Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “validation dataset”

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 829 records · Page 46
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Evaporation Dataset from a Temperate Closed-basin Lake in Minnesota 2014-2016

This data set contains observed and modeled evaporation data, eddy covariance data, and supporting information measured at White Bear Lake in the Twin Cities Metropolitan Area near St. Paul, Minnesota, USA. The data package includes csv files containing half-hourly latent heat flux, sensible heat flux, and meteorological data from 2014 to 2016. It also includes csv files containing half-hourly observed evaporation measurements, along with supporting data for validation and retrospective modeling of evaporation. This data was collected to examine current, past, and future lake evaporation to better understand how climate will impact the water balance of lakes within this region.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

A generative artificial intelligence framework based on a molecular diffusion model for the design of metal-organic frameworks for carbon capture

Metal-organic frameworks (MOFs) exhibit great promise for CO 2 capture. However, finding the best performing materials poses computational and experimental grand challenges in view of the vast chemical space of potential building blocks. Here, we introduce GHP-MOFassemble, a generative artificial intelligence (AI), high performance framework for the rational and accelerated design of MOFs with high CO 2 adsorption capacity and synthesizable linkers. GHP-MOFassemble generates novel linkers, assembled with one of three pre-selected metal nodes (Cu paddlewheel, Zn paddlewheel, Zn tetramer) into MOFs in a primitive cubic topology. GHP-MOFassemble screens and validates AI-generated MOFs for uniqueness, synthesizability, structural validity, uses molecular dynamics simulations to study their stability and chemical consistency, and crystal graph neural networks and Grand Canonical Monte Carlo simulations to quantify their CO 2 adsorption capacities. We present the top six AI-generated MOFs with CO 2 capacities greater than 2m mol g -1 , i.e., higher than 96.9% of structures in the hypothetical MOF dataset.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Identifying Vehicle Signals in Continuous Seismic Data Using Unsupervised Machine-Learning Techniques

Seismic sensors deployed near roadways effectively capture ground vibrations generated by passing vehicles. Although both traditional and machine‐learning algorithms have been utilized for analyzing such signals, independent validation of detected vehicle events remains limited. We applied two unsupervised machine‐learning algorithms, uniform manifold approximation and projection for dimension reduction, and hierarchical density‐based spatial clustering of applications with noise, to continuous seismic data collected along a road on the main campus of Oak Ridge National Laboratory. The algorithms identified seven distinct cluster labels across the entire dataset. By comparing these cluster labels with precipitation records from a nearby weather station and image‐derived labels from a local camera system, we identified one cluster associated with rainfall and another with vehicle activity. Our algorithms identified a greater number of vehicle‐related labels compared to the camera‐derived labels because seismic data are unaffected by poor lighting conditions. The arrival times of the newly detected vehicle signals corresponded well with the road’s speed limit, supporting our findings. Our algorithm outperformed the short‐term average/long‐term average method and k‐means clustering. Our results suggest that seismic data, when analyzed with machine‐learning algorithms, can complement existing vehicle monitoring systems, particularly under challenging environmental conditions.

Chai, Chengping [Oak Ridge National Laboratory (OR↗
DOE OSTI2022

Machine Learning Prediction of the Critical Cooling Rate for Metallic Glasses from Expanded Datasets and Elemental Features

In this study, we use a random forest (RF) model to predict the critical cooling rate (R C ) for glass formation of various alloys from features of their constituent elements. The RF model was trained on a database that integrates multiple sources of direct and indirect R C data for metallic glasses to expand the directly measured R C database of less than 100 values to a training set of over 2000 values. The model error on 5-fold cross-validation (CV) is 0.66 orders of magnitude in K/s. The error on leave-out-one-group CV on alloy system groups is 0.59 log units in K/s when the target alloy constituents appear more than 500 times in training data. Using this model, we make predictions for the set of compositions with melt-spun glasses in the database and for the full set of quaternary alloys that have constituents which appear more than 500 times in training data. These predictions identify a number of potential new bulk metallic glass systems for future study, but the model is most useful for the identification of alloy systems likely to contain good glass formers rather than detailed discovery of bulk glass composition regions within known glassy systems.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Correlation approach for quality assurance of additive manufactured parts based on optical metrology

Surface topography and surface finish are two significant factors for evaluating the quality of products in additive manufacturing (AM). AM parts are fabricated layer by layer, which is quite different from traditional formative or subtractive methods. Despite rapid progress in additive manufacturing and associated optical metrology for quality control and in-situ monitoring, limited research has been conducted to investigate the reliability of 3D surface measurement data. The surface topologies scanned by multiple optical systems demonstrated significant differences due to varying sampling mechanisms, resolutions, system noises, etc. The 3D datasets should be trustworthy in order to extract parameters for quality assurance or feedback control from 3D surface measurements. In this paper, we set up new standards to evaluate the reliability of 3D surface measurement data and analyze the variation in the topographical profile. In this study, two non-contact optical methods based on Focus Variation Microscopy (FVM) and Structured Light System (SLS) were adopted to measure the surface topography of the target components. The two optical metrology systems generated two entirely different point cloud datasets. Statistical methods were applied to test the difference between the data obtained from the two systems. By using a data analytics approach for comparison, it was found that the surface roughness estimated from the point cloud data sets of FVM and SLS has no significant difference, though the point cloud data sets were completely different. This paper provides standard validation approach to evaluate the plausibility of metrology data from in-situ real-time surface analysis for process planning of AM.

36 MATERIALS SCIENCE↗
DOE OSTI2025

The Lyman- α forest from LBGs: First 3D correlation measurement with DESI and prospects for cosmology

The Lyman-α (Lyα) forest is a key tracer of large-scale structure at redshifts z > 2, traditionally studied using the spectra of luminous but relatively rare quasars. In this work, we explore the viability of using the fainter yet significantly more abundant Lyman Break Galaxies (LBGs) as alternative background sources for Lyα forest studies. We analyze 4,151 Lyα forest skewers extracted from LBG spectra obtained in the DESI pilot surveys conducted in the COSMOS and XMM-LSS fields. From this dataset, we present the first measurement of the Lyα forest auto-correlation function derived exclusively from LBG spectra, probing comoving separations up to 48 h -1 Mpc at an effective redshift of z eff = 2.70. The measured LBG Lyα forest auto-correlation is consistent with that derived from DESI DR2 quasar Lyα forest spectra at a comparable redshift, validating the use of LBGs as reliable background sources for Lyα forest analyses. In addition, we measure the cross-correlation between the LBG Lyα forest and the positions of 13,362 galaxies, demonstrating that this observable serves as a sensitive diagnostic for assessing the precision and accuracy of galaxy redshift estimates, and for identifying and correcting systematic offsets. Finally, using both synthetic LBG spectra and Fisher matrix forecasts, we show that a future wide-area survey covering ∼5,000 deg 2 , targeting 1,000 LBGs per square degree at signal-to-noise levels comparable to our sample, could enable LBG-based Lyα forest baryon acoustic oscillation (BAO) measurements with expected uncertainties of σ α ISO = 0.4% (isotropic) and σ α AP = 1.3% (Alcock-Paczynski). This performance is further enhanced when combining the BAO analysis with a Lyα forest Full Shape (FS) approach, yielding a predicted uncertainty of σ α ISO FS = 0.6%. These results open a new avenue for precision cosmology at high redshift using the Lyα forest in dense LBG samples.

Lyman alpha forest↗
DOE OSTI2022

Application of machine learning to sporadic experimental data for understanding epitaxial strain relaxation

Understanding epitaxial strain relaxation is one of the key challenges in functional thin films with strong structure–property relations. Herein, we employ an emerging data analytics approach to quantitatively evaluate the underlying relationships between critical thickness ($h_c$) of strain relaxation and various physical and chemical features, despite the sporadic experimental data points available. First, we have collected and refined the reported $h_c$ of the perovskite oxide thin film/substrate system to construct a consistent sub-dataset which captures a common trend among the varying experimental details. Then, we employ correlation analyses and feature engineering to find the most relevant feature set which includes Poisson's ratio and lattice mismatch. With the insight offered by correlation analyses and feature engineering, machine learning (ML) models have been trained to deduce a decent accuracy, which has been further validated experimentally. In this work, the demonstrated framework is expected to be efficiently extended to the other classes of thin films in understanding $h_c$.

36 MATERIALS SCIENCE↗
DOE OSTI2026

WINDPROF: Merged Best-Estimate Wind Profile Data – Cape Cod (WFIP3 Campaign)

WINDPROF provides 10-minute wind and turbulence profiles that integrate a scanning Doppler lidar, a profiling Doppler lidar, and two sonic anemometers across Northeast U.S. coastal and offshore sites during the WFIP3 campaign. Variables include wind speed, wind direction, vertical velocity, turbulence intensity, and turbulent kinetic energy, each with per-instrument quality control and inter-instrument agreement validation. Profiles are mapped to a standardized height grid – dedicated near-surface levels at the sonic measurement heights (4 m and 10 m AGL), 20 m spacing to 100 m, and 30 m spacing above – and carry component and derived uncertainty estimates. Heights are reported above ground level; the site's ground elevation is stored separately. The Cape Cod dataset covers 1 February 2024 – 6 September 2025.

17 WIND ENERGY↗
DOE OSTI2025

Flood Susceptibility Mapping Using Machine Learning and Geospatial-Sentinel-1 SAR Integration for Enhanced Early Warning Systems

This study presents a comprehensive framework for flood susceptibility mapping by integrating geospatial factors with both statistical and machine learning models. Thirteen Flood-related factors, including DEM, slope, TWI, NDVI, etc., are extracted as features of models, and historical flood data derived from Sentinel-1 SAR from 2018 to 2023 are used as the target variables of the models. These datasets are analyzed using a frequency-based statistical model and three machine learning models, including Random Forest, XGBoost, and CNN, to generate flood susceptibility maps. The performance of each model is evaluated through AUC; and SHAP scores are separately generated for Machine learning (ML) models to explain each feature contribution in the ML model. The generated susceptibility maps are validated by high-flood-risk locations monitored by flood sensors, BLE inundation models, and flood-prone areas suggested by the Local Community Task Force. The results indicate that the XGBoost model outperforms all other models, with an AUC of 0.92 and demonstrates the highest alignment with recommended high-flood-risk locations, while the frequency-based statistical model showed the weakest performance with an AUC of 0.65. SHAP value graphs highlight the elevation, slope, and TWI as the most influential features across all models. The susceptibility maps generated by the machine learning model show strong agreement with the BLE map and high-flood-risk areas identified by the local Community Task Force.

Google Engine↗
DOE OSTI2022

Field validation of data-driven BSDF and peak extraction models for light-scattering fabric shades

Shading and daylighting systems affect cooling, heating, and lighting energy use by modulating solar radiation through the building façade. Characterizing shading systems holistically and accurately helps designers and engineers evaluate shading systems to achieve energy and non-energy performance goals. These complex fenestration systems can be modeled using Bidirectional Scattering Distribution Functions (BSDF), which map incident radiation to hemispherical distributions of outgoing radiation. Data-driven, tabulated BSDFs are derived from interpolated goniophotometer measured data, then sampled during the raytracing calculation. A peak extraction (PE) algorithm was developed to circumvent limits in BSDF angular resolution, where the specular peak is extracted during simulation by evaluating the BSDF in the through direction and surrounding region. The objective of this study was to validate this measurement and modeling workflow using field monitored data from a full scale testbed with eleven installed fabrics of different weaves, openness factors, and colors and assess the accuracy of the workflow under different adaptation and contrast conditions. Test conditions were limited to clear sky conditions with the sun in the field of view. Results showed that, for tensor tree datasets, vertical illuminance, solar luminance (2.5° apex), and daylight glare probability (DGP) were predicted to within a mean bias error (MBE) error of -456 lx (-12.3%), -3.46e5 (-38.4%), and -0.042 (-7.8%) when full PE occurred. With a binary classification of glare/ no glare, DGP was predicted accurately with a true positive rate of 0.98 and true negative rate of 1.0 using tensor tree data and less accurately with Klems BSDF data, particularly for cases of no glare. The workflow may be of insufficient accuracy to distinguish borderline performance between fabrics using the four-point glare scale, particularly under low adaptation, high contrast daylit conditions. Errors were due to reductions in peak shape and intensity across the BSDF interpolation and data reduction workflow. Future work is needed to better preserve measurement fidelity during interpolation and sampling, which in turn will improve PE performance.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Deep learning model for fast, science-based forecasting of fluid migration along faults in geologic carbon storage scenarios

Effective long-term geologic storage depends on robust site selection and credible, science-based forecasting of subsurface behavior to ensure storage integrity. For this work, we develop a deep learning–based reduced-order model (ROM) to quantify potential carbon dioxide (CO₂) and brine migration through geological faults. The ROM combines a Transformer model for binary classification and a Stacked Ensemble for regression, trained on a comprehensive dataset generated from 1400 physics-based reservoir simulations. Key geologic and operational parameters—including fault geometry, reservoir structure, and injection conditions—were systematically varied to capture a wide range of fluid migration scenarios. The ROM accurately predicts the onset of migration, cumulative migration volumes of both CO₂ and brine, and associated migration rates, as compared to an independent set of validation simulations, while significantly reducing computational cost compared to traditional simulation methods. Model performance was evaluated across diverse fault configurations, revealing that shallow reservoir geometry and fault angle are among the most influential factors governing migration behavior. Sensitivity analysis using SHapley Additive exPlanations (SHAP) provided interpretability, revealing distinct patterns in how geological and operational features drive transient versus cumulative migration outcomes. The ROM’s ability to rapidly simulate fault migration scenarios enables efficient sensitivity analyses, scenario evaluations, and decision support for site selection and monitoring design. This approach enhances the safety, scalability, and long-term operational performance of geologic carbon storage (GCS) systems by providing a robust, interpretable tool for predicting subsurface fluid migration and assessing fault-related migration potential.

42 ENGINEERING↗
DOE OSTI2020

Analysis of alternative climate datasets and evapotranspiration methods for the Upper Mississippi River Basin using SWAT within HAWQS

This study reports the application of Soil andWater Assessment Tool (SWAT) within the Hydrologic andWater Quality System(HAWQS) on-line platform, for the UpperMississippi River Basin (UMRB). TheUMRB is an important ecosystem located in the north central U.S. that is experiencing a range of ecological stresses. Specifically, testing of SWAT was performed for: (1) Hargreaves (HG) and Penman-Monteith (PM) PET methods, and (2) Livneh, National Climatic Data Center (NCDC) and Parameter-elevation Regressions on Independent Slopes Model (PRISM) climate datasets. The Livneh-PM combination resulted in the highest average annual water yield of 380.6 mm versus the lowest estimated water yield of 193.9 mm for the Livneh-HG combination, in response to 23-year uncalibrated simulations. Higher annual ET and PET values were predicted with HG method versus the PM method for all three weather datasets in response to the uncalibrated simulations, due primarily to higher HG-based estimates during the growing season. Based on these results, it was found that the HG method is the preferred PET option for the UMRB. Initial calibration of SWAT was performed using the Livneh data and HGmethod for threeMississippi Rivermain stem gauge sites, which was followed by spatial validation at 10 other gauge sites located within the UMRB streamnetwork. Overall satisfactory results were found for the calibration and validation gauge sites, with the majority of R2 values ranging between 0.61 and 0.82, Nash- Sutcliffe modeling efficiency (NSE) values ranging between 0.50 and 0.79, and Kling-Gupta efficiency (KGE) values ranging between 0.61 and 0.84. The results of an additional experimental suite of six scenarios,which represented different combinations of climate data sets and calibrated parameters, revealed that suggested statistical criteria were again satisfied by the different scenario combinations. Overall, the PRISM data exhibited the strongest reliability for the UMRB.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Next-generation large-scale binary protein interaction network for Drosophila melanogaster

Generating reference maps of interactome networks illuminates genetic studies by providing a protein-centric approach to finding new components of existing pathways, complexes, and processes. We apply state-of-the-art methods to identify binary protein-protein interactions (PPIs) for Drosophila melanogaster. Four all-by-all yeast two-hybrid (Y2H) screens of > 10,000 Drosophila proteins result in the ‘FlyBi’ dataset of 8723 PPIs among 2939 proteins. Testing subsets of data from FlyBi and previous PPI studies using an orthogonal assay allows for normalization of data quality; subsequent integration of FlyBi and previous data results in an expanded binary Drosophila reference interaction network, DroRI, comprising 17,232 interactions among 6511 proteins. We use FlyBi data to generate an autophagy network, then validate in vivo using autophagy-related assays. The deformed wings (dwg) gene encodes a protein that is both a regulator and a target of autophagy. Altogether, these resources provide a foundation for building new hypotheses regarding protein networks and function.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Moving lens effect: Simulations, forecasts, and foreground mitigation

The peculiar motion of massive objects across the line of sight imprints a dipolar temperature anisotropy pattern on the cosmic microwave background known as the moving lens effect. This effect provides a unique probe of the transverse components of the peculiar velocity field, but has not yet been detected due to its small size. We implement and validate a stacking estimator for the moving lens signal using a galaxy catalog as a tracer of massive haloes combined with reconstructed velocities from the galaxy number density field. Using simulations, we forecast detection prospects for the moving lens signal from current and upcoming microwave background and galaxy surveys. Here, we demonstrate a new foreground mitigation strategy likely sufficient for current datasets, and discuss various sources of systematic error and noise. Upcoming galaxy surveys will provide high-significance statistical detections of the moving lens effect.

Astrophysical & cosmological simulations↗
DOE OSTI2026

Daily, 30 m Resolution NDSI Data for the East River Watershed, CO for 2000-2020

This dataset contains daily Normalized Difference Snow Index (NDSI) values at 30 m spatial resolution for the East River watershed in Colorado, USA. The temporal range of these data includes water years 2001-2020. These data were created using the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM). This model fuses low spatial and high temporal resolution data from MODIS (500 m, daily) with high spatial and low temporal resolution data from Landsat (30 m, 16 days) to create a 30m synthetic daily snow product. This product allows for the analysis of historical snow covered area trends in the East River Watershed at fine spatiotemporal resolutions where it was not available previously. This research was performed as a part of the Department of Energy’s Subsurface Biogeochemical Research Program with the primary intent of better understanding the timing and spatial patterns of water delivery to the Critical Zone in mountain watersheds. Each .zip file contains one "water year" of data (October 1 - September 30; i.e., water year 2010 starts October 1, 2010 and ends September 30, 2011). Each zip file contains the following: STARFM daily Normalized Difference Snow Index (NDSI) fusion data files in GeoTiff format with one layer for each day between Landsat data acquisition dates (i.e., for dates of Landsat acquisition, the Landsat image is included for that date). The study area is located in an area of Landsat path overlap, so Landsat dates acquisitions are every 7-9 days. Landsat NDSI files containing the high spatial (30m), low temporal (7-9 days due to Landsat path overlap) resolution data used as input to STARFM in GeoTiff format with one layer for each day. Dates for which no Landsat data were obtained are included as NoData layers. MODIS NDSI files containing the high temporal (daily), low spatial (500m) resolution data used as input to STARFM in GeoTiff format with one layer for each day. Please note the MODIS data were resampled to 30m pixels for input into the STARFM model. The data have a scale factor of 10,000 and a no data value of -32767. The projection of all datasets is WGS 84 (EPSG: 4326), which has a latitude/longitude based degree resolution of 0.0002694946 X 0.0002694946, and approximates to the 30 m spatial resolution mentioned above. The Layer Index files in .csv format. They contain information for each layer in the above GeoTiff files regarding the corresponding date for each layer, the fraction of pixels in the image that contain valid data (missing data is due to either cloud cover or poor data quality; these values are not percent snow cover). Dates of Landsat overpass are indicated in these files. If no Landsat data were able to be obtained due to cloud cover or lack of Landsat Tier 1 data available on Google Earth Engine, this is also noted.

EARTH SCIENCE > CRYOSPHERE > SNOW/ICE↗
DOE OSTI2023

Few-Shot Learning Enables Population-Scale Analysis of Leaf Traits in Populus trichocarpa

Plant phenotyping is typically a time-consuming and expensive endeavor, requiring large groups of researchers to meticulously measure biologically relevant plant traits, and is the main bottleneck in understanding plant adaptation and the genetic architecture underlying complex traits at population scale. In this work, we address these challenges by leveraging few-shot learning with convolutional neural networks to segment the leaf body and visible venation of 2,906 Populus trichocarpa leaf images obtained in the field. In contrast to previous methods, our approach (a) does not require experimental or image preprocessing, (b) uses the raw RGB images at full resolution, and (c) requires very few samples for training (e.g., just 8 images for vein segmentation). Traits relating to leaf morphology and vein topology are extracted from the resulting segmentations using traditional open-source image-processing tools, validated using real-world physical measurements, and used to conduct a genome-wide association study to identify genes controlling the traits. In this way, the current work is designed to provide the plant phenotyping community with (a) methods for fast and accurate image-based feature extraction that require minimal training data and (b) a new population-scale dataset, including 68 different leaf phenotypes, for domain scientists and machine learning researchers. All of the few-shot learning code, data, and results are made publicly available.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

A Novel Segmentation Algorithm for the ARM User Facility All-Sky Imagers Using Machine Learning Applications

Cloud cover plays a pivotal role in modulating the Earth's energy budget through the reflection of incoming solar radiation and the trapping of outgoing longwave radiation. Ground-based all-sky imagers offer an objective assessment of cloud cover that can be used to estimate solar irradiance, classify cloud types, track cloud movement, and serve as a benchmark 10 for the evaluation of satellite and reanalysis data products. The Atmospheric Radiation Measurement (ARM) user facility has utilized all-sky imagers for more than 25 years to monitor cloud cover and augment its comprehensive suite of atmospheric measurements. Following the retirement of its Total Sky Imager (TSI), ARM recently deployed the TSI’s successor, the All Sky Imager (ASI-16 camera systems). To provide a smooth transition and continuity to the vast amount of knowledge gathered by the TSI over the years, while addressing typical deployment issues, we developed a novel pixel segmentation algorithm, 15 the ASI Sky Cover (ASISKYCOVER). ASISKYCOVER builds on the different strengths and properties of the TSI processing algorithm while integrating machine learning techniques, ensuring data validity and accuracy across diverse atmospheric conditions. It enhances cloud cover characterization with new features such as artifact detection and uncertainty quantification. ASISKYCOVER also includes cloud cover estimates for near-zenith (narrow field-of-view) and reduces susceptibility to false detections. This study introduces ASISKYCOVER, details its algorithm framework, and demonstrates its capabilities using a 20 year-long dataset from the ARM Southern Great Plains site. Comparisons with co-located TSI data and other ARM measurements, such as zenith-pointing radars and lidars, are presented, underscoring the ASISKYCOVER’s potential to improve cloud cover analyses and data evaluation efforts, as well as to be integrated into higher-level data products that synergize instrument suites to generate new and insightful information

Silber, Israel↗
DOE OSTI2021

CyProduct: A software tool for accurately predicting the byproducts of human cytochrome P450 metabolism

In silico metabolism prediction is a cheminformatic task of autonomously predicting the set of metabolic byproducts produced from a specified molecule and a set of enzymes or reactions. Here we describe a novel machine-learned in silico cytochrome P450 (CYP450) metabolism prediction suite, called CyProduct, that accurately predicts metabolic byproducts for a specified molecule and a human CYP450 isoform. It includes three modules: (1) CypReact, a tool that predicts if the query compound reacts with a given CYP450 enzyme; (2) CypBoM, a tool that accurately predicts the “bond site” of the reaction (i.e., which specific bonds within the query molecule react with the CYP isoform); and (3) MetaboGen, a tool that generates the metabolic byproducts based on CypBoM’s bond-site prediction. CyProduct predicts metabolic biotransformation products for each of the nine most important human CYP450 enzymes. CypBoM uses an important new concept called “Bond of Metabolism” (BoM), which extends the traditional “Site of Metabolism" (SoM) by specifying the information about the set of chemical bonds that is modified or formed in a metabolic reaction (rather than the specific atom). We created a BoM database for 3487 CYP450-mediated Phase I reactions, then used this to train the CypBoM Predictor to predict the reactive bond locations on substrate molecules. CypBoM Predictor’s cross-validated Jaccard score for reactive bond prediction ranged from 0.380 to 0.452 over the nine CYP450 enzymes. Over variants of a test set of 72 known CYP450 substrates and 30 non-reactants, CyProduct outperformed the other packages -- including ADMET Predictor, BioTransformer and GLORY -- by an average of 200% (wrt Jaccard score) in terms of predicting metabolites. The CyProduct suite and the datasets are freely available at https://bitbucket.org/wishartlab/cyproduct/src/master/.

Machine learning, Cytochrome P450, Metabolism pred↗