SEARCH · Search NASA

Results for “Variable selection”

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.

SourceResearch connectionFrom yearThrough year

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

At least 325 records · Page 18

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Say where you sample: Increasing site selection transparency in urban ecology

Urban ecological studies have the potential to extend our understanding of socio-ecological systems beyond that of an individual city or region. Cross-comparative empirical work and synthesis are imperative to develop a general urban ecological theory. This can be achieved only if studies are replicable and generalizable. Transparency in methods reporting facilitates generalizability and replicability by documenting the decisions scientists make during the various steps of research design; this is particularly true for sampling design and selection because of their impact on both internal and external validity and the potential to unintentionally introduce bias. Three interdependent aspects of sample design are study sample selection (e.g., specific organisms, soils, or water), sample specification (measurement of specific variable of interest), and site selection (locations sampled). Of these, documentation of site selection—the where component of sample design—is underrepresented in the urban ecology literature. Using a stratified random sample of 158 papers from 12 major urban ecology journals, we investigated how researchers selected study sites in urban ecosystems and evaluated whether their site selection methods were transparent. We extracted data from these papers using a 50-question, theory-based questionnaire and a multiple-reviewer approach. Our sample represented almost 45 years of urban ecology research across 40 different countries. We found that more than 80% of the papers read were not transparent in their site selection methodology. We do not believe site selection methods are replicable for 70% of the papers read. Key weaknesses include incomplete descriptions of populations and sampling frames, urban gradients, sample selection methods, and property access. Low transparency in reporting the where methodology limits urban ecologists’ ability to assess the internal and external validity of studies’ findings and to replicate published studies; it also limits the generalizability of existing studies. The challenges of low transparency are particularly relevant in urban ecology, a field where standard protocols for site selection and delineation are still being developed. These limitations interfere with the fields’ ability to build theory and inform policy. We conclude by offering a set of recommendations to increase transparency, replicability, and generalizability.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Application of FARM to an IES scenario within the FORCE ecosystem

The FARM (Feasible Actuator Range Modifier) software module is a component of the RAVEN-based FORCE framework for analysis of Integrated Energy Systems (IES). FARM supports the HERON software module in the evaluation of the optimal dispatch by evaluating feasible set-points for the different IES unit components. Set-points are required to satisfy limits on both production variables (i.e., the variables to be optimized such as the electrical power, the hydrogen production rate, etc.) and process variables tied to the service life of equipment (e.g., steam flowrate, vessel pressure, turbine firing temperature, etc.). This problem is addressed by adopting a two-stage approach. First, the HERON power dispatcher determines set-points that meet the constraints on the former variables (e.g., power levels and power ramp rate limits). These constraints are called explicit constraints. Then, FARM adjusts these set-points to ensure the respect of the limits on the latter variables given the knowledge of the system physics acquired through machine learning algorithms. These constraints are called implicit constraints. From this standpoint, FARM constitutes a bridge between the HERON power dispatcher that adopts a simplified description of the IES unit (low-resolution physics) and the HYBRID high-fidelity models (high-resolution physics). The original version of the FARM software module (FARM-Alpha) was released by Argonne National Laboratory in January 2021. In the latest version of the code released in July 2022 (FARM-Delta), the Reference Governor (RG) algorithm was upgraded to a Multi-Input Multi-Output version from its original Single-Input Single-Output form. The RG algorithm acts to enforce constraints. With this improvement an IES unit is now treated as a single dynamic system from the standpoint of control. The crosstalk among components in an IES unit is now fully considered thereby ensuring a true optimization is obtained for those units that have multiple set-points. In this report, the capabilities of FARM-Delta operating within the FORCE ecosystem are demonstrated for an IES test case. The specific configuration of IES unit for this case was selected by the IES team with consultation from the Advanced Reactor IES Expert Group. A full TEA analysis that invoked HERON, HYBRID, FARM, and RAVEN was performed and serves to demonstrate how the latest modification to FARM algorithms (i.e., state variable selection, state-space matrices derivation, set-point verification) can shape setpoints that might otherwise compromise the health of equipment through accelerated wear and tear. In this specific test case, it was demonstrated that these algorithms ensure a more efficient utilization of steam resources to be shared by two different subsystems, namely Balance of Plant (BOP) and High-Temperature Steam Electrolysis (HTSE). Finally, some code improvements that can further enhance the user-friendliness are suggested.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Session 3: Agrivoltaics Pathways [Slides]

This presentation was developed for a webinar series for the USAID's South Asia Group for Energy (SAGE) and South Asia Regional Energy Partnership (SAREP). The presentation includes insights into agrivoltaic site assessment planning, including what is needed for a farm assessment, solar panel system design from the agricultural perspective, variables that go into crop selection, and environmental impact, sustainability, and agritourism, technical parameters for developing an agrivoltaics project, including what makes up a PV feasibility study, understanding variables that go into technology and equipment selection, agrivoltaic installation and agricultural integration, and monitoring and maintenance throughout the life of the project, and agrivoltaic financial planning, risk mitigation, and debt equity issues.

14 SOLAR ENERGY↗

NASA NTRS2005

Large Engine Technology Program. Task 22: Variable Geometry Concepts for Rich-Quench-Lean Combustors

The objective of the task reported herein was to define, evaluate, and optimize variable geometry concepts suitable for use with a Rich-Quench-Lean (RQL) combustor. The specific intent was to identify approaches that would satisfy High Speed Civil Transport (HSCT) cycle operational requirements with regard to fuel-air ratio turndown capability, ignition, and stability margin without compromising the stringent emissions, performance, and reliability goals that this combustor would have to achieve. Four potential configurations were identified and three of these were refined and tested in a high-pressure modular RQL combustor rig. The tools used in the evolution of these concepts included models built with rapid fabrication techniques that were tested for airflow characteristics to confirm sizing and airflow management capability, spray patternation, and atomization characterization tests of these models and studies that were supported by Computational Fluid Dynamics analyses. Combustion tests were performed with each of the concepts at supersonic cruise conditions and at other critical conditions in the flight envelope, including the transition points of the variable geometry system, to identify performance, emissions, and operability impacts. Based upon the cold flow characterization, emissions results, acoustic behavior observed during the tests and consideration of mechanical, reliability, and implementation issues, the tri-swirler configuration was selected as the best variable geometry concept for incorporation in the RQL combustor evolution efforts for the HSCT.

Tacina, Robert R.↗

DOE OSTI2024

Machine learning models of intermittent operation of RO wellhead water treatment for salinity reduction and nitrate removal

Machine learning models were developed for intermittent multi-mode operation of a wellhead reverse osmosis water purification and desalination system to predict salt passage, nitrate passage, and permeate flux. The models, based on long short-term memory (LSTM) recurrent neural network (RNN) architecture, included an attention mechanism to increase model performance in proximity of the regulatory limit for nitrate. Training and testing of the models for the Startup, Production, Shutdown and Flushing operational modes were based on operational data (consisting of 22 process variables per data sample) acquired every 2–5 s over a six-month period. The significant sets of model input attributes for the different operational modes were assessed via Spearman ranking correlation, Self-Organizing Map (SOM) analysis and feed forward feature selection (FFFS). Although the variability of nitrate passage, salt passage and permeate flux was significant over the four operational modes, prediction performance for the three outcomes were with R2 and Average Absolute Relative Error (AARE) of 0.78–0.95 and 2.96–6.16 %, respectively. Model updates post membrane elements replacement demonstrated similar levels of prediction accuracy. The study results suggest that there is merit in exploring the utility of multi-mode models for sensor fault detection, data imputation, and for potential use in model-predictive control.

Intermittent RO operation↗

NASA NTRS1991

Multichannel astrometric photometer-based parallaxes of evolved stars - Chi Cygni, 51 Andromedae, and OP Andromedae

Using data from Allegheny Observatory's Thaw/MAP astrometric system, absolute parallaxes are presented for Chi Cygni, 51 Andromedae, the newly discovered RS CVn variable OP Andromedae, and selected stars in these fields. Using existing angular diameter measures and the new parallaxes, linear diameters for Chi Cygni and 51 And are given. Chi Cygni has a perturbation in declination having a period of 0.75 yr and an amplitude of +/- 5 mas. Owing to the extreme variability of Chi Cygni, no further characterization of the system is possible at this time. Other results of spectral interest include the detection of a perturbation in the motion of BD +47 deg 466. This system was found to have a parallax of 14.4 +/- 1.2 mas.

Stein, John W.↗

DOE OSTI2025

Uncertainty estimation of bifurcated solutions in the Rayleigh–Bénard problem for advanced nuclear reactors applications

Multiphysics models of nuclear reactors frequently comprise nonlinear systems of equations. The nonlinear nature of these models could lead to solution bifurcations, where a small change in a certain parameter, e.g., the thermophysical properties of the coolant, can lead to a sudden change in the system’s behavior. At the point in parameter space where this happens, called a critical point, the Jacobian matrix of the model’s nonlinear operator becomes singular potentially permitting multiple solutions to coexist. In this paper, we perform uncertainty estimation (UE) in a parameter range that includes bifurcated solutions within the context of Rayleigh–Bénard problem. We perform this analysis assuming uncertain temperature difference, and tilt angle for the iterative solution algorithm with a unit Prandtl number (Pr = 1). Also, we perform this analysis under uncertain thermophysical properties for both FLiBe molten salt and liquid sodium as working fluid. We deploy two approaches to compute statistical moments for the resulting distributions of selected flow-field variables. The first approach is the blind computation of the mean and the standard deviation without any consideration of solution bifurcation, while the second approach utilizes k-means clustering to cluster each branch’s solutions together and compute separate statistical moments for each branch. The statistical distributions are obtained by perturbing the selected parameters about nominal values that correspond to a solution on one of the valid branches, and that solution is used as initial guess for the iterative solution algorithm. We found that perturbation of any parameter when its nominal value is close to its critical point always leads to branch jumping, i.e., the iterations converge to a solution on a branch different from the branch of the initial guess. This produces a statistical ensemble comprised of fundamentally different solutions leading to wrong mean values and uncertainty estimates, whereas clustering provides an efficient way to deal with this type of computation. This work is important for developing Gen IV nuclear systems because many of these systems rely on natural convection for cooling especially in accident conditions.

97 - MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Tailoring Hierarchical Structure and Rare Earth Affinity of Compositionally Identical Polymers via Sequence Control

Macromolecule sequence, structure, and function are inherently intertwined. While well-established relationships exist in proteins, they are more challenging to define for synthetic polymer nanoparticles due to their molecular weight, sequence, and conformational dispersities. Furthermore, to explore the impact of sequence on nanoparticle structure, we synthesized a set of 16 compositionally identical, sequence-controlled polymers with distinct monomer patterning of dimethyl acrylamide and a bioinspired, structure-driving di(phenylalanine) acrylamide (FF). Sequence control was achieved through multiblock polymerizations, yielding unique ensembles of polymer sequences which were simulated by kinetic Monte Carlo simulations. Systematic analysis of the global (tertiary- and quaternary-like) structure in this amphiphilic copolymer series revealed the effect of multiple sequence descriptors: the number of domains, the hydropathy of terminal domains, and the patchiness (density) of FF within a domain, each of which impacted both chain collapse and the distribution of single- and multichain assemblies. Furthermore, both the conformational freedom of chain segments and local-scale, β-sheet-like interactions were sensitive to the patchiness of FF. To connect sequence, structure, and target function, we evaluated an additional series of nine sequence-controlled copolymers as sequestrants for rare earth elements (REEs) by incorporating a functional acrylic acid monomer into select polymer scaffolds. We identified key sequence variables that influence the binding affinity, capacity, and selectivity of the polymers for REEs. Collectively, these results highlight the potential of and boundaries of sequence control via multiblock polymerizations to drive primary sequence ensembles hierarchical structures, and ultimately the functionality of compositionally identical polymeric materials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Using low volume eDNA methods to sample pelagic marine animal assemblages

Environmental DNA (eDNA) is an increasingly useful method for detecting pelagic animals in the ocean but typically requires large water volumes to sample diverse assemblages. Ship-based pelagic sampling programs that could implement eDNA methods generally have restrictive water budgets. Studies that quantify how eDNA methods perform on low water volumes in the ocean are limited, especially in deep-sea habitats with low animal biomass and poorly described species assemblages. Using 12S rRNA and COI gene primers, we quantified assemblages comprised of micronekton, coastal forage fishes, and zooplankton from low volume eDNA seawater samples (n = 436, 380–1800 mL) collected at depths of 0–2200 m in the southern California Current. We compared diversity in eDNA samples to concurrently collected pelagic trawl samples (n = 27), detecting a higher diversity of vertebrate and invertebrate groups in the eDNA samples. Differences in assemblage composition could be explained by variability in size-selectivity among methods and DNA primer suitability across taxonomic groups. The number of reads and amplicon sequences variants (ASVs) did not vary substantially among shallow (<200 m) and deep samples (>600 m), but the proportion of invertebrate ASVs that could be assigned a species-level identification decreased with sampling depth. Using hierarchical clustering, we resolved horizontal and vertical variability in marine animal assemblages from samples characterized by a relatively low diversity of ecologically important species. Low volume eDNA samples will quantify greater taxonomic diversity as reference libraries, especially for deep-dwelling invertebrate species, continue to expand.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

Data for Yield from Iowa’s first commercial miscanthus fields: implications of spatial variability for productivity and sustainability beyond research plots

This dataset contains biomass yield measurements and associated vegetation index data collected from commercial Miscanthus × giganteus fields in eastern Iowa during the 2022–2023 growing seasons. The data support the analyses presented in the article: “Yield From Iowa's First Commercial Miscanthus Fields: Implications of Spatial Variability for Productivity and Sustainability Beyond Research Plots.” We collected 105 ground-truth biomass samples from four mature commercial fields (>4 years old) covering 92.81 ha. Samples were taken from 3 m² quadrats that were hand-harvested in alignment with commercial harvest timing. Stem biomass (excluding leaves) was weighed, moisture-corrected, and converted to dry-matter yield expressed in Mg DM ha⁻¹. Sampling locations were selected to capture spatial variability visible in aerial imagery and were recorded using RTK GPS. Each biomass observation was paired with vegetation indices derived from high-resolution PlanetScope satellite imagery (3 m resolution). Images were acquired throughout the growing season, and indices were calculated to evaluate their ability to predict end-of-season biomass yield. Statistical and machine learning approaches were used to identify key predictors, and a linear regression model based on end-of-July Green Normalized Difference Vegetation Index (GNDVI) was developed and evaluated. This repository includes the data used in that modeling workflow. Management practices, economic data, full imagery time series, and additional methodological details are described in the associated publication and are not included here. The dataset consists of three comma-separated value (CSV) files: 1. Combine_Groundtruth_Yield_VI_22_23.csv This file contains ground-truth biomass yield measurements and associated key vegetation index values collected during the 2022 and 2023 growing seasons. Rows: 105 observations Columns: Year — Year of observation (2022 or 2023) Field — Field location identifier Sample_number — Unique sample identifier GNDVI_End_Jul — Green Normalized Difference Vegetation Index calculated at end of July GNDVI_End_Aug — Green Normalized Difference Vegetation Index calculated at end of August NDRE_End_Aug — Normalized Difference Red Edge index calculated at end of August Biomass_Stem_Yield_MgDM/ha — Measured stem biomass yield (megagrams dry matter per hectare) 2. trainData_GNDVI.csv This file contains the subset of observations used to train the predictive relationship between July GNDVI and biomass yield. Rows: 76 observations Columns: Unnamed: 0 — Row index retained from the original data processing workflow GNDVI_End_Jul — GNDVI at end of July Stem_Yield_MgDM/ha — Observed stem biomass yield (Mg DM ha⁻¹) 3. testData_GNDVI.csv This file contains the test dataset used to evaluate model performance. Rows: 29 observations Columns: Unnamed: 0 — Row index retained from the original data processing workflow GNDVI_End_Jul — GNDVI at end of July Predicted_Yield_MgDM/ha — Model-predicted stem biomass yield (Mg DM ha⁻¹) Observed_Yield_MgDM/ha — Measured stem biomass yield (Mg DM ha⁻¹)

Potential yield, yield gap, in-field management, y↗

NASA NTRS1991

General shape optimization capability

A method is described for calculating shape sensitivities, within MSC/NASTRAN, in a simple manner without resort to external programs. The method uses natural design variables to define the shape changes in a given structure. Once the shape sensitivities are obtained, the shape optimization process is carried out in a manner similar to property optimization processes. The capability of this method is illustrated by two examples: the shape optimization of a cantilever beam with holes, loaded by a point load at the free end (with the shape of the holes and the thickness of the beam selected as the design variables), and the shape optimization of a connecting rod subjected to several different loading and boundary conditions.

Chargin, Mladen K.↗

NASA NTRS2016

Optimization of Variable-Depth Liner Configurations for Increased Broadband Noise Reduction

This paper employs three acoustic propagation codes to explore variable-depth liner configurations for the NASA Langley Grazing Flow Impedance Tube (GFIT). The initial study demonstrates that a variable impedance can acceptably be treated as a uniform impedance if the spatial extent over which this variable impedance occurs is less than one-third of a wavelength of the incident sound. A constrained optimization study is used to design a variable-depth liner and to select an optimization metric. It also provides insight regarding how much attenuation can be achieved with variable-depth liners. Another optimization study is used to design a liner with much finer chamber depth resolution for the Mach 0.0 and 0.3 test conditions. Two liners are designed based on spatial rearrangement of chambers from this liner to determine whether the order is critical. Propagation code predictions suggest this is not the case. Both liners are fabricated via additive manufacturing and tested in the GFIT for the Mach 0.0 condition. Predicted and measured attenuations compare favorably across the full frequency range. These results clearly suggest that the chambers can be arranged in any order, thus offering the potential for innovative liner designs to minimize depth and weight.

Jones, M. G.↗

NASA NTRS2010

Analysis of Sting Balance Calibration Data Using Optimized Regression Models

Calibration data of a wind tunnel sting balance was processed using a candidate math model search algorithm that recommends an optimized regression model for the data analysis. During the calibration the normal force and the moment at the balance moment center were selected as independent calibration variables. The sting balance itself had two moment gages. Therefore, after analyzing the connection between calibration loads and gage outputs, it was decided to choose the difference and the sum of the gage outputs as the two responses that best describe the behavior of the balance. The math model search algorithm was applied to these two responses. An optimized regression model was obtained for each response. Classical strain gage balance load transformations and the equations of the deflection of a cantilever beam under load are used to show that the search algorithm s two optimized regression models are supported by a theoretical analysis of the relationship between the applied calibration loads and the measured gage outputs. The analysis of the sting balance calibration data set is a rare example of a situation when terms of a regression model of a balance can directly be derived from first principles of physics. In addition, it is interesting to note that the search algorithm recommended the correct regression model term combinations using only a set of statistical quality metrics that were applied to the experimental data during the algorithm s term selection process.

Ulbrich, N.↗

DOE OSTI2020

Follow-up Studies of Five Cataclysmic Variable Candidates Discovered by LAMOST

We report follow-up observations of five cataclysmic variable candidates from the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) published by Hou et al. LAMOST J024048.51+195226.9 is the most unusual of the five; an early-M type secondary star contributes strongly to its spectrum, and its spectral and photometric behavior are strikingly reminiscent of the hitherto-unique propeller system AE Aqr. We confirm that a 7.34 hr period discovered in the Catalina survey data is orbital. Another object, LAMOST J204305.95+341340.6, appears to be a near twin of the novalike variable V795 Her, with an orbital period in the so-called 2–3 hr “gap.” LAMOST J035913.61+405035.0 is evidently an eclipsing, weakly outbursting dwarf nova with a 5.48 hr period. Our spectrum of LAMOST J090150.09+375444.3 is dominated by a late-type secondary and shows weak, narrow Balmer emission moving in phase with the absorption lines, but at lower amplitude; we do not see the He ii λ4686 emission evident in the published discovery spectrum. We again confirm that a period from the Catalina data, in this case 6.80 hr, is orbital. LAMOST J033940.98+414805.7 yields a radial-velocity period of 3.54 hr, and its spectrum appears to be typical of novalike variables in this period range. The spectroscopically selected sample from LAMOST evidently includes some interesting cataclysmic variables that have been unrecognized until now, apparently because of the relatively modest range of their photometric variations.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2021

ThickBrick: optimal event selection and categorization in high energy physics. Part I. Signal discovery

We provide a prescription called ThickBrick to train optimal machine-learning-based event selectors and categorizers that maximize the statistical significance of a potential signal excess in high energy physics (HEP) experiments, as quantified by any of six different performance measures. For analyses where the signal search is performed in the distribution of some event variables, our prescription ensures that only the information complementary to those event variables is used in event selection and categorization. This eliminates a major misalignment with the physics goals of the analysis (maximizing the significance of an excess) that exists in the training of typical ML-based event selectors and categorizers. In addition, this decorrelation of event selectors from the relevant event variables prevents the background distribution from becoming peaked in the signal region as a result of event selection, thereby ameliorating the challenges imposed on signal searches by systematic uncertainties. Our event selectors (categorizers) use the output of machine-learning-based classifiers as input and apply optimal selection cutoffs (categorization thresholds) that are functions of the event variables being analyzed, as opposed to flat cutoffs (thresholds). These optimal cutoffs and thresholds are learned iteratively, using a novel approach with connections to Lloyd’s k-means clustering algorithm. We provide a public, Python implementation of our prescription, also called ThickBrick, along with usage examples.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2023

Distributed out-of-memory NMF on CPU/GPU architectures

We propose an efficient distributed out-of-memory implementation of the non-negative matrix factorization (NMF) algorithm for heterogeneous high-performance-computing systems. The proposed implementation is based on prior work on NMFk, which can perform automatic model selection and extract latent variables and patterns from data. In this work, we extend NMFk by adding support for dense and sparse matrix operation on multi-node, multi-GPU systems. The resulting algorithm is optimized for out-of-memory problems where the memory required to factorize a given matrix is greater than the available GPU memory. Memory complexity is reduced by batching/tiling strategies, and sparse and dense matrix operations are significantly accelerated with GPU cores (or tensor cores when available). Input/output latency associated with batch copies between host and device is hidden using CUDA streams to overlap data transfers and compute asynchronously, and latency associated with collective communications (both intra-node and inter-node) is reduced using optimized NVIDIA Collective Communication Library (NCCL) based communicators. Benchmark results show significant improvement, from 32X to 76x speedup, with the new implementation using GPUs over the CPU-based NMFk. Good weak scaling was demonstrated on up to 4096 multi-GPU cluster nodes with approximately 25,000 GPUs when decomposing a dense 340 Terabyte-size matrix and an 11 Exabyte-size sparse matrix of density 10 -6 .

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Data-Driven Mapping of the Cesium Cadmium Bromide Phase Space Utilizing a Soft-Chemistry Approach

Soft-chemistry techniques provide a versatile approach to synthesizing inorganic materials under mild conditions, enabling access to compositions and structures that are challenging to achieve through traditional thermodynamically driven solid-state methods. However, these solution-based routes often result in phase competition, requiring precise control over reaction conditions to achieve selective product formation. While one-variable-at-a-time (OVAT) approaches have traditionally been used for phase selection, data-driven strategies are emerging as more efficient methods for navigating complex synthetic spaces. Ternary metal halides, such as cesium cadmium bromides (Cs–Cd–Br), are of growing interest due to their potential in wide and ultrawide band gap applications. Unlike the well-studied cesium lead halide phases, the compositional diversity and solution-based synthesis of ternary Cs–Cd–Br phases remain largely unexplored. This study systematically investigates the synthetic phase space of the Cs–Cd–Br system by constructing a data-driven phase map. Using a common set of precursors and a standardized experimental procedure, we successfully synthesize all four known Cs–Cd–Br phases—CsCdBr 3 , Cs 2 CdBr 4 , Cs 3 CdBr 5 , and Cs 7 Cd 3 Br 13 —each exhibiting distinct structures, morphologies, and optical properties. Our findings highlight the potential of soft-chemistry methods for expanding the library of ternary metal halides and provide key insights into the thermodynamic and kinetic factors governing phase formation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Protein–Protein Interaction Networks Derived from Classical and Machine Learning-Based Natural Language Processing Tools

The study of protein-protein interactions (PPIs) provides insight into various biological mechanisms, including the binding of antibodies to antigens, enzymes to inhibitors or promoters, and receptors to ligands. Recent studies of PPIs have led to significant biological breakthroughs. For example, the study of PPIs involved in the human:SARS-CoV-2 viral infection mechanism aided in the development of the SARS-CoV-2 vaccines. Though several databases exist for the manual curation of PPI networks, text mining methods have been routinely demonstrated as useful alternatives for newly studied or understudied species where databases are incomplete. Here, the relationship extraction (RE) performance of several open-source classical text processing, machine learning (ML)-based natural language processing (NLP), and large language model (LLM)-based NLP tools were compared. Overall, our results indicated that networks derived from classical methods tend to have high true positive rates at the expense of having overconnected-networks, ML-based NLP methods have lower true positive rates but networks with the closest structures to the target network, and LLM-based NLP methods tend to exist in-between the two other approaches, with variable performances. Finally, the selection of a specific NLP approach should be tied to the needs of a study and text availability, as models varied in performance due to the amount of text provided.

59 BASIC BIOLOGICAL SCIENCES↗