJECTED CENTRAL DARK MATTER FRACTIONS AND DENSITIES IN MASSIVE EARLY-TYPE GALAXIES FROM THE SLOAN DIGITAL SKY SURVEY

We investigate in massive early-type galaxies the variation of their two-dimensional central fraction of dark over total mass and dark matter density as a function of stellar mass, central stellar velocity dispersion, effective radius, and central surface stellar mass density. We use a sample of approximately 1.7 x 10{sup 5} galaxies from the Sloan Digital Sky Survey Data Release Seven (SDSS DR7) at redshift smaller than 0.33. We apply conservative photometric and spectroscopic cuts on the SDSS DR7 and the MPA/JHU value-added galaxy catalogs, to select galaxies with physical properties similar to those of the lenses studied in the Sloan Lens ACS Survey. The values of the galaxy stellar and total mass projected inside a cylinder of radius equal to the effective radius are obtained, respectively, by fitting the SDSS multicolor photometry with stellar population synthesis models, under the assumption of a Chabrier stellar initial mass function (IMF), and adopting a one-component isothermal total mass model with effective velocity dispersion approximated by the central stellar velocity dispersion. The plausibility of an isothermal model to represent the galaxy total mass distribution is supported by independent gravitational lensing and stellar-dynamical analyses performed in the lens subsample, which is found here to represent nicely the entire galaxy sample. We find that within the effective radius the stellar mass estimates differ from the total ones by only a relatively constant proportionality factor. In detail, we observe that the values of the projected fraction of dark over total mass and the logarithmic values of the central surface dark matter density (measured in M{sub sun} kpc{sup -2}) have almost Gaussian probability distribution functions, with median values of 0.64{sup +0.08}{sub -0.11} and 9.1{sup +0.2}{sub -0.2}, respectively. We discuss the observed correlations between these quantities and other galaxy global parameters and show that our results disfavor an interpretation of the tilt of the fundamental plane in terms of differences in the galaxy dark matter content and give useful information on the possible variations of the galaxy stellar IMF and dark matter density profile. Finally, we provide some observational evidence on the likely significant contribution of dry minor mergers, feedback from active galactic nuclei, and/or coalescence of binary black holes on the formation and evolution of massive early-type galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Human Gut Microbiota from Autism Spectrum Disorder Promote Behavioral Symptoms in Mice

Large-scale phosphoproteomics with coverage of over 10,000 sites of phosphorylation have now been routinely achieved with advanced mass spectrometry (MS)-based workflows. However, accurate targeted MS-based quantification of phosphorylation dynamics, an important direction for gaining quantitative understanding of signaling pathways or networks, has been much less investigated. Herein, we report an assessment of the targeted workflow in the context of signal transduction pathways, using the epidermal growth factor receptor (EGFR)–mitogen-activated protein kinase (MAPK) pathway as our model. 43 phosphopeptides from the EGFR–MAPK pathway were selected for the study. The recovery and sensitivity of a workflow consisted of two commonly used enrichment methods, immobilized metal affinity chromatography (IMAC) and titanium oxide (TiO2), combined with selected reaction monitoring (SRM)-MS, were evaluated. The recovery of phosphopeptides by IMAC and TiO2 enrichment was quantified to be 38 ± 5% and 58 ± 20%, respectively, based on internal standards. Moreover, both enrichment methods provided comparable sensitivity from 1-100 g starting peptides. Robust quantification was consistently achieved for most targeted phosphopeptides when starting with 25-100 g peptides. However, the numbers of quantified targets significantly dropped when peptide samples were in the 1-25g range. Finally, IMAC-SRM was applied to quantify signaling dynamics of EGFR-MAPK pathway in Hs578T cells following 3 ng/mL EGF treatment. The kinetics of phosphorylation clearly revealed early and late phases of phosphorylation, even for very low abundance proteins. These results demonstrate the feasibility of robust targeted quantification of phosphorylation dynamics for specific pathways, even starting with relatively small amounts of protein.

Sharon, Gil↗

Targeted Quantification of Phosphorylation Dynamics in the Context of EGFR-MAPK Pathway

Large-scale phosphoproteomics with coverage of over 10,000 sites of phosphorylation have now been routinely achieved with advanced mass spectrometry (MS)-based workflows. However, accurate targeted MS-based quantification of phosphorylation dynamics, an important direction for gaining quantitative understanding of signaling pathways or networks, has been much less investigated. Herein, we report an assessment of the targeted workflow in the context of signal transduction pathways, using the epidermal growth factor receptor (EGFR)–mitogen-activated protein kinase (MAPK) pathway as our model. 43 phosphopeptides from the EGFR–MAPK pathway were selected for the study. The recovery and sensitivity of a workflow consisted of two commonly used enrichment methods, immobilized metal affinity chromatography (IMAC) and titanium oxide (TiO2), combined with selected reaction monitoring (SRM)-MS, were evaluated. The recovery of phosphopeptides by IMAC and TiO2 enrichment was quantified to be 38 ± 5% and 58 ± 20%, respectively, based on internal standards. Moreover, both enrichment methods provided comparable sensitivity from 1-100 g starting peptides. Robust quantification was consistently achieved for most targeted phosphopeptides when starting with 25-100 g peptides. However, the numbers of quantified targets significantly dropped when peptide samples were in the 1-25g range. Finally, IMAC-SRM was applied to quantify signaling dynamics of EGFR-MAPK pathway in Hs578T cells following 3 ng/mL EGF treatment. The kinetics of phosphorylation clearly revealed early and late phases of phosphorylation, even for very low abundance proteins. These results demonstrate the feasibility of robust targeted quantification of phosphorylation dynamics for specific pathways, even starting with relatively small amounts of protein.

Environmental Molecular Sciences Laboratory↗

Small Angle Neutron Scattering Studies of R67 Dihydrofolate Reductase, a Tetrameric Protein with Intrinsically Disordered N-Termini

R67 dihydrofolate reductase (DHFR) is a homotetramer with a single active site pore and no sequence or structural homology with chromosomal DHFRs. The R67 enzyme provides resistance to trimethoprim, an active site-directed inhibitor of Escherichia coli DHFR. Sixteen to twenty N-terminal amino acids are intrinsically disordered in the R67 dimer crystal structure. Chymotrypsin cleavage of 16 N-terminal residues results in an active enzyme with a decreased stability. We investigated the space sampled by the disordered N-termini of R67 DHFR using small angle neutron scattering. From a combined analysis using molecular dynamics and the program SASSIE (http://www.smallangles.net/sassie/SASSIE_HOME.html), the apoenzyme displays a radius of gyration (Rg) of 21.46 ± 0.50 Å. Addition of glycine betaine, an osmolyte, does not result in folding of the termini as the Rg increases slightly to 22.78 ± 0.87 Å. SASSIE fits of the latter SANS data indicate that the disordered N-termini sample larger regions of space and remain disordered, suggesting they might function as entropic bristles. Pressure perturbation calorimetry also indicated that the volume of R67 DHFR increases upon addition of 10% betaine and decreased at 20% betaine because of the dehydration of the protein. Studies of the hydration of full-length R67 DHFR in the presence of the osmolytes betaine and dimethyl sulfoxide find around 1250 water molecules hydrating the protein. Similar studies with truncated R67 DHFR yield around 400 water molecules hydrating the protein in the presence of betaine. The difference of ~900 waters indicates the N-termini are well-hydrated.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Pattern-enhanced Resonant Soft X-ray Scattering for Operando monitoring of electrochemical solid-liquid interfaces

Unveiling interfaces at sub-nanometer scales is essential for advancing the understanding of complex chemical transformations. However, characterizing solid-liquid interfaces with high dimensional sensitivity and temporal resolution remains challenging, due to their dynamic nature and inaccessibility by conventional probes. Here we present an approach, Pattern-enhanced Resonant Soft X-ray Scattering, to overcome the challenges. Rooted in a “sample-as-optics” philosophy, this technique utilizes precisely engineered line-grating nanopatterns to modulate near-field X-ray illumination, coherently enhancing scattering signals from the line-gratings. We implement the method using Ni line-grating nanopatterns in electrochemical water oxidation. The periodic nanostructures serve as diffractive optical elements to reveal the Ni oxidation gradients and structural dynamics at the electrode-electrolyte interfaces. Finite-element simulations corroborate the observed trends by modeling variations in compositions and structures during electrocatalysis. Through integrating advanced sample design with coherent wave nature of soft X-rays, our approach opens accessible pathways to operando exploring chemical evolution and sub-nanometer dimensional variations simultaneously in electrochemical systems. This non-destructive method is efficient and element-specific, making it valuable for probing chemical and dimensional dynamics with appropriate modeling.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Li Intercalation into Graphite: Direct Optical Imaging and Cahn–Hilliard Reaction Dynamics

Lithium intercalation into graphite is a critical process in energy storage technology. Studies of Li intercalation kinetics have proved challenging due to structural and phase complexity, and sample heterogeneity. Here we report direct time- and space-resolved, all-optical measurement of Li intercalation. We use a single crystal graphite electrode with lithographically defined disc geometry. All-optical, Raman and reflectance measurements distinguish the intrinsic intercalation process from side reactions, and provide new insight into the microscopic intercalation process. The recently proposed Cahn–Hilliard reaction (CHR) theory quantitatively captures the observed phase front spatial patterns and dynamics, using a two-layer free-energy model with novel, generalized Butler–Volmer kinetics. This approach unites Cahn–Hilliard and electrochemical kinetics, using a thermodynamically consistent description of the Li injection reaction at the crystal edge that involves a cooperative opening of graphene planes. Here, the excellent agreement between experiment and theory presented here, with single-crystal resolution, provides strong support for the CHR theory of solid-state reactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

THE SAMI GALAXY SURVEY: TOWARD A UNIFIED DYNAMICAL SCALING RELATION FOR GALAXIES OF ALL TYPES

We take advantage of the first data from the Sydney-AAO Multi-object Integral field Galaxy Survey to investigate the relation between the kinematics of gas and stars, and stellar mass in a comprehensive sample of nearby galaxies. We find that all 235 objects in our sample, regardless of their morphology, lie on a tight relation linking stellar mass (M {sub *}) to internal velocity quantified by the S {sub 0.5} parameter, which combines the contribution of both dispersion (σ) and rotational velocity (V {sub rot}) to the dynamical support of a galaxy (S{sub 0.5}=√(0.5 V{sub rot}{sup 2}+σ{sup 2})). Our results are independent of the baryonic component from which σ and V {sub rot} are estimated, as the S {sub 0.5} of stars and gas agree remarkably well. This represents a significant improvement compared to the canonical M {sub *} versus V {sub rot} and M {sub *} versus σ relations. Not only is no sample pruning necessary, but also stellar and gas kinematics can be used simultaneously, as the effect of asymmetric drift is taken into account once V {sub rot} and σ are combined. Our findings illustrate how the combination of dispersion and rotational velocities for both gas and stars can provide us with a single dynamical scaling relation valid for galaxies of all morphologies across at least the stellar mass range 8.5 <log (M {sub *}/M {sub ☉}) < 11. Such relation appears to be more general and at least as tight as any other dynamical scaling relation, representing a unique tool for investigating the link between galaxy kinematics and baryonic content, and a less biased comparison with theoretical models.

79 ASTRONOMY AND ASTROPHYSICS↗

Coalescence of silver clusters by immersion in diluted HF solution

The galvanic displacement deposition of silver on H-terminated Si (100) in the time scale of seconds is instantaneous and characterized by a cluster density of 10{sup 11}-10{sup 12} cm{sup −2}. The amount of deposited Ag follows a t{sup 1/2} dependence in agreement with a Cottrell diffusion limited mechanism. At the same time, during the deposition, the cluster density reduces by a factor 5. This behavior is in contrast with the assumption of immobile clusters. We show in the present work that coalescence and aggregation occur also in the samples immersed in the diluted hydrofluoric acid (HF) solution without the presence of Ag{sup +}. Clusters agglomerate according to a process of dynamic coalescence, typical of colloids, followed by atomic redistribution at the contact regions with the generation of multiple internal twins and stacking-faults. The normalized size distributions in terms of r/r{sub mean} follow also the prediction of the Smoluchowski ripening mechanism. No variation of the cluster density occurs for samples immersed in pure H{sub 2}O solution. The different behavior might be associated to the strong attraction of clusters to oxide-terminated Si surface in presence of water. The silver clusters are instead weakly bound to hydrophobic H-terminated Si in presence of HF. HF causes then the detachment of clusters and a random movement on the silicon surface with mobility of about 10{sup −13} cm{sup 2}/s. Attractive interaction (probably van der Waals) among particles promotes coarsening.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Π4U: A high performance computing framework for Bayesian uncertainty quantification of complex models

We present Π4U,{sup 1} an extensible framework, for non-intrusive Bayesian Uncertainty Quantification and Propagation (UQ+P) of complex and computationally demanding physical models, that can exploit massively parallel computer architectures. The framework incorporates Laplace asymptotic approximations as well as stochastic algorithms, along with distributed numerical differentiation and task-based parallelism for heterogeneous clusters. Sampling is based on the Transitional Markov Chain Monte Carlo (TMCMC) algorithm and its variants. The optimization tasks associated with the asymptotic approximations are treated via the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). A modified subset simulation method is used for posterior reliability measurements of rare events. The framework accommodates scheduling of multiple physical model evaluations based on an adaptive load balancing library and shows excellent scalability. In addition to the software framework, we also provide guidelines as to the applicability and efficiency of Bayesian tools when applied to computationally demanding physical models. Theoretical and computational developments are demonstrated with applications drawn from molecular dynamics, structural dynamics and granular flow.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Impact of Soret effect on hydrogen and helium retention in PFC tungsten under ELM-like conditions

Abstract In our previous work, we have demonstrated using nonequilibrium molecular-dynamics simulations that the fluxes of helium and self-interstitial atoms in the presence of a thermal gradient in tungsten are directed opposite to the heat flux, indicating that species transport is governed by a Soret effect, namely, thermal-gradient-driven diffusion, characterized by a negative heat of transport that drives species transport uphill, i.e. from the cooler to the hot regions of the tungsten sample. In this work, the findings of our thermal and species transport analysis have been implemented in our cluster-dynamics code, Xolotl, which has been used to compute temperature and species profiles over spatiotemporal scales representative of plasma-facing component (PFC) tungsten under typical reactor operating conditions, including extreme heat loads at the plasma-facing surface characteristic of plasma instabilities that induce edge localized modes (ELMs). We demonstrate that the steady-state species profiles, when properly accounting for the Soret effect, vary significantly from those where temperature-gradient-driven transport is not accounted for and discuss the implications of such a Soret effect on the response to plasma exposure of plasma-facing tungsten. Although our cluster-dynamics simulations do not yet include self-clustering of helium or hydrogen blister formation, our simulation results show that the Soret effect substantially reduces helium and hydrogenic species retention inside PFC tungsten.

36 MATERIALS SCIENCE↗

Efficient learning of power grid voltage control strategies via model-based deep reinforcement learning

Here this article proposes a model-based deep reinforcement learning (DRL) method to design emergency control strategies for short-term voltage stability problems in power systems. Recent advances show promising results for model-free DRL-based methods in power systems control problems. But in power systems applications, these model-free methods have certain issues related to training time (clock time) and sample efficiency; both are critical for making state-of-the-art DRL algorithms practically applicable. DRL-agent learns an optimal policy via a trial-and-error method while interacting with the real-world environment. It is also desirable to minimize the direct interaction of the DRL agent with the real-world power grid due to its safety-critical nature. Additionally, the state-of-the-art DRL-based policies are mostly trained using a physics-based grid simulator where dynamic simulation is computationally intensive, lowering the training efficiency. We propose a novel model-based DRL framework where a deep neural network (DNN)-based dynamic surrogate model (SM), instead of a real-world power grid or physics-based simulation, is utilized within the policy learning framework, making the process faster and more sample efficient. However, having stable training in model-based DRL is challenging because of the complex system dynamics of large-scale power systems. We addressed these issues by incorporating imitation learning to have a warm start in policy learning, reward-shaping, and multi-step loss in surrogate model training. Finally, we achieved 97.5% reduction in samples and 87.7% reduction in training time for an application to the IEEE 300-bus test system.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Electron-phonon coupling and superconductivity-induced distortion of the phonon lineshape in V 3 Si

Phonon measurements in the A15-type superconductors were complicated in the past because of the unavailability of large single crystals for inelastic neutron scattering, e.g., in the case of Nb 3 Sn, or unfavorable neutron scattering properties in the case of V 3 Si. Hence, only few studies of the lattice dynamical properties with momentum resolved methods were published, in particular below the superconducting transition temperature T c . Here, we overcome these problems by employing inelastic x-ray scattering and report a combined experimental and theoretical investigation of lattice dynamics in V 3 Si with the focus on the temperature-dependent properties of low-energy acoustic phonon modes in several high-symmetry directions. We paid particular attention to the evolution of the soft phonon mode of the structural phase transition observed in our sample at T s = 18.9 K, i.e., just above the measured superconducting phase transition at T c = 16.8 K. Theoretically, we predict lattice dynamics including electron-phonon coupling based on density-functional-perturbation theory and discuss the relevance of the soft phonon mode with regard to the value of T c . Furthermore, we explain superconductivity-induced anomalies in the lineshape of several acoustic phonon modes using a model proposed by Allen et al., [Phys. Rev. B 56, 5552 (19 7)].

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Experimental and Numerical Investigation of Flow Boiling in Additive Manufactured Foam Structures With Vapor Pathways

Abstract The unique properties of metal foams make them potential candidates for a range of applications, including microsystem thermal management. Using additive manufacturing to create foam-type structures can improve upon prior thermal solutions by eliminating thermal interface materials and allowing for customization/local control of parameters. In the present investigation, flow boiling in additive-manufactured metal foams is investigated both experimentally and numerically. Two test samples, one with uniform structure and the other with pathways for vapor removal, are compared both experimentally and numerically. A conjugate computational fluid dynamics and heat transfer (CFD-HT) model utilizing a three-dimensional volume of fluid (VOF) model with accompanying evaporation/condensation model provided in-depth visualization of the boiling flow phenomena. The experiments generated the thermohydraulic performance over a range of heat fluxes, demonstrating that the sample incorporating dedicated vapor pathways performed better in both pressure and heat transfer performance metrics compared to the uniform foam. Additionally, negative system-level effects (i.e., hydraulic oscillations) were shown to be abated using the vapor removal structures. The numerical model yielded further insight into the factors contributing to the improved performance. Results indicated the pathways functioned as vapor removal channels, allowing the generated vapor to vent from the foam structure into the lanes. Further computational investigations demonstrated changes in flow regimes, where the addition of vapor channels caused the flow to change from churn to annular. Bubble behavior unique to the vapor pathway structure was studied, showing stagnant regions that eject vapor into the channel.

Engineering↗

Detection of significant antiviral drug effects on COVID-19 with reasonable sample sizes in randomized controlled trials: A modeling study

Development of an effective antiviral drug for Coronavirus Disease 2019 (COVID-19) is a global health priority. Although several candidate drugs have been identified through in vitro and in vivo models, consistent and compelling evidence from clinical studies is limited. The lack of evidence from clinical trials may stem in part from the imperfect design of the trials. We investigated how clinical trials for antivirals need to be designed, especially focusing on the sample size in randomized controlled trials. A modeling study was conducted to help understand the reasons behind inconsistent clinical trial findings and to design better clinical trials. We first analyzed longitudinal viral load data for Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) without antiviral treatment by use of a within-host virus dynamics model. The fitted viral load was categorized into 3 different groups by a clustering approach. Comparison of the estimated parameters showed that the 3 distinct groups were characterized by different virus decay rates (p-value < 0.001). The mean decay rates were 1.17 d -1 (95% CI: 1.06 to 1.27 d -1 ), 0.777 d -1 (0.716 to 0.838 d -1 ), and 0.450 d -1 (0.378 to 0.522 d -1 ) for the 3 groups, respectively. Such heterogeneity in virus dynamics could be a confounding variable if it is associated with treatment allocation in compassionate use programs (i.e., observational studies). Subsequently, we mimicked randomized controlled trials of antivirals by simulation. An antiviral effect causing a 95% to 99% reduction in viral replication was added to the model. To be realistic, we assumed that randomization and treatment are initiated with some time lag after symptom onset. Using the duration of virus shedding as an outcome, the sample size to detect a statistically significant mean difference between the treatment and placebo groups (1:1 allocation) was 13,603 and 11,670 (when the antiviral effect was 95% and 99%, respectively) per group if all patients are enrolled regardless of timing of randomization. The sample size was reduced to 584 and 458 (when the antiviral effect was 95% and 99%, respectively) if only patients who are treated within 1 day of symptom onset are enrolled. We confirmed the sample size was similarly reduced when using cumulative viral load in log scale as an outcome. We used a conventional virus dynamics model, which may not fully reflect the detailed mechanisms of viral dynamics of SARS-CoV-2. The model needs to be calibrated in terms of both parameter settings and model structure, which would yield more reliable sample size calculation. In this study, we found that estimated association in observational studies can be biased due to large heterogeneity in viral dynamics among infected individuals, and statistically significant effect in randomized controlled trials may be difficult to be detected due to small sample size. The sample size can be dramatically reduced by recruiting patients immediately after developing symptoms. We believe this is the first study investigated the study design of clinical trials for antiviral treatment using the viral dynamics model.

60 APPLIED LIFE SCIENCES↗

Generative learning for slow manifolds and bifurcation diagrams

In dynamical systems characterized by separation of time scales, the approximation of so called “slow manifolds”, on which the long term dynamics lie, is a useful step for model reduction. Initializing on such slow manifolds is a useful step in modeling, since it circumvents fast transients, and is crucial in multiscale algorithms (like the equation-free approach) alternating between fine scale (fast) and coarser scale (slow) simulations. In a similar spirit, when one studies the infinite time dynamics of systems depending on parameters, the system attractors (e.g., its steady states) lie on bifurcation diagrams (curves for one-parameter continuation, and more generally, on manifolds in state parameter space. Sampling these manifolds gives us representative attractors (here, steady states of ODEs or PDEs) at different parameter values. Algorithms for the systematic construction of these manifolds (slow manifolds, bifurcation diagrams) are required parts of the “traditional” numerical nonlinear dynamics toolkit. In more recent years, as the field of Machine Learning develops, conditional score-based generative models (cSGMs) have been demonstrated to exhibit remarkable capabilities in generating plausible data from target distributions that are conditioned on some given label. It is tempting to exploit such generative models to produce samples of data distributions (points on a slow manifold, steady states on a bifurcation surface) conditioned on (consistent with) some quantity of interest (QoI, observable). In this work, we present a framework for using cSGMs to quickly (a) initialize on a low-dimensional (reduced-order) slow manifold of a multi-time-scale system consistent with desired value(s) of a QoI (a “label”) on the manifold, and (b) approximate steady states in a bifurcation diagram consistent with a (new, out-of-sample) parameter value. This conditional sampling can help uncover the geometry of the reduced slow-manifold and/or approximately “fill in” missing segments of steady states in a bifurcation diagram. Finally, the quantity of interest, which determines how the sampling is conditioned, is either known a priori or identified using manifold learning-based dimensionality reduction techniques applied to the training data.

Dynamical systems↗