Search NASA⌕ Search

SEARCH · Search NASA

Results for “Domain knowledge”

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 343 records · Page 19

Ontology Language to Support Description of Experiment Control System Semantics, Collaborative Knowledge-Base Design and Ontology Reuse

In this paper we discuss the control domain specific ontology that is built on top of the domain-neutral Resource Definition Framework (RDF). Specifically, we will discuss the relevant set of ontology concepts along with the relationships among them in order to describe experiment control components and generic event-based state machines. Control Oriented Ontology Language (COOL) is a meta-data modeling language that provides generic means for representation of physics experiment control processes and components, and their relationships, rules and axioms. It provides a semantic reference frame that is useful for automating the communication of information for configuration, deployment and operation. COOL has been successfully used to develop a complete and dynamic knowledge-base for experiment control systems, developed using the AFECS framework.

43 PARTICLE ACCELERATORS↗

Social Network: JAZ Protein Interactions Expand Our Knowledge of Jasmonate Signaling

Members of the family of JASMONATE ZIM-DOMAIN (JAZ) proteins are key regulators of the jasmonate (JA) hormonal response. The 12-member family is characterized by three conserved domains, an N-terminal domain, a TIFY-containing ZINC-FINGER EXPRESSED IN INFLORESCENCE MERISTEM domain, and a C-terminal Jas domain. JAZ proteins regulate JA-responsive gene transcription by inhibiting DNA-binding transcription factors in the absence of JA. JAZ proteins interact in a hormone-dependent manner with CORONATINE INSENSITIVE 1 (COI1), the recognition component of the E3 ubiquitin ligase, SCFCOI1, resulting in the ubiquitination and subsequent degradation of JAZs via the 26S proteasome pathway. Since their discovery in 2007, JAZ proteins have been implicated in protein–protein interactions with multiple transcription factors. These studies have shed light on the mechanism by which JAZs repress transcription, are targeted for degradation, modulate the JA signaling response, and participate in crosstalk with other hormone signaling pathways. In this review, we will take a close look at the recent discoveries made possible by the characterization JAZ protein–protein interactions.

54 ENVIRONMENTAL SCIENCES↗

Real-time, digital pulse-shape discrimination in non-hazardous fast liquid scintillation detectors: Prospects for safety and security

Pulse-shape discrimination (PSD) in fast, organic scintillation detectors is a long-established technique used to separate neutrons and {gamma} rays in mixed radiation fields. In the analogue domain the method can achieve separation in real time, but all knowledge of the pulses themselves is lost thereby preventing the possibility of any post- or repeated analysis. Also, it is typically reliant on electronic systems that are largely obsolete and which require significant experience to set up. In the digital domain, PSD is often more flexible but significant post-processing has usually been necessary to obtain neutron/{gamma}-ray separation. Moreover, the scintillation media on which the technique relies usually have a low flash point and are thus deemed hazardous. This complicates the ease with which they are used in industrial applications. In this paper, results obtained with a new portable digital pulse-shape discrimination instrument are described. This instrument provides real-time, digital neutron/{gamma} separation whilst preserving the synchronization with the time-of-arrival for each event, and realizing throughputs of 3 x 10{sup 6} events per second. Furthermore, this system has been tested with a scintillation medium that is non-flammable and not hazardous. (authors)

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Calculation of Higher Eigenpairs of the Transport Equation Using IRAM Based on Domain Decomposition

The higher eigenpairs of the neutron transport (or diffusion) equation are useful in core flux distribution reconstruction, transient analysis, stability analysis, and loading pattern optimization. There are various methods for the calculation of higher eigenpairs, such as the power iteration method with filtering technique, the Generalized Modified Power Method (GMPM), the power iteration method with innovative boundary conditions for symmetric cores, and Krylov subspace methods. In regard to Krylov subspace methods, the Implicitly Restarted Arnoldi Method (IRAM) and ORTHOMIN were implemented to calculate higher eigenpairs for both transport and diffusion equations. In Warsa's work, however, there was a significant drawback that problems with energy upscatter could not be treated efficiently. In Modak's work, the mode to which his method was likely to converge depended on the shape of initial guess flux, and only one mode could be computed each time, making it impractical for real use. Furthermore, the computational burden for higher eigenpairs is generally much larger than for only the first eigenpair, especially for neutron transport equations. Therefore, the implementation of parallel computation for higher eigenpairs is essential to improve the efficiency, and no such work has been reported in literature so far. In this summary, we propose an innovative method to calculate higher eigenpairs of the neutron transport equation based on high performance computation. In this method, 2D Matrix Method of Characteristic (MMOC) is used to discretize the neutron transport equation; IRAM is used to evaluate higher eigenpairs; spatial domain decomposition is used to realize parallel computation. The highlight of this method is using the MGMD PGMRES (Multi-Group and Multi-Domain coupled Parallel Generalized Minimal Residual Method) algorithm to simultaneously accelerate the upscatter iterations and the inner interface angular fluxes iterations, avoiding the drawback in Warsa's method and the efficiency problem of conventional Jacobi iteration. To the best of our knowledge, our work represents the first application of IRAM based on domain decomposition for higher eigenpairs.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Staphylococcus aureus FtsZ and PBP4 bind to the conformationally dynamic N-terminal domain of GpsB

In the Firmicutes phylum, GpsB is a membrane associated protein that coordinates peptidoglycan synthesis with cell growth and division. Although GpsB has been studied in several bacteria, the structure, function, and interactome of Staphylococcus aureus GpsB is largely uncharacterized. To address this knowledge gap, we solved the crystal structure of the N-terminal domain of S. aureus GpsB, which adopts an atypical, asymmetric dimer, and demonstrates major conformational flexibility that can be mapped to a hinge region formed by a three-residue insertion exclusive to Staphylococci. When this three-residue insertion is excised, its thermal stability increases, and the mutant no longer produces a previously reported lethal phenotype when overexpressed in Bacillus subtilis. In S. aureus, we show that these hinge mutants are less functional and speculate that the conformational flexibility imparted by the hinge region may serve as a dynamic switch to fine-tune the function of the GpsB complex and/or to promote interaction with its various partners. Furthermore, we provide the first biochemical, biophysical, and crystallographic evidence that the N-terminal domain of GpsB binds not only PBP4, but also FtsZ, through a conserved recognition motif located on their C-termini, thus coupling peptidoglycan synthesis to cell division. Taken together, the unique structure of S. aureus GpsB and its direct interaction with FtsZ/PBP4 provide deeper insight into the central role of GpsB in S. aureus cell division.

59 BASIC BIOLOGICAL SCIENCES↗

Predictive Indicators of the Performance of Large Language Models

In several mission contexts, it is desirable to estimate the performance of large language models (LLMs) on tasks that we cannot run directly. In light of published “scaling laws” our hypothesis is that some tasks should be consistently more challenging than others based on characteristics of the task. The goal of this project was to begin quantifying how much information about LLM performance can be gained from the features of a model and a task. Two of our statistical models struggled to converge. Pass/fail test results may provide limited information for inference beyond model quality and task difficulty, but we see no evidence at this time for significant feature interaction effect sizes, arguing for simple models. Future work extending the models to capitalize on perplexity of ground truth answers is suggested. This project also introduces “Depth of Knowledge Variant Testing” as a strategy for more finely assessing language models on open domain question and answer tasks. We developed sets of questions that ask a language model to produce similar information while demonstrating increasing depth of knowledge, and also relabeled existing Q&A test questions with their depth of knowledge. Our results suggest further consideration of Bloom’s taxonomy and further refinement of prompts to properly elicit information at varying depths. In the course of this work, we set up a basic infrastructure for standardizing tasks and testing many language models on these tasks. In addition to testing the predictive quality of model features and performance across test suites, with this project we have introduced two new task features to contextualize each test question: the Dewey Classification main category of information covered, and the Bloom’s taxonomy level that corresponds to the depth of knowledge probed by the question. Splits across these and other features produced over five hundred task subtypes with distinct feature vectors, which we tested on half a dozen models.

97 MATHEMATICS AND COMPUTING↗

A new estimate of the China temperature anomaly series and uncertainty assessment in 1900–2006

While global warming during the last century has been well recognized, the magnitude of the climate warming in regions such as China over the past 100 years still has some uncertainty due to limited observations during the early years. Several series of temperature anomalies for the 20th century in China have been independently developed by different groups. The uncertainty arises mainly from the sparse observations before 1950, where statistics are sensitive to the small and potentially biased sample. In this study, BSHADE–MSN (Biased Sentinel Hospitals Areal Disease Estimation and Means of Stratified Nonhomogeneous Surface), a combination of two novel distinct statistical methods that are applicable with different sample situations to a spatial heterogeneous surface, is applied to estimate annual mean temperature anomalies for China. This method takes into account prior knowledge of geographical spatial autocorrelation and nonhomogeneity of target domains, remedies the biased sample, and maximizes an objective function for the best linear unbiased estimation (BLUE) of the regional mean quantity. For the period 1900–1999, the overall trend estimated by BSHADE–MSN is 0.80°C with a 95% confidential interval between 0.41°C and 1.18°C. This is significantly lower than that calculated by Climate Anomaly Method (CAM) and Block Kriging. The new temperature anomaly series for China exhibits slightly warmer conditions for the period before 1950 than existing studies. All the methods applied so far agree well with each other for the period after 1950, when there are sufficient stations across the country for the estimation of temperature anomaly series. Cross validation shows that the new regional mean temperature anomaly series has smaller estimation error variance and higher accuracy than those based on the other methods assessed in this study.

54 ENVIRONMENTAL SCIENCES↗

Inositol Pentakisphosphate Isomers Bind PH Domains with Varying Specificity and Inhibit Phosphoinositide Interactions

PH domains represent one of the most common domains in the human proteome. These domains are recognized as important mediators of protein-phosphoinositide and protein-protein interactions. Phosphoinositides are lipid components of the membrane that function as signaling molecules by targeting proteins to their sites of action. Phosphoinositide based signaling pathways govern a diverse range of important cellular processes including membrane remodeling, differentiation, proliferation and survival. Myo-Inositol phosphates are soluble signaling molecules that are structurally similar to the head groups of phosphoinositides. These molecules have been proposed to function, at least in part, by regulating PH domain-phosphoinositide interactions. Given the structural similarity of inositol phosphates we were interested in examining the specificity of PH domains towards the family of myo-inositol pentakisphosphate isomers. In work reported here we demonstrate that the C-terminal PH domain of pleckstrin possesses the specificity required to discriminate between different myo-inositol pentakisphosphate isomers. The structural basis for this specificity was determined using high-resolution crystal structures. Moreover, we show that while the PH domain of Grp1 does not possess this high degree of specificity, the PH domain of protein kinase B does. These results demonstrate that some PH domains possess enough specificity to discriminate between myo-inositol pentakisphosphate isomers allowing for these molecules to differentially regulate interactions with phosphoinositides. Furthermore, this work contributes to the growing body of evidence supporting myo-inositol phosphates as regulators of important PH domain-phosphoinositide interactions. Finally, in addition to expanding our knowledge of cellular signaling, these results provide a basis for developing tools to probe biological pathway.

59 BASIC BIOLOGICAL SCIENCES↗

Community ecology across bacteria, archaea and microbial eukaryotes in the sediment and seawater of coastal Puerto Nuevo, Baja California

Microbial communities control numerous biogeochemical processes critical for ecosystem function and health. Most analyses of coastal microbial communities focus on the characterization of bacteria present in either sediment or seawater, with fewer studies characterizing both sediment and seawater together at a given site, and even fewer studies including information about non-bacterial microbial communities. As a result, knowledge about the ecological patterns of microbial biodiversity across domains and habitats in coastal communities is limited-despite the fact that archaea, bacteria, and microbial eukaryotes are present and known to interact in coastal habitats. To better understand microbial biodiversity patterns in coastal ecosystems, we characterized sediment and seawater microbial communities for three sites along the coastline of Puerto Nuevo, Baja California, Mexico using both 16S and 18S rRNA gene amplicon sequencing. We found that sediment hosted approximately 500-fold more operational taxonomic units (OTUs) for bacteria, archaea, and microbial eukaryotes than seawater (p < 0.001). Distinct phyla were found in sediment versus seawater samples. Of the top ten most abundant classes, Cytophagia (bacterial) and Chromadorea (eukaryal) were specific to the sediment environment, whereas Cyanobacteria and Bacteroidia (bacterial) and Chlorophyceae (eukaryal) were specific to the seawater environment. A total of 47 unique genera were observed to comprise the core taxa community across environment types and sites. No archaeal taxa were observed as part of either the abundant or core taxa. No significant differences were observed for sediment community composition across domains or between sites. For seawater, the bacterial and archaeal community composition was statistically different for the Major Outlet site (p < 0.05), the site closest to a residential area, and the eukaryal community composition was statistically different between all sites (p < 0.05). Our findings highlight the distinct patterns and spatial heterogeneity in microbial communities of a coastal region in Baja California, Mexico.

59 BASIC BIOLOGICAL SCIENCES↗

A Data Processing Pipeline for Adversarial Socio-Technical Network Analysis

With the rapid adoption of emerging technologies, there is a need to catalog and model sociotechnical interdependencies that have been historically used to influence the operation of Critical Infrastructure networks including the impacts of mergers and acquisitions, hostile takeovers, and foreign investment. Our research intends to address this need with two primary contributions. First, we have developed a data curation and processing pipeline to generate sociotechnical networks extracted from a variety of data sources including SEC filings and infrastructure asset databases. The pipeline, implemented in Apache Airflow, extracts and normalizes the representation of entities and relations, specified within ontologies. Our intent is to provide an extensible, machine-actionable approach to quickly communicate such models, reproduce previous results, and adapt them to new, unanticipated situations. Second, networks produced by our pipeline enable the development of graph-theoretic metrics that consider the properties of network components in addition to its topology. Metadata associated with network components---whether semantic, temporal, or geospatial---affects the alignment of generated networks with assumptions underlying complexity metrics. Validation of generated networks relative to component types defined by an ontology, may allow the research community to adapt metrics to the semantics of the domains being studied. Generated networks may be processed as knowledge, dynamic, or spatial graphs and enables a variety of analyses including automated reasoning and measures of network complexity. Automated reasoning views extracted entities and relations as a knowledge graph; this enables application of inference rules that represent historically-attested adversarial business methods and applies that behavior to a specific geographic context. Measures of network complexity, including degree distribution, reachability analyses, temporal analysis, and community detection can be adapted to indicate adversarial organizational influence.

97 MATHEMATICS AND COMPUTING↗

Semantic Property Graph for Scalable Knowledge Graph Analytics

Graphs are a natural and fundamental representation to describe entities, relationships, activities, and evolution of complex systems. Many domains such as communication, citation, procurement, biology, social media, and transportation can be modeled as a set of entities and their relationships. Resource Description Framework (RDF) and Labeled Property Graph (LPG) are two of the most used data models to encode information in a graph. Both models are similar in terms of using basic graph elements such as nodes and edges but differ in terms of the modeling approach, expressibility, serialization, and target applications. RDF is a flexible data exchange model for expressing information about entities but it tends to a have high memory footprint and inefficient storage, which does not make it a natural choice to perform scalable graph analytics. In contrast, LPG has gained traction as a reliable model to perform scalable graph analytic tasks such as sub-graph matching, network alignment, and real-time knowledge graph query. It provides efficient storage, fast traversal, and flexibility to model various real-world domains. At the same time, the LPG lacks the support of a formal knowledge representation such as an ontology to provide automated knowledge inference. We propose Semantic Property Graph (SPG) as a logical projection of reified RDF into the LPG model. SPG continues to use RDF ontology to define the type hierarchy of the projected graph and validate it against a given ontology. We present a framework to convert reified RDF graphs into SPG using two different computing environments. We also present cloud-based graph migration capabilities using Amazon Web Services.

Purohit, Sumit↗

SU-E-J-71: Spatially Preserving Prior Knowledge-Based Treatment Planning

Purpose: Prior knowledge-based treatment planning is impeded by the use of a single dose volume histogram (DVH) curve. Critical spatial information is lost from collapsing the dose distribution into a histogram. Even similar patients possess geometric variations that becomes inaccessible in the form of a single DVH. We propose a simple prior knowledge-based planning scheme that extracts features from prior dose distribution while still preserving the spatial information. Methods: A prior patient plan is not used as a mere starting point for a new patient but rather stopping criteria are constructed. Each structure from the prior patient is partitioned into multiple shells. For instance, the PTV is partitioned into an inner, middle, and outer shell. Prior dose statistics are then extracted for each shell and translated into the appropriate Dmin and Dmax parameters for the new patient. Results: The partitioned dose information from a prior case has been applied onto 14 2-D prostate cases. Using prior case yielded final DVHs that was comparable to manual planning, even though the DVH for the prior case was different from the DVH for the 14 cases. Solely using a single DVH for the entire organ was also performed for comparison but showed a much poorer performance. Different ways of translating the prior dose statistics into parameters for the new patient was also tested. Conclusion: Prior knowledge-based treatment planning need to salvage the spatial information without transforming the patients on a voxel to voxel basis. An efficient balance between the anatomy and dose domain is gained through partitioning the organs into multiple shells. The use of prior knowledge not only serves as a starting point for a new case but the information extracted from the partitioned shells are also translated into stopping criteria for the optimization problem at hand.

60 APPLIED LIFE SCIENCES↗

Structural basis for antibody binding to adenylate cyclase toxin reveals RTX linkers as neutralization-sensitive epitopes

RTX leukotoxins are a diverse family of prokaryotic virulence factors that are secreted by the type 1 secretion system (T1SS) and target leukocytes to subvert host defenses. T1SS substrates all contain a C-terminal RTX domain that mediates recruitment to the T1SS and drives secretion via a Brownian ratchet mechanism. Neutralizing antibodies against the Bordetella pertussis adenylate cyclase toxin, an RTX leukotoxin essential for B . pertussis colonization, have been shown to target the RTX domain and prevent binding to the α M β 2 integrin receptor. Knowledge of the mechanisms by which antibodies bind and neutralize RTX leukotoxins is required to inform structure-based design of bacterial vaccines, however, no structural data are available for antibody binding to any T1SS substrate. Here, we determine the crystal structure of an engineered RTX domain fragment containing the α M β 2 -binding site bound to two neutralizing antibodies. Notably, the receptor-blocking antibodies bind to the linker regions of RTX blocks I–III, suggesting they are key neutralization-sensitive sites within the RTX domain and are likely involved in binding the α M β 2 receptor. As the engineered RTX fragment contained these key epitopes, we assessed its immunogenicity in mice and showed that it elicits similar neutralizing antibody titers to the full RTX domain. The results from these studies will support the development of bacterial vaccines targeting RTX leukotoxins, as well as next-generation B . pertussis vaccines.

59 BASIC BIOLOGICAL SCIENCES↗

Towards Geospatial Knowledge Graph Infused Neuro-Symbolic AI for Remote Sensing Scene Understanding

Deep learning has proven its effectiveness in numerous tasks for remote sensing scene understanding. However there is an increasing interest to explore fusion of domain-specific background information to the deep neural network to further improve its performance. Remote sensing researchers are also working towards developing models that generalize and adapt to multiple applications. Generalization challenges coupled with the scarcity of large corpora of high-quality noise-free labelled data, have together fueled an interest for leveraging background information. Knowledge graphs serve as excellent choice to represent domain-specific information in a structured, standardized and extensible manner. Integrating symbolic knowledge representations in the form of Knowledge Graph Embedding (KGE) to perform neuro-symbolic reasoning is an emerging research direction promising significant impacts. This vision paper seeks to position ideas and provoke early thoughts toward advancing neuro-symbolic artificial intelligence in the context of geospatial challenges. Specifically, it conceptualizes and elaborates on an architecture for infusing geospatial knowledge from knowledge graph in a deep neural network pipeline. As guiding case studies - land-use land-cover classification, object detection and instance segmentation can benefit from infusing spatio-contextual information with remote sensing imagery. The discussion further reflects on and articulates the challenges and explainable AI opportunities anticipated when scaling and maintaining large-scale geospatial knowledge graphs.

Potnis, Abhishek↗

Data-Driven Cyber-Attack Detection for PV Farms via Time-Frequency Domain Features

The internetworking of grid-connected power electronics converters (PECs) in photovoltaic (PV) farms has inevitably expanded the cyber-attack surfaces. Here this paper presents a comprehensive study on cyber-attack detection and diagnosis for PEC-enabled PV farms via single waveform sensor to distinguish between normal conditions, open-circuit faults, short-circuit faults, and cyber-attacks. To our knowledge, this has not been attempted before. Firstly, we propose frequency-domain magnitude-based residuals to identify short-circuit faults and a time-domain mean current vector-based feature to distinguish open-circuit faults from other threats. These features can fully reflect the specific physical characteristics of PV farms during threat duration. Secondly, unlike micro phasor measurement units (µPMU) and raw electric waveform-based methods, the proposed innovative features can address novel cyber-attacks that are excluded from the training process. Thirdly, an online hardware-in-the-loop (HIL) testbed using the OPAL-RT real-time digital simulator has verified the effectiveness. The monitoring system runs in real-time while using HIL as an operational solar farm and a National Instruments (NI) data acquisition card as the electric waveform sensor at the point of coupling.

42 ENGINEERING↗

Foundation Models of Scientific Knowledge for Chemistry: Opportunities, Challenges and Lessons Learned

Foundation models pre-trained on large corpora demonstrate significant gains across many natural language processing tasks and domains e.g., law, healthcare, education, etc. However, only limited efforts have investigated the opportunities and limitations of applying these powerful models to science and security applications. In this work we develop foundation models of scientific knowledge for chemistry to augment scientists with the advanced ability to perceive and reason at scale previously unimagined. Specifically, we build large-scale (1.47B parameter) general-purpose models for chemistry that can be effectively used to perform a wide range of in-domain and out-of-domain tasks. Evaluating these models in a zero-shot setting, we analyze the effect of model and data scaling, knowledge depth, and temporality on model performance in context of model training efficiency. Our novel findings demonstrate that (1) model size significantly contributes to the task performance when evaluated in a zero-shot setting; (2) data quality (aka diversity) affects model performance more than data quantity; (3) similarly, unlike previous work (Luu et al., 2021) temporal order of the documents in the corpus boosts model performance only for specific tasks, e.g., SciQ; and (4) models pre-trained from scratch perform better on in-domain tasks than those tuned from general-purpose models like Open AI’s GPT-2.

Foundation Models, Chemistry↗

Domain-aware Control-oriented Neural Models for Autonomous Underwater Vehicles

Conventional physics-based modeling is a time-consuming bottleneck in control design for complex nonlinear systems like autonomous underwater vehicles (AUVs). In contrast, purely data-driven models, require a large number of observations and lack operational guarantees for safety-critical systems. Data-driven models leveraging available partially characterized dynamics have potential to provide reliable systems models in a typical data-limited scenario for high value complex systems, thereby avoiding months of expensive expert modeling time. In this work we explore this middle-ground between expert-modeled and pure data-driven modeling. We present control-oriented parametric models with varying levels of domain-awareness that exploit known system structure and prior physics knowledge to create constrained deep neural dynamical system models. We employ universal differential equations to construct data-driven blackbox and graybox representations of the AUV dynamics. In addition, we explore a hybrid formulation that explicitly models the residual error related to imperfect graybox models. We compare the prediction performance of the learned models for different distributions of initial conditions and control inputs to assess their suitability for control.

Shaw Cortez, Wenceslao E.↗