Search NASA⌕ Search

SEARCH · Search NASA

Results for “missing data”

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 631 records · Page 35

Ultra-large-scale ab initio quantum chemical computation of bio-molecular systems: The case of spike protein of SARS-CoV-2 virus

The COVID-19 pandemic poses a severe threat to human health with an unprecedented social and economic disruption. Spike (S) glycoprotein of the SARS-CoV-2 virus is pivotal in understanding the virus anatomy, since it initiates the first contact with the ACE2 receptor in the human cell. We report results of ab initio computation of the spike protein, the largest ab initio quantum chemical computation to date on any bio-molecular system, using a divide and conquer strategy by focusing on individual structural domains. In this approach we divided the S-protein into seven structural domains: N-terminal domain (NTD), receptor binding domain (RBD), subdomain 1 (SD1), subdomain 2 (SD2), fusion peptide (FP), heptad repeat 1 with central helix (HR1-CH) and connector domain (CD). The entire Chain A has 14,488 atoms including the hydrogen atoms but excluding the amino acids with missing coordinates based on the PDB data (ID: 6VSB). The results include structural refinement, ab initio calculation of intramolecular bonding mechanism, 3- dimensional non-local inter-amino acid interaction with implications for the inter-domain interaction. Details of the electronic structure, interatomic bonding, partial charge distribution and the role played by hydrogen bond network are discussed. In the interaction among structural domains, we present new insights for crucial hinge-like movement and fusion process. Extension of such calculation to the interface between the S-protein binding domain and ACE2 receptor can provide a pathway for computational understanding of mutations and the design of therapeutic drugs to combat the COVID-19 pandemic.

59 BASIC BIOLOGICAL SCIENCES↗

Leveraging large language models to address data scarcity in machine learning for graphene synthesis

Machine learning in experimental materials science faces significant challenges due to the scarcity of data, which are costly and time-consuming to generate, particularly when relying on in-house experiments. Literature data mining offers a potential solution but introduces issues like mixed data quality, inconsistent formats, and non-uniform reporting of synthesis parameters, resulting in partially missing and heterogeneous features across the dataset. Here, we propose data imputation and feature engineering methods that employ pre-trained large language models (LLMs) to enhance machine learning performance on scarce, heterogeneous datasets, demonstrated on graphene CVD synthesis data and the ML-HydPARK hydrogen storage dataset. GPT models perform data imputation via tailored prompting and semantic normalization of inconsistently reported features through embeddings, for example, to harmonize the complex nomenclature of CVD substrates. Beyond yielding more diverse and richer feature representations than traditional methods such as K-nearest neighbors (KNN) and Multivariate Imputation by Chained Equations (MICE), LLM-based data imputation is evaluated against dataset characteristics and prompting strategies. We vary the level of autonomy granted to the LLM, from generic prompting that leverages pre-trained knowledge for autonomous data generation to data-informed prompting that constrains outputs using target-specific information, and demonstrate which level of autonomy yields superior imputation performance across datasets and feature types. The proposed data engineering methods markedly improve downstream performance; for example, in graphene layer number classification using a support vector machine (SVM), binary accuracy increases from 39% to 65% and ternary accuracy from 52% to 72%. Fine-tuning experiments on both datasets show that combining our proposed LLM-based data imputation and feature encoding methods with numerical machine learning predictors outperforms standalone fine-tuned LLM predictors in data-scarce settings. The proposed strategies emphasize data enhancement techniques rather than refining learning architectures or regularizing loss functions, offering a broadly applicable framework for improving machine learning performance on scarce, inhomogeneous datasets.

Chemical vapor deposition↗

Scientific data interpolation with low dimensional manifold model

Here, we propose to apply a low dimensional manifold model to scientific data interpolation from regular and irregular samplings with a significant amount of missing information. The low dimensionality of the patch manifold for general scientific data sets has been used as a regularizer in a variational formulation. The problem is solved via alternating minimization with respect to the manifold and the data set, and the Laplace–Beltrami operator in the Euler–Lagrange equation is discretized using the weighted graph Laplacian. Various scientific data sets from different fields of study are used to illustrate the performance of the proposed algorithm on data compression and interpolation from both regular and irregular samplings.

97 MATHEMATICS AND COMPUTING↗

HPC formulations of optimization algorithms for tensor completion

Tensor completion is a powerful tool used to estimate or recover missing values in multi-way data. It has seen great success in domains such as product recommendation and healthcare. Tensor completion is most often accomplished via low-rank sparse tensor factorization, a computationally expensive non-convex optimization problem which has only recently been studied in the context of parallel computing. In this work, we study three optimization algorithms that have been successfully applied to tensor completion: alternating least squares (ALS), stochastic gradient descent (SGD), and coordinate descent (CCD++). We explore opportunities for parallelism on shared- and distributed-memory systems and address challenges such as memory- and operation-efficiency, load balance, cache locality, and communication. Among our advancements are a communication-efficient CCD++ algorithm, an ALS algorithm rich in level-3 BLAS routines, and an SGD algorithm which combines stratification with asynchronous communication. Furthermore, we show that introducing randomization during ALS and CCD++ can accelerate convergence. We evaluate our parallel formulations on a variety of real datasets on a modern supercomputer and demonstrate speedups through 16384 cores. Further, these improvements reduce time-to-solution from hours to seconds on real-world datasets. We show that after our optimizations, ALS is advantageous on parallel systems of small-to-moderate scale, while both ALS and CCD++ provide the lowest time-to-solution on large-scale distributed systems.

97 MATHEMATICS AND COMPUTING↗

Search for gauge-mediated supersymmetry in events with at least one photon and missing transverse momentum in pp collisions at s = 13 TeV

A search for gauge-mediated supersymmetry (SUSY) in final states with photons and large missing transverse momentum is presented. The data sample of pp collisions at $\sqrt{s} = $ 13 TeV was collected with the CMS detector at the CERN LHC and corresponds to an integrated luminosity of 35.9 fb$^{-1}$. Data are compared with models in which the lightest neutralino has bino- or wino-like components, resulting in decays to photons and gravitinos, where the gravitinos escape detection. The event selection is optimized for both electroweak (EWK) and strong production SUSY scenarios. The observed data are consistent with standard model predictions, and limits are set in the context of a general gauge mediation model in which gaugino masses up to 980 GeV are excluded at 95% confidence level. Gaugino masses below 780 and 950 GeV are excluded in two simplified models with EWK production of mass-degenerate charginos and neutralinos. Stringent limits are set on simplified models based on gluino and squark pair production, excluding gluino (squark) masses up to 2100 (1750) GeV depending on the assumptions made for the decay modes and intermediate particle masses. This analysis sets the highest mass limits to date in the studied EWK models, and in the considered strong production models when the mass difference between the gauginos and the squarks or gluinos is small.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Particle-Phase Chemistry of Secondary Organic Material: Modeled Compared to Measured O:C and H:C Elemental Ratios Provide Constraints

Chemical mechanisms for the production of secondary organic material (SOM) are developed in focused laboratory studies but widely used in the complex modeling context of the atmosphere. Given this extrapolation, a stringent testing of the mechanisms is important. In addition to particle mass yield as a typical standard for model-measurement comparison, particle composition expressed as O:C and H:C elemental ratios can serve as a higher dimensional constraint. A paradigm for doing so is developed herein for SOM production from a C5 C10 C15 terpene sequence, namely isoprene, R-pinene, and β-caryopyhllene. The model MCM-SIMPOL is introduced based on the Master Chemical Mechanism (MCM v3.2) and a group contribution method for vapor pressures (SIMPOL). The O:C and H:C ratios of the SOM are measured using an Aerosol Mass Spectrometer (AMS). Detailed SOM-specific AMS calibrations for the organic contribution to the H2Oþ and COþ ions indicate that published O:C and H:C ratios for SOM are systematically too low. Overall, the measurement-model gap was small for particle mass yield but significant for particle-average elemental composition. The implication is that a key chemical pathway is missing from the chemical mechanism. The data can be explained by the particle-phase homolytic decomposition of organic hydroperoxides and subsequent alkyl-radical-promoted oligomerization.

54 ENVIRONMENTAL SCIENCES↗

The Need for a New LLNL Pulsed Sphere Neutron Leakage Spectra Series

Here, it is shown that spectra measured as part of the Lawrence Livermore National Laboratory Pulsed Sphere (LPS) program offer decisive information to locate formatting or physics issues in nuclear data of key interest for fusion reactor simulations. However, experiments from this measurement series are not benchmarks. For instance, their uncertainties are incomplete. There are also many open questions—e.g., on the setup, the detector response, and whether LPS are accurately modeled—that cannot be answered anymore given the limited documentation and that many of the experimenters are no longer actively working. This limited knowledge has implications when one tries to adjust nuclear data to LPS spectra. Usually, one adjusts to benchmarks representing an application with the hope to get more precise nuclear data for the application of interest where differential data might be scarce and/or to reduce nuclear data uncertainties in the application simulations. However, it is demonstrated that adjustment with LPS spectra without accounting for missing uncertainties and modeling potential biases in the experimental data leads to adjusted data that are highly unphysical. That means adjusted data differ significantly from evaluated data based on information from differential experiments; also, application quantities predicted with the adjusted data deviate distinctly from experimental ones. While we can approximate our limited knowledge on these experiments with Gaussian processes in the adjustment process, this modeling of bias is arbitrary rather than based on a physics explanation, calling into doubt the validity of resulting adjusted data. Thus, we discuss here the need for a new measurement series, learning from the strengths and weaknesses of the LPS program, to yield decisive and well-benchmarked integral experiments to support fusion reactor research.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

ATCA survey of ammonia in the galactic center: The temperatures of dense gas clumps between Sgr A* and Sgr B2

We present a large-scale, interferometric survey of ammonia (1, 1) and (2, 2) toward the Galactic center observed with the Australia Telescope Compact Array. The survey covers Δℓ ∼ 1° (∼150 pc at an assumed distance of 8.5 kpc) and Δb ∼ 0.°2 (∼30 pc) which spans the region between the supermassive black hole Sgr A* and the massive star forming region Sgr B2. The resolution is ∼20'' (∼0.8 pc) and emission at scales ≳ 2' (≳ 3.2 pc) is filtered out due to missing interferometric short spacings. Consequently, the data represent the denser, compact clouds and disregards the large-scale, diffuse gas. Many of the clumps align with the 100 pc dust ring and mostly anti-correlate with 1.2 cm continuum emission. We present a kinetic temperature map of the dense gas. The temperature distribution peaks at ∼38 K with a width at half maximum between 18 K and 61 K (measurements sensitive within T {sub kin} ∼ 10-80 K). Larger clumps are on average warmer than smaller clumps which suggests internal heating sources. Our observations indicate that the circumnuclear disk ∼1.5 pc around Sgr A* is supplied with gas from the 20 km s{sup –1} molecular cloud. This gas is substantially cooler than gas ∼3-15 pc away from Sgr A*. We find a strong temperature gradient across Sgr B2. Ammonia column densities correlate well with SCUBA 850 μm fluxes, but the relation is shifted from the origin, which may indicate a requirement for a minimum amount of dust to form and shield ammonia. Around the Arches and Quintuplet clusters we find shell morphologies with UV-influenced gas in their centers, followed by ammonia and radio continuum layers.

79 ASTRONOMY AND ASTROPHYSICS↗

Correlations of π N partial waves for multireaction analyses

In the search for missing baryonic resonances, many analyses include data from a variety of pion- and photon-induced reactions. For elastic πN scattering, however, usually the partial waves of the SAID (Scattering Analysis Interactive Database) or other groups are fitted, instead of data. We provide the partial-wave covariance matrices needed to perform correlated χ 2 fits, in which the obtained χ 2 equals the actual χ 2 up to nonlinear and normalization corrections. For any analysis relying on partial waves extracted from elastic pion scattering, this is a prerequisite to assess the significance of resonance signals and to assign any uncertainty on results. Lastly, the influence of systematic errors is also considered.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Search for New Bottomlike Quark Pair Decays QQ->(tW{sup +}-)(tW{sup +}-) in Same-Charge Dilepton Events

We report the most restrictive direct limits on masses of fourth-generation down-type quarks b{sup '}, and quarklike composite fermions (B or T{sub 5/3}), decaying promptly to tW{sup +}-. We search for a significant excess of events with two same-charge leptons (e, mu), several hadronic jets, and missing transverse energy. An analysis of data from pp collisions with an integrated luminosity of 2.7 fb{sup -1} collected with the CDF II detector at Fermilab yields no evidence for such a signal, setting mass limits m{sub b}{sup '}, m{sub B}>338 GeV/c{sup 2} and m{sub T{sub 5{sub /{sub 3}}}}>365 GeV/c{sup 2} at 95% confidence level.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Search for Diphoton Events with Large Missing Transverse Energy in 7 TeV Proton-Proton Collisions with the ATLAS Detector

A search for diphoton events with large missing transverse energy is presented. The data were collected with the ATLAS detector in proton-proton collisions at {radical}(s)=7 TeV at the CERN Large Hadron Collider and correspond to an integrated luminosity of 3.1 pb{sup -1}. No excess of such events is observed above the standard model background prediction. In the context of a specific model with one universal extra dimension with compactification radius R and gravity-induced decays, values of 1/R<729 GeV are excluded at 95% C. L., providing the most sensitive limit on this model to date.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

LaueMatching: an approach for rapid and robust indexing of Laue diffraction patterns

Traditional Laue diffraction pattern indexing often struggles with noisy data, weak signals, peak overlap and missing reflections, particularly from complex or deformed microstructures. Here, we introduce LaueMatching, a high-throughput indexing algorithm designed to overcome these limitations. LaueMatching utilizes a fundamentally different approach based on direct pattern correlation: experimentally pre-processed images are compared against a comprehensive pre-computed library of simulated diffraction patterns corresponding to a dense grid of possible orientations. This approach bypasses the need for explicit peak identification and fitting, steps that are often a failure point for traditional methods. The algorithm rapidly and robustly indexes multiple crystallographic orientations and crystal systems simultaneously, even from challenging patterns. LaueMatching's effectiveness and accuracy have been rigorously tested and validated on diverse experimental (Ni, Al, EuAl 2 O 4 ) and simulated diffraction patterns, demonstrating high-fidelity orientation refinement. Code to implement this approach on both CPU and GPU resources can be downloaded from https://github.com/AdvancedPhotonSource/LaueMatching.

36 MATERIALS SCIENCE↗

Role of the Hydrophobic Bridge in the Carbapenemase Activity of Class D β-Lactamases

Class D carbapenemases are enzymes of the utmost clinical importance due to their ability to confer resistance to the last-resort carbapenem antibiotics. We investigated the role of the conserved hydrophobic bridge in the carbapenemase activity of OXA-23, the major carbapenemase of the important pathogen Acinetobacter baumannii . We show that substitution of the bridge residue Phe110 affects resistance to meropenem and doripenem and has little effect on MICs of imipenem. The opposite effect was observed upon substitution of the other bridge residue Met221. Complete disruption of the bridge by the F110A/M221A substitution resulted in a significant loss of affinity for doripenem and meropenem and to a lesser extent for imipenem, which is reflected in the reduced MICs of these antibiotics. In the wild-type OXA-23, the pyrrolidine ring of the meropenem tail forms a hydrophobic interaction with Phe110 of the bridge. Similar interactions would ensue with ring-containing doripenem but not with imipenem, which lacks this ring. Our structural studies showed that this interaction with the meropenem tail is missing in the F110A/M221A mutant. These data explain why disruption of the interaction between the enzyme and the carbapenem substrate impacts the affinity and MICs of meropenem and doripenem to a larger degree than those of imipenem. Our structures also show that the bridge directs the acylated carbapenem into a specific tautomeric conformation. Furthermore, it is not this conformation but rather the stabilizing interaction between the tail of the antibiotic and the hydrophobic bridge that contributes to the carbapenemase activity of class D β-lactamases.

59 BASIC BIOLOGICAL SCIENCES↗

Demonstration of neutron identification in neutrino interactions in the MicroBooNE liquid argon time projection chamber

A significant challenge in measurements of neutrino oscillations is reconstructing the incoming neutrino energies. While modern fully-active tracking calorimeters such as liquid argon time projection chambers in principle allow the measurement of all final state particles above some detection threshold, undetected neutrons remain a considerable source of missing energy with little to no data constraining their production rates and kinematics. We present the first demonstration of tagging neutrino-induced neutrons in liquid argon time projection chambers using secondary protons emitted from neutron-argon interactions in the MicroBooNE detector. We describe the method developed to identify neutrino-induced neutrons and demonstrate its performance using neutrons produced in muon-neutrino charged current interactions. The method is validated using a small subset of MicroBooNE’s total dataset. The selection yields a sample with 60% of selected tracks corresponding to neutron-induced secondary protons. At this purity, the integrated efficiency is 8.4% for neutrons that produce a detectable proton.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Quality-Controlled Meteorological Data from the Flood Control District of Maricopa County (FCDMC) Network, Phoenix, Arizona (1987-2024)

This dataset contains 15- or 30-minute interval meteorological data from the Flood Control District of Maricopa County (FCDMC), Arizona, USA, covering eight key variables across multiple sensor stations between 1987 and 2024. Each variable is stored as a separate CSV file, containing time-series data that have undergone rigorous quality control (QC) procedures and, where appropriate, short-gap interpolation for consistency. The quality control (QC) pipeline consisted of four sequential tests: (1) a range test to ensure all values fall within physically realistic limits, (2) a step test to identify abrupt and implausible changes between consecutive records, (3) a proximity test that validates flagged values from step test using data from nearby stations and exceedance probability thresholds, and (4) a persistence test to detect and remove periods of unrealistically constant readings. These thresholds were calibrated to Arizona’s environmental conditions and sensor specifications. After QC, short gaps (≤2 hours) were linearly interpolated to ensure consistent temporal resolution, except for wind variables. Due to a major upgrade in FCDMC’s data transmission system, only ALERT-2 protocol data (2016–2024) for wind variables are included; earlier ALERT-1 data were excluded because of irregular sampling and high missing rates. This dataset supports regional climate and infrastructure resilience studies by providing standardized, high-resolution meteorological data for the greater Phoenix metropolitan area.

54 ENVIRONMENTAL SCIENCES↗

Detecting Neutrons in MicroBooNE

A significant challenge in measurements of neutrino oscillations is reconstructing the incoming neutrino energies. While modern fully-active tracking calorimeters such as liquid argon time projection chambers in principle allow the measurement of all final state particles above some detection threshold, undetected neutrons remain a considerable source of missing energy with little to no data constraining their production rates and kinematics. We present the first demonstration of tagging neutrino-induced neutrons in liquid argon time projection chambers using secondary protons emitted from neutron-argon interactions in the MicroBooNE detector. We describe the method developed to identify neutrino-induced neutrons and demonstrate its performance using neutrons produced in muon-neutrino charged current interactions. The method is validated using a small subset of MicroBooNE's total dataset. The selection yields a sample with $60\%$ of selected tracks corresponding to neutron-induced secondary protons.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Transformer Health Monitoring Using Dissolved Gas Analysis

As integral components of any power plant, transformers supply the generated electricity to the grid. However, a transformer’s cellulose-based paper insulation and the mineral oil in which it is immersed break down over time under standard operating conditions—or more rapidly due to potential faults within the system. As the transformer’s mineral oil breaks down, gases are released that can be measured and monitored. This technical brief exhibits a collection of diagnostic and prognostic techniques that utilities can adopt in lieu of labor-intensive periodic preventive maintenance routines. Furthermore, prognostic models have been incorporated using the latest version of the Institute of Electrical and Electronics Engineers (IEEE) standard (IEEE, 2019) for dissolved gas analysis (DGA), thus expanding it to include estimation of the time to maintenance. Overall, four different methodologies are explained, each of which aids in determining a transformer’s state of health. These methodologies include the Chendong model, the IEEE thermal life consumption model (IEEE, 2012), a diagnostic model for DGA, and a prognostic model for DGA that uses an autoregressive integrated moving average (ARIMA) model. An additional improvement for estimating missing system parameters by using monitoring data (i.e., a tool for parameter estimation utilizing Powell’s method) is presented, enabling the IEEE thermal life consumption model to benefit not only the collaborating power plant, but also the power industry at large.

24 POWER TRANSMISSION AND DISTRIBUTION↗