SEARCH · Search NASA

Results for “Model interpretation”

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 793 records · Page 44

All sources NASA NTRS DOE OSTI

NASA NTRS1993

bpshape wk4: a computer program that implements a physiological model for analyzing the shape of blood pressure waveforms

We describe the theory and computer implementation of a newly-derived mathematical model for analyzing the shape of blood pressure waveforms. Input to the program consists of an ECG signal, plus a single continuous channel of peripheral blood pressure, which is often obtained invasively from an indwelling catheter during intensive-care monitoring or non-invasively from a tonometer. Output from the program includes a set of parameter estimates, made for every heart beat. Parameters of the model can be interpreted in terms of the capacitance of large arteries, the capacitance of peripheral arteries, the inertance of blood flow, the peripheral resistance, and arterial pressure due to basal vascular tone. Aortic flow due to contraction of the left ventricle is represented by a forcing function in the form of a descending ramp, the area under which represents the stroke volume. Differential equations describing the model are solved by the method of Laplace transforms, permitting rapid parameter estimation by the Levenberg-Marquardt algorithm. Parameter estimates and their confidence intervals are given in six examples, which are chosen to represent a variety of pressure waveforms that are observed during intensive-care monitoring. The examples demonstrate that some of the parameters may fluctuate markedly from beat to beat. Our program will find application in projects that are intended to correlate the details of the blood pressure waveform with other physiological variables, pathological conditions, and the effects of interventions.

Non-NASA Center↗

DOE OSTI2020

Prediction of Hydroxymethylfurfural Yield in Glucose Conversion through Investigation of Lewis Acid and Organic Solvent Effects

Hydroxymethylfurfural (HMF) is one of the important renewable platform compounds that can be obtained from biomass feedstocks through glucose conversion catalyzed by Brønsted and Lewis acids. However, it is challenging to enhance the HMF yield due to side reactions. In this study, a systematic approach combining theory and experiment was performed to investigate the influence of Lewis acids and organic solvents on the HMF yield. For the Lewis acid effect, a relationship between chemical hardness and experimental HMF yields was found in the rate-limiting step of glucose-to-fructose isomerization for six metal chlorides; HMF production was promoted when the metal chloride and a substrate had a similar chemical hardness. To study the organic solvent effect, a multivariate model was developed based on the insights gained from the mechanistic study of fructose dehydration, to predict HMF yields in a given water-organic cosolvent system. It showed a reliable accuracy in evaluating HMF yields with a mean absolute error (MAE) of 3.0% with respect to experimental HMF yields for 13 solvents, and also predicted HMF yields with a MAE of 10.7% for four new solvents. Chemical interpretation of the model revealed that it is desirable to use a solvent capable of stabilizing the carbocation intermediates with low proton transfer activity and high hydrogen bond basicity, to maximize the HMF yield. This multivariate model informs experimentalists about rational selection of solvents with very low computational costs needed to calculate only six variables for each solvent. It can be expanded to other catalytic systems such as heterogeneous Brønsted–Lewis bifunctional catalysts and enables optimization of reaction conditions to obtain other useful platform molecules through biomass conversion.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2019

Validation of Wichner Predictive Model for Chronic Oxidation by Moisture of Nuclear Graphite

This document reports the results of multi-year extensive research on chronic oxidation of nuclear graphite by traces of moisture that might be present in the helium coolant of high-temperature gas-cooled reactors (HTGRs). The research focuses on four grades of nuclear graphite that are considered potential candidates for component fabrication in advanced nuclear reactors. The four grades are (1) PCEA (medium-fine grain, extruded, manufactured by GrafTech Inc., USA); (2) NBG-17 (medium-fine grain, vibrationally molded, manufactured by SGL Carbon, Germany/France); (3) IG-110 (microfine grain, isostatically molded, manufactured by Toyo Tanso, Japan); and (4) 2114 (microfine grain, isostatically pressed, manufactured by Mersen, USA). The experimental part of this research comprised three objectives that were consistently pursued for each graphite material: (1) accelerated oxidation tests in well-controlled conditions (temperature, gas composition) and analysis of chemical kinetic models and of their parameters for each graphite grade; (2) measurements of water vapor transport rates from H 2 O/He mixtures through thin graphite slabs for estimation of the water effective diffusivity specific to each material; (3) controlled oxidation of graphite specimens in H 2 O/He mixtures and characterization of the effects of oxidation temperatures on the density profiles of oxidized layers. These three groups of experimental results were interpreted using the model developed by Wichner et al. for predicting the penetration depth of oxidants (moisture or oxygen) as a function of oxidation conditions. Basically, starting from the kinetic and transport properties of each graphite grade, the ultimate goal was predicting the stable density profile in the oxidized layer and its dependence on the oxidation temperature. The profiles obtained this way were then compared against direct observations of oxidized layers in specimens obtained in well-controlled oxidation conditions. The general agreement between predictions and observations validated the general applicability of the Wichner model. The results obtained in this project were presented at annual Department of Energy–Advanced Reactor Technologies program reviews and at specialist meetings and conferences and were published in Oak Ridge National Laboratory (ORNL) reports and in professional journals. The results published in 2018 had an immediate impact in 2019, when a group from three US universities used ORNL kinetic data as input information in a complex simulation of chronic oxidation by moisture using the same graphite grades during long-time operation in a modular high-temperature gas-cooled reactor. Their conclusion, in line with the conclusions obtained at the end of this project, was that chronic graphite oxidation by moisture during normal operation conditions of an HTGR will be limited to a thin surface layer of graphite blocks at the reactor bottom, where the highest temperatures are. That will not affect the integrity of graphite components and reactor safety, provided the impurity concentration in the helium coolant is rigorously controlled and maintained at the design specifications.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2018

Search for new physics in final states with an energetic jet or a hadronically decaying $W$ or $Z$ boson and transverse momentum imbalance at $\sqrt{s}=13\text{ }\text{ }\mathrm{TeV}$

A search for new physics using events containing an imbalance in transverse momentum and one or more energetic jets arising from initial-state radiation or the hadronic decay of W or Z bosons is presented. A data sample of proton-proton collisions at s=13 TeV, collected with the CMS detector at the LHC and corresponding to an integrated luminosity of 35.9 fb-1, is used. The observed data are found to be in agreement with the expectation from standard model processes. The results are interpreted as limits on the dark matter production cross section in simplified models with vector, axial-vector, scalar, and pseudoscalar mediators. Interpretations in the context of fermion portal and nonthermal dark matter models are also provided. In addition, the results are interpreted in terms of invisible decays of the Higgs boson and set stringent limits on the fundamental Planck scale in the Arkani-Hamed, Dimopoulos, and Dvali model with large extra spatial dimensions.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2023

Carbon isotope fractionation by an ancestral rubisco suggests that biological proxies for CO 2 through geologic time should be reevaluated

The history of Earth’s carbon cycle reflects trends in atmospheric composition convolved with the evolution of photosynthesis. Fortunately, key parts of the carbon cycle have been recorded in the carbon isotope ratios of sedimentary rocks. The dominant model used to interpret this record as a proxy for ancient atmospheric CO 2 is based on carbon isotope fractionations of modern photoautotrophs, and longstanding questions remain about how their evolution might have impacted the record. Therefore, we measured both biomass (ε p ) and enzymatic (ε Rubisco ) carbon isotope fractionations of a cyanobacterial strain ( Synechococcus elongatus PCC 7942) solely expressing a putative ancestral Form 1B rubisco dating to ≫1 Ga. This strain, nicknamed ANC, grows in ambient pCO 2 and displays larger ε p values than WT, despite having a much smaller ε Rubisco (17.23 ± 0.61‰ vs. 25.18 ± 0.31‰, respectively). Surprisingly, ANC ε p exceeded ANC ε Rubisco in all conditions tested, contradicting prevailing models of cyanobacterial carbon isotope fractionation. Such models can be rectified by introducing additional isotopic fractionation associated with powered inorganic carbon uptake mechanisms present in Cyanobacteria, but this amendment hinders the ability to accurately estimate historical pCO 2 from geological data. Understanding the evolution of rubisco and the CO 2 concentrating mechanism is therefore critical for interpreting the carbon isotope record, and fluctuations in the record may reflect the evolving efficiency of carbon fixing metabolisms in addition to changes in atmospheric CO 2 .

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS2002

Mass Conservation in a Chemical Transport Model and its Effect on CO2 and SF6 Simulations

Chemical transport models (CTMs) must conserve mass to be useful for applications involving assessment of the effect of various pollutants on the troposphere and stratosphere. Furthermore, calculations of the evolution of constituents such as SF6 are used to evaluate overall model transport, and interpretation of such simulations is clouded if mass conservation is not assured. For realistic simulations or predictions, it is crucial that constituents are not produced or lost by transport or other processes in the CTMs. Analysis of CO2 and SF6 experiments using a CTM shows that problems with mass conservation can seriously degrade the simulations. Failure to conserve mass results from inconsistency of the surface pressure tendency and the divergence of horizontal mass flux when the model is forced by assimilated meteorological data. We have developed an effective method to eliminate the inconsistency by modifying the divergent part of the wind field. The changes in the wind fields are quite small but the impact on mass conservation is large. Parameterizations of physical processes such as convection or turbulent transport can also affect mass conservation. The lack of conservation is small but accumulates when integrations are lengthy such as required for SF6. This lack of conservation is found using winds from either a GCM or from an assimilation system. A simple adjustment removes much of the inaccuracy in the convective parameterization. A CO2 simulation using assimilated winds from the most recent version of the Goddard Earth Observing System Data Assimilation System will be used to illustrate the impact of these transport improvements.

Zhu, Z.↗

DOE OSTI2025

Interpretation of Ion Irradiation and Neutron Irradiation Damage in Additively Manufactured 316 Stainless Steel using Multiscale Modeling

The accelerated adoption of nuclear energy necessitates advanced manufacturing technologies, such as additive manufacturing, to meet heightened supply chain requirements and support innovative reactor technologies. Due to the unique microstructural characteristics of additively manufactured materials under distinct solidification conditions, comprehensive evaluation of their performance in reactor environments is essential. The Advanced Materials and Manufacturing Technologies program under the Department of Energy's Office of Nuclear Energy focuses on understanding the irradiation performance and damage evolution of laser powder bed fusion 316 stainless steel, with an emphasis on integrating ion and neutron irradiation data to accelerate the development and qualification of materials for advanced nuclear reactor applications. While ion irradiation is a cost- and time-effective method, modeling and simulation are required to interpret the data for the broader range of irradiation conditions encountered in advanced reactors. In fiscal year 2025, integrated multiscale modeling and simulations were conducted to assess irradiation damage in additively manufactured 316 stainless steel. Key outcomes include predictions of chromium enrichment at grain boundaries, nickel enrichment at dislocation cell walls and void surfaces, and heterogeneous void evolution under ion and neutron irradiation conditions. Cluster dynamics simulations revealed the coarsening of voids at high irradiation temperatures and the suppression of void growth by high network dislocation density, while also demonstrating significant growth and coarsening of voids and self-interstitial atom loops at low dose rates. Machine learning-accelerated atomistic simulations highlighted the impact of the local environment and chromium concentration on vacancy diffusivity, providing key insights on the influence of composition on void swelling and radiation-induced segregation. Additionally, molecular dynamics simulations demonstrated the presence of defect production bias and a significant effect of carbon content on defect cluster behavior. These combined efforts aim to predict the performance of additively manufactured materials under various reactor conditions, supporting their qualification for nuclear reactor applications by interpreting ion irradiation data. This report underscores the potential of integrated multiscale modeling to analyze ion irradiation data in the effort to accelerate the qualification of additively manufactured materials for nuclear reactor components.

316 stainless steel↗

DOE OSTI2018

Comparative Study of Finite Element Methods Using the Time-Accuracy-Size(TAS) Spectrum Analysis

We present a performance analysis appropriate for comparing algorithms using different numerical discretizations. By taking into account the total time-to-solution, numerical accuracy with respect to an error norm, and the computation rate, a cost-benefit analysis can be performed to determine which algorithm and discretization are particularly suited for an application. This work extends the performance spectrum model in for interpretation of hardware and algorithmic tradeoffs in numerical PDE simulation. As a proof-of-concept, popular finite element software packages are used to illustrate this analysis for Poisson’s equation.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Charge Radii of Neutron-Rich Scandium Isotopes and the Seniority Symmetry in the 0⁢𝑓 7/2 Shell

Nuclear charge radii of neutron-rich 47–49 Sc isotopes were measured using collinear laser spectroscopy at CERN-ISOLDE. The new data reveal that the charge radii of scandium isotopes exhibit a distinct trend between 𝑁 = 20 and 𝑁 = 28, with 41 Sc and 49 Sc isotopes having similar values, mirroring the closeness of the charge radii of 40 Ca and 48 Ca. Theoretical models that successfully interpret the radii of calcium isotopes could not account for the observed behavior in scandium radii, in particular the reduced odd-even staggering. Remarkably, the inclusion of the new 49 Sc radius data has unveiled a similar trend in the charge radii of 𝑁 = 28 isotones and 𝑍 = 20 isotopes when adding the protons atop the 48 Ca core and the neutrons atop the 40 Ca core, respectively. We demonstrate that this trend is consistent with the prediction of the seniority model.

Nuclear charge distribution↗

NASA NTRS1992

SAVS: A Space Analysis and Visualization System

We propose to develop, test, demonstrate, and deliver to NASA a powerful and versatile data acquisition, manipulation, analysis and visualization system which will enhance scientific capabilities in the display and interpretation of diverse and distributed data within an integrated user-friendly environment. Our approach exploits existing technologies and combines three major elements into an easy-to-use interactive package: (1) innovative visualization software; (2) advanced database techniques; and (3) a rich set of mathematical and image processing tools. Visualization capabilities will include one-, two-, and three-dimensional displays, along with animation, compression, warping and slicing functions. Analysis tools will include generic mathematical and statistical techniques along with the ability to use large scale models for interactive interpretation of large volume data sets. Our system will be implemented on Sun and DEC UNIX workstations and on the Stardent Graphics Super computer. Our final deliverable will include complete documentation and a NASA/NSF-CDAW/SUNDIAL campaign demonstration.

Szuszczewicz, Edward P.↗

NASA NTRS1993

SAVS: A Space Analysis and Visualization System

We propose to develop, test, demonstrate, and deliver to NASA a powerful and versatile data acquisition, manipulation, analysis and visualization system which will enhance scientific capabilities in the display and interpretation of diverse and distributed data within an integrated user-friendly environment. Our approach exploits existing technologies and combines three major elements into an easy-to-use interactive package: (1) innovative visualization software; (2) advanced database techniques; and (3) a rich set of mathematical and image processing tools. Visualization capabilities will include one-, two-, and three-dimensinal displays, along with animation, compression, warping and slicing functions. Analysis tools will include generic mathematical and statistical techniques along with the ability to use large scale models for interactive interpretation of large volume data sets. Our system will be implemented on Sun and EC UNIX workstations and on the Stardent Graphics Supercomputer. Our final deliverable will include complete documentation and a NASA/NSF-CDAW/SUNDIAL campaign demonstration.

Szuszczewicz, Edward P.↗

NASA NTRS2001

Low Temperature Reflectance Spectra of Titan Tholins

Compositional interpretation of remotely obtained reflectance spectra of outer solar system surfaces is achieved by a variety of methods. These include matching spectral curves, matching spectral features, quantitative spectral interpretation, and theoretical modeling of spectra. All of these approaches rely upon laboratory measurements of one kind or another. The bulk of these laboratory measurements are obtained with the sample of interest at ambient temperatures and pressures. However, surface temperatures of planets, satellites, and asteroids in the outer solar system are significantly cooler than ambient laboratory conditions on Earth. The infrared spectra of many materials change as a function of temperature. As has been recently demonstrated it is important to assess what effects colder temperatures have on spectral properties and hence, compositional interpretations. Titan tholin is a solid residue created by energetic processing of H-, C-, and N-bearing gases. Such residues can also be created by energetic processing if the gases are condensed into ices. Titan tholin has been suggested as a coloring agent for several surfaces in the outer solar system. Here we report laboratory measurements of Titan tholin at a temperature of 100 K and compare these to measurements of the same sample near room temperature. At low temperature the absorption features beyond 1 micrometer narrow slightly. At wavelengths greater than approx. 0.8 micrometer the overall reflectance of the sample decreases slightly making the sample less red at low temperatures. We will discuss the implications of the laboratory measurements for interpretation of cold outer solar system surfaces.

Roush, T. L.↗

NASA NTRS2013

Galactic Cosmic Ray Event-Based Risk Model (GERM) Code

This software describes the transport and energy deposition of the passage of galactic cosmic rays in astronaut tissues during space travel, or heavy ion beams in patients in cancer therapy. Space radiation risk is a probability distribution, and time-dependent biological events must be accounted for physical description of space radiation transport in tissues and cells. A stochastic model can calculate the probability density directly without unverified assumptions about shape of probability density function. The prior art of transport codes calculates the average flux and dose of particles behind spacecraft and tissue shielding. Because of the signaling times for activation and relaxation in the cell and tissue, transport code must describe temporal and microspatial density of functions to correlate DNA and oxidative damage with non-targeted effects of signals, bystander, etc. These are absolutely ignored or impossible in the prior art. The GERM code provides scientists data interpretation of experiments; modeling of beam line, shielding of target samples, and sample holders; and estimation of basic physical and biological outputs of their experiments. For mono-energetic ion beams, basic physical and biological properties are calculated for a selected ion type, such as kinetic energy, mass, charge number, absorbed dose, or fluence. Evaluated quantities are linear energy transfer (LET), range (R), absorption and fragmentation cross-sections, and the probability of nuclear interactions after 1 or 5 cm of water equivalent material. In addition, a set of biophysical properties is evaluated, such as the Poisson distribution for a specified cellular area, cell survival curves, and DNA damage yields per cell. Also, the GERM code calculates the radiation transport of the beam line for either a fixed number of user-specified depths or at multiple positions along the Bragg curve of the particle in a selected material. The GERM code makes the numerical estimates of basic physical and biophysical quantities of high-energy protons and heavy ions that have been studied at the NASA Space Radiation Laboratory (NSRL) for the purpose of simulating space radiation biological effects. In the first option, properties of monoenergetic beams are treated. In the second option, the transport of beams in different materials is treated. Similar biophysical properties as in the first option are evaluated for the primary ion and its secondary particles. Additional properties related to the nuclear fragmentation of the beam are evaluated. The GERM code is a computationally efficient Monte-Carlo heavy-ion-beam model. It includes accurate models of LET, range, residual energy, and straggling, and the quantum multiple scattering fragmentation (QMSGRG) nuclear database.

Cucinotta, Francis A.↗

DOE OSTI2018

Exact excited states of nonintegrable models

We discuss a method of numerically identifying exact energy eigenstates for a finite system, whose form can then be obtained analytically. We demonstrate our method by identifying and deriving exact analytic expressions for several excited states, including an infinite tower, of the one-dimensional spin-1 Affleck-Kennedy-Lieb-Tasaki (AKLT) model, a celebrated nonintegrable model. The states thus obtained for the AKLT model can be interpreted as from one to an extensive number of quasiparticles on the ground state or on the highest excited state when written in terms of dimers. Included in these exact states is a tower of states spanning energies from the ground state to the highest excited state. Here, some of the states of the tower appear to be in the bulk of the energy spectrum, allowing us to make conjectures on the strong eigenstate thermalization hypothesis. We also generalize these exact states including the tower of states to the generalized integer spin AKLT models. Furthermore, we establish a correspondence between some of our states and those of the Majumdar-Ghosh model, yet another nonintegrable model, and extend our construction to the generalized integer spin AKLT models.

1-dimensional spin chains↗

DOE OSTI2017

Molecular dynamics simulations of NMR relaxation and diffusion of bulk hydrocarbons and water

Molecular dynamics (MD) simulations are used to investigate 1 H nuclear magnetic resonance (NMR) relaxation and diffusion of bulk n-C 5 H 12 to n-C 17 H 36 hydrocarbons and bulk water. The MD simulations of the 1H NMR relaxation times T 1,2 in the fast motion regime where T 1 = T 2 agree with measured (de-oxygenated) T2 data at ambient conditions, without any adjustable parameters in the interpretation of the simulation data. Likewise, the translational diffusion DT coefficients calculated using simulation con gurations are well-correlated with measured diffusion data at ambient conditions. The agreement between the predicted and experimentally measured NMR relaxation times and diffusion coefficient also validate the force elds used in the simulation. The molecular simulations naturally separate intramolecular from intermolecular dipole-dipole interactions helping bring new insight into the two NMR relaxation mechanisms as a function of molecular chain-length (i.e. carbon number). Comparison of the MD simulation results of the two relaxation mechanisms with traditional hard-sphere models used in interpreting NMR data reveals important limitations in the latter. With increasing chain length, there is substantial deviation in the molecular size inferred on the basis of the radius of gyration from simulation and the tted hard-sphere radii required to rationalize the relaxation times. This deviation is characteristic of the local nature of the NMR measurement, one that is well-captured by molecular simulations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2019

Search for heavy neutrinos and third-generation leptoquarks in hadronic states of two τ leptons and two jets in proton-proton collisions at $ \sqrt{s}=13 $ TeV

A search for new particles has been conducted using events with two high transverse momentum $\tau$ leptons that decay hadronically and at least two energetic jets. The analysis is performed using data from proton-proton collisions at $\sqrt{s}=$ 13 TeV, collected by the CMS experiment at the LHC in 2016 and corresponding to an integrated luminosity of 35.9 fb$^{-1}$. The observed data are consistent with standard model expectations. The results are interpreted in the context of two physics models. The first model involves right-handed charged bosons, W$_\mathrm{R}$, that decay to heavy right-handed Majorana neutrinos, N$_\ell$ $(\ell= $e, $\mu,\tau)$, arising in a left-right symmetric extension of the standard model. The model considers that N$_\mathrm{e}$ and N$_\mu$ are too heavy to be detected at the LHC. Assuming that the N$_\tau$ mass is half of the W$_\mathrm{R}$ mass, masses of the W$_\mathrm{R}$ boson below 3.50 TeV are excluded at 95% confidence level. Exclusion limits are also presented considering different scenarios for the mass ratio between N$_\tau$ and W$_\mathrm{R}$, as a function of W$_\mathrm{R}$ mass. In the second model, pair production of third-generation scalar leptoquarks that decay into $\tau\tau$bb is considered, resulting in an observed exclusion region with leptoquark masses below 1.02 TeV, assuming a 100% branching fraction for the leptoquark decay to a $\tau$ lepton and a bottom quark. These results represent the most stringent limits to date on these models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2026

Sim-to-real supervised domain adaptation for radioisotope identification

Machine learning has the potential to improve the speed and reliability of radioisotope identification using gamma spectroscopy. However, meticulously labeling an experimental dataset for training is often prohibitively expensive, while training models purely on synthetic data is risky due to the domain gap between simulated and experimental measurements. In this research, we demonstrate that supervised domain adaptation can substantially improve the performance of radioisotope identification models by transferring knowledge between synthetic and experimental data domains. We consider two domain adaptation scenarios: (1) a simulation-to-simulation adaptation, where we perform multi-label proportion estimation using simulated high-purity germanium detectors, and (2) a simulation-to-experimental adaptation, where we perform multi-class, single-label classification using measured spectra from handheld lanthanum bromide (LaBr) and sodium iodide (NaI) detectors. We begin by pretraining a spectral classifier on synthetic data using a custom transformer-based neural network. After subsequent fine-tuning on just 64 labeled experimental spectra, we achieve a test accuracy of 96% in the sim-to-real scenario with a LaBr detector, far surpassing a synthetic-only baseline model (75%) and a model trained from scratch (80%) on the same 64 spectra. Furthermore, we demonstrate that domain-adapted models learn more human-interpretable features than experiment-only baseline models. Overall, our results highlight the potential for supervised domain adaptation techniques to bridge the sim-to-real gap in radioisotope identification, enabling the development of accurate and explainable classifiers even in real-world scenarios where access to experimental data is limited.

Lalor, Peter W.↗

DOE OSTI2025

Decoding substrate specificity determining factors in glycosyltransferase-B enzymes – insights from machine learning models

Substrate specificity is an essential characteristic of any enzyme's function and an understanding of the factors that determine this specificity is crucial for enzyme engineering. Unlike the structure of an enzyme which is directly impacted by its sequence, substrate specificity as an enzyme attribute involves a rather indirect relationship with sequence as it also depends on structural aspects that dictate substrate accessibility and active site dynamics. In this study, we explore the performance of classifier-based machine learning models trained on curated sequence and structural data for a class of glycosyltransferases (GTs), namely GT-Bs, to understand their substrate specificity determining factors. GTs enable the transfer of sugar moieties to other biomolecules such as oligosaccharides or proteins and are found in all kingdoms of life. In plants, GTs participate in the biosynthesis of plant cell wall biopolymers (e.g.: hemicelluloses and pectins) and are an integral part of the enzymatic machinery that enables the storage of carbon and energy as plant biomass. To elucidate the substrate specificity of uncharacterized GT-Bs, we constructed multi-label machine learning models (Support Vector Classifier, K-Nearest Neighbors, Gaussian Naïve-Bayes, Random Forest) that incorporate both sequence and structural features. These models achieve good predictive accuracies on test datasets. However, despite our use of structural information, we highlight that there is further scope for improvement in training these models to draw interpretable relationships between sequence, structure and substrate specificity determining motifs in GT-Bs.

97 MATHEMATICS AND COMPUTING↗