Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Statistical accuracy”

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 469 records · Page 26
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Comparison of Results from Recent NNSA and CEA Measurements of the 239 Pu(n, f) Prompt Fission Neutron Spectrum

The National Nuclear Security Administration (NNSA)/DP French Alternative Energies and Atomic Energy Commission (CEA)/DAM agreement on cooperation on fundamental science is a U.S.-French collaborative effort to combine intellectual and experimental resources and further the relevant nuclear science. Recently, both the NNSA and CEA experimental teams performed high-statistics measurements of the 239 Pu(n, f) prompt fission neutron spectrum (PFNS) at the Los Alamos Neutron Science Center, both of which were recently published in the journal Physical Review C. These separate measurements used the same experimental area and a common neutron detector array, but differ in many aspects, including background assessments, data acquisition systems and philosophies, fission detectors, and PFNS extraction techniques. Hence, some aspects of the experimental methods and associated uncertainties are highly correlated while others are independent. The results from both measurements broke new ground for PFNS measurements given their higher accuracy and more detailed study of corrections necessary for the measured quantity compared to existing literature measurements, and both will significantly impact PFNS nuclear data evaluations for the foreseeable future. Here, the focus of this work is to document a comparison of the results from these distinct measurements in terms of the acquired data, the PFNS results, and the measured average PFNS energies. While systematic differences between the PFNS results are present on the 1–3% level, the acquired data relative to each respective measurement at low incident neutron energies are in remarkable agreement, as are the conclusions regarding the magnitude and position of features in the PFNS relating to second-chance fission, third-chance fission, and pre-equilibrium neutron emission.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2020

Grid-based diffusion Monte Carlo for fermions without the fixed-node approximation

A diffusion Monte Carlo algorithm is introduced that can determine the correct nodal structure of the wave function of a few-fermion system and its ground-state energy without an uncontrolled bias. This is achieved by confining signed random walkers to the points of a uniform infinite spatial grid, allowing them to meet and annihilate one another to establish the nodal structure without the fixed-node approximation. An imaginary-time propagator is derived rigorously from a discretized Hamiltonian, governing a non-Gaussian, sign-flipping, branching, and mutually annihilating random walk of particles. The accuracy of the resulting stochastic representations of a fermion wave function is limited only by the grid and imaginary-time resolutions and can be improved in a controlled manner. Here, the method is tested for a series of model problems including fermions in a harmonic trap as well as the He atom in its singlet or triplet ground state. For the latter case, the energies approach from above with increasing grid resolution and converge within 0.015 E h of the exact basis-set-limit value for the grid spacing of 0.08 a.u. with a statistical uncertainty of 10 –5 E h without an importance sampling or Jastrow factor.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2023

Machine-learning-based investigation of the variables affecting summertime lightning occurrence over the Southern Great Plains

Lightning is affected by many factors, many of which are not routinely measured, well understood, or accounted for in physical models. Several commonly used machine learning (ML) models have been applied to analyze the relationship between Atmospheric Radiation Measurement (ARM) data and lightning data from the Earth Networks Total Lightning Network (ENTLN) in order to identify important variables affecting lightning occurrence in the vicinity of the Southern Great Plains (SGP) ARM site during the summer months (June, July, August and September) of 2012 to 2020. Testing various ML models, we found that the random forest model is the best predictor among common classifiers. When convective clouds were detected, it predicts lightning occurrence with an accuracy of 76.9 % and an area under the curve (AUC) of 0.850. Using this model, we further ranked the variables in terms of their effectiveness in nowcasting lightning and identified geometric cloud thickness, rain rate and convective available potential energy (CAPE) as the most effective predictors. The contrast in meteorological variables between no-lightning and frequent-lightning periods was examined for hours with CAPE values conducive to thunderstorm formation. Besides the variables considered for the ML models, surface variables and mid-altitude variables (e.g., equivalent potential temperature and minimum equivalent potential temperature, respectively) have statistically significant contrasts between no-lightning and frequent-lightning hours. For example, the minimum equivalent potential temperature from 700 to 500 hPa is significantly lower during frequent-lightning hours compared with no-lightning hours. Finally, a notable positive relationship between the intracloud (IC) flash fraction and the square root of CAPE ($\sqrt{CAPE}$) was found, suggesting that stronger updrafts increase the height of the electrification zone, resulting in fewer flashes reaching the surface and consequently a greater IC flash fraction.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Evaluation of obstacle modelling approaches for resource assessment and small wind turbine siting: case study in the northern Netherlands

Abstract. Growth in adoption of distributed wind turbines for energy generation is significantly impacted by challenges associated with siting and accurate estimation of the wind resource. Small turbines, at hub heights of 40 m or less, are greatly impacted by terrestrial obstacles such as built structures and vegetation that can cause complex wake effects. While some progress in high-fidelity complex fluid dynamics (CFD) models has increased the potential accuracy for modelling the impacts of obstacles on turbulent wind flow, these models are too computationally expensive for practical siting and resource assessment applications. To understand the efficacy of available models in situ, this study evaluates classic and commonly used methods alongside new state-of-the-art lower-order models derived from CFD simulations and machine learning approaches. This evaluation is conducted using a subset of an extensive original dataset of measurements from more than 300 operational wind turbines in the northern Netherlands. The results show that data-driven methods (e.g. machine learning and statistical modelling) are most effective at predicting production at real sites with an average error in annual energy production of 2.5 %. When sufficient data may not be available de novo to support these data-driven approaches, models derived from high-fidelity simulations show promise and reliably outperform classic methods. On average these models have 6.3 %–11.5 % error compared with 26 % for classic methods and 27 % baseline error for reanalysis data without obstacle correction. While more performant on average, these methods are also sensitive to the quality of obstacle descriptions and reanalysis inputs.

17 WIND ENERGY↗
DOE OSTI2020

Sensitivity of future liquid argon dark matter search experiments to core-collapse supernova neutrinos

Future liquid-argon DarkSide-20k and ARGO detectors, designed for direct dark matter search, will be sensitive also to core-collapse supernova neutrinos, via coherent elastic neutrino-nucleus scattering. This interaction channel is flavor-insensitive with a high-cross section, enabling for a high-statistics neutrino detection with target masses of $\sim$50~t and $\sim$360~t for DarkSide-20k and ARGO, respectively. Thanks to the low-energy threshold of $\sim$0.5~keV$_{nr}$ achievable by exploiting the ionization channel, DarkSide-20k and ARGO have the potential to discover supernova bursts throughout our galaxy and up to the Small Magellanic Cloud, respectively, assuming a 11-M$_{\odot}$ progenitor star. We report also on the sensitivity to the neutronization burst, whose electron neutrino flux is suppressed by oscillations when detected via charged current and elastic scattering. Finally, the accuracies in the reconstruction of the average and total neutrino energy in the different phases of the supernova burst, as well as its time profile, are also discussed, taking into account the expected background and the detector response.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2021

Inter-well connectivity detection in CO 2 WAG projects using statistical recurrent unit models

Routine well-wise injection and production measurements contain significant information on subsurface structure and properties. Data-driven technology that interprets surface data into subsurface structure or properties can assist operators in making informed decisions by providing a better understanding of field assets. Our machine-learning framework is built on the statistical recurrent unit (SRU) model and interprets well-based injection/production data into inter-well connectivity without relying on a geologic model. We test it on synthetic and field-scale CO 2 EOR projects utilizing the water-alternating-gas (WAG) process. SRU is a special type of recurrent neural network (RNN) that allows for better characterization of temporal trends, by learning various statistics of the input at different time scales. In our application, the complete states (injection rate, pressure and cumulative injection) at injectors and pressure states at producers are fed to SRU as the input and the phase rates at producers are treated as the output. Once the SRU is trained and validated, it is then used to assess the connectivity of each injector to any producer using permutation variable importance method, wherein inputs corresponding to an injector are shuffled and the increase in prediction error at a given producer is recorded as the importance (connectivity metric) of the injector to the producer. This method is tested in both synthetic and field-scale cases. The validation of the proposed data-driven inter-well connectivity assessment is performed using synthetic data from simulation models where inter-well connectivity can be easily measured using the streamline-based flux allocation. The SRU model is shown to offer excellent prediction performance on the synthetic case. Despite significant measurement noise and frequent well shut-ins imposed in the field-scale case, the SRU model offers good prediction accuracy, the overall relative error of the phase production rates at most producers ranges from 10% to 30%. It is shown that the dominant connections identified by the data-driven method and streamline method are in close agreement. This significantly improves confidence in our data-driven procedure. The novelty of this work is that it is purely data-driven method and can directly interpret routine surface measurements to intuitive subsurface knowledge. Furthermore, the streamline-based validation procedure provides physics-based backing to the results obtained from data analytics. This study results in a reliable and efficient data analytics framework that is well-suited for large field applications.

42 ENGINEERING↗
DOE OSTI2020

Mining a human transcriptome database for chemical modulators of NRF2

Nuclear factor erythroid-2 related factor 2 (NRF2) encoded by the NFE2L2 gene is a transcription factor critical for protecting cells from chemically-induced oxidative stress. We developed computational procedures to identify chemical modulators of NRF2 in a large database of human microarray data. A gene expression biomarker was built from statistically-filtered gene lists derived from microarray experiments in primary human hepatocytes and cancer cell lines exposed to NRF2-activating chemicals (oltipraz, sulforaphane, CDDOIm) or in which the NRF2 suppressor Keap1 was knocked down by siRNA. Directionally consistent biomarker genes were further filtered for those dependent on NRF2 using a microarray dataset from cells after NFE2L2 siRNA knockdown. The resulting 143-gene biomarker was evaluated as a predictive tool using the correlation-based Running Fisher algorithm. Using 59 gene expression comparisons from chemically-treated cells with known NRF2 activating potential, the biomarker gave a balanced accuracy of 93%. The biomarker was comprised of many well-known NRF2 target genes (AKR1B10, AKR1C1, NQO1, TXNRD1, SRXN1, GCLC, GCLM), 69% of which were found to be bound directly by NRF2 using ChIP-Seq. NRF2 activity was assessed across ~9840 microarray comparisons from ~1460 studies examining the effects of ~2260 chemicals in human cell lines. A total of 260 and 43 chemicals were found to activate or suppress NRF2, respectively, most of which have not been previously reported to modulate NRF2 activity. Using a NRF2-responsive reporter gene in HepG2 cells, we confirmed the activity of a set of chemicals predicted using the biomarker. The biomarker will be useful for future gene expression screening studies of environmentally-relevant chemicals.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Data Summarization and Inference at Scale

This is the final report for the DOE ASCR grant SC-0022260, Data Summarization and Inference at Scale, PI: Alex Pothen, Purdue University. The goal of the project was to solve data-intensive and compute-intensive problems in the physical sciences, engineering, information science, data science, etc. by designing and implementing new algorithms that could work with a subset of the data. The four subgoals were: (a) The solution of problems where the data is too large to be stored in the memory of a computer. In this streaming model of computation, the data arrives as a stream of elements to the computer, each element is processed as it arrives, and a decision is made to discard the data or to store it; only a small subset of the data proportional to the size of the output solution is stored, and when all the data has been streamed, a solution to the problem is computed from the stored subset. (b) The use of machine learning methods to compute solutions to data-intensive problems. The use of GPUs is critical to obtain high performance on machine learning tasks, but their memory sizes are smaller relative to that of CPUs. For large-scale problems, the data is sampled many times, and small samples are used with repetition, for robustness, to compute solutions to inference tasks. This sampling reduces the memory required to solve the problem, but attention is needed to avoid slow convergence to the solutions, and reduced accuracy of inference. We propose submodular optimization, Large Language Models, and physics-informed neural networks to enable GPU computations here. (c) Modeling and visualization of high-dimensional data using interpretable features. Clinical proteomic data sets from immunology for the detection of cancer and other diseases are temporal and high-dimensional, and algorithms for visualizing these data sets using clinically interpretable features are lacking. We propose methods that compute distances based on the optimal transportation problem and graph edit distances to address this problem. We also propose the use of optimal transport-based distances, spatial statistics, and network structure to classify image data sets, We apply these algorithms to electron micrographs of the peripheral nervous system in the digestive tract. (d) The design of data-intensive algorithms on emerging architectures, specifically, noisy, intermediate-scale quantum (NISQ) devices. Quantum computers offer the possibility of exploring large solution spaces due to the principle of superposition, but current quantum computers are limited by few qubits, short coherence times due to noise, poor interconections among the qubits, etc. We propose the use of the divide and conquer paradigm to solve large-scale problems, wherein collections of small subproblems are solved on the quantum devices, and the solutions to the subproblems are integrated into a solution for the original problem on a classical computer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Comparative study of machine learning techniques for post-combustion carbon capture systems

Computational analysis of countercurrent flows in packed absorption columns, often used in solvent-based post-combustion carbon capture systems (CCSs), is challenging. Typically, computational fluid dynamics (CFD) approaches are used to simulate the interactions between a solvent, gas, and column's packing geometry while accounting for the thermodynamics, kinetics, heat, and mass transfer effects of the absorption process. These simulations can then be used explain a column's hydrodynamic characteristics and evaluate its CO 2 -capture efficiency. However, these approaches are computationally expensive, making it difficult to evaluate numerous designs and operating conditions to improve efficiency at industrial scales. In this work, we comprehensively explore the application of statistical ML methods, convolutional neural networks (CNNs), and graph neural networks (GNNs) to aid and accelerate the scale-up and design optimization of solvent-based post-combustion CCSs. We apply these methods to CFD datasets of countercurrent flows in absorption columns with structured packings characterized by several geometric parameters. We train models to use these parameters, inlet velocity conditions, and other model-specific representations of the column to estimate key determinants of CO 2 -capture efficiency without having to simulate additional CFD datasets. We also evaluate the impact of different input types on the accuracy and generalizability of each model. We discuss the strengths and limitations of each approach to further elucidate the role of CNNs, GNNs, and other machine learning approaches for CO 2 -capture property prediction and design optimization.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Deep Learning of Dark Energy Spectroscopic Instrument Mock Spectra to Find Damped Lyα Systems

We have updated and applied a convolutional neural network (CNN) machine-learning model to discover and characterize damped Ly α systems (DLAs) based on Dark Energy Spectroscopic Instrument (DESI) mock spectra. We have optimized the training process and constructed a CNN model that yields a DLA classification accuracy above 99% for spectra that have signal-to-noise ratios (S/N) above 5 per pixel. The classification accuracy is the rate of correct classifications. This accuracy remains above 97% for lower S/N ≈1 spectra. This CNN model provides estimations for redshift and H i column density with standard deviations of 0.002 and 0.17 dex for spectra with S/N above 3 pixel -1 . Also, this DLA finder is able to identify overlapping DLAs and sub-DLAs. Further, the impact of different DLA catalogs on the measurement of baryon acoustic oscillations (BAO) is investigated. The cosmological fitting parameter result for BAO has less than 0.61% difference compared to analysis of the mock results with perfect knowledge of DLAs. This difference is lower than the statistical error for the first year estimated from the mock spectra: above 1.7%. We also compared the performances of the CNN and Gaussian Process (GP) models. Our improved CNN model has moderately 14% higher purity and 7% higher completeness than an older version of the GP code, for S/N > 3. Both codes provide good DLA redshift estimates, but the GP produces a better column density estimate by 24% less standard deviation. A credible DLA catalog for the DESI main survey can be provided by combining these two algorithms.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

A review of coal heating value correlations with application to coal char, tar, and other fuels

Accurately measuring a fuel’s heating value is one of the first steps in the classification of a new fuel. Heating values are widely used in coal combustion research and are becoming more useful in other fuel types as well. Many different empirical correlations to predict heating values based on primary organic (CHONS) elemental composition are found in the literature, many of which were originally created to predict heating values of parent coal, and some of which have been extended for biomass. However, no correlations exist for heating values of coal chars and tars. Thirteen literature heating value correlations (10 model forms with 13 unique sets of coefficients) were evaluated for their predictive accuracy of a wide variety of fuel types, including 353 coals. This comparative analysis showed that the predictive capabilities of many of these literature models are very similar for coal-based fuels, with good correlation of coal char heating values. Correlation of coal tar heating values was hampered by the lack of heating value data. The choice of which model to use becomes more important for biomass fuels as well as for more unique fuel types like propellants and explosives, municipal solid waste, etc. Finally, the best model forms for each fuel type are suggested based on several statistical measures of fit.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2022

Development of an Open-source Alloy Selection and Lifetime Assessment Tool for Structural Components in CSP

Lack of sufficient data on high temperature mechanical and corrosion behavior of structural materials is a huge barrier in the technological maturity of current and future Concentrating Solar Power (CSP) technologies. Rapid development and selection of materials cannot be achieved by expensive and time-consuming acquisition of experimental data. The goal of the proposed work is development of an open-source alloy selection and lifetime prediction tool that will integrate validated physics-based models to describe influence of temperature, alloy composition, environment and component geometry (thickness) on mechanical and corrosion behavior of Ni and Fe-based alloys employed in molten salts/sCO 2 heat exchangers. This one-year project leveraged the extensive dataset on the creep\corrosion behavior of candidate materials generated at ORNL through past projects and input from current collaborations with industrial partners. Based on previous experience and the feedback provided by industry (Brayton Energy and Echogen), three candidate materials of interest, Ni-based alloys 740H, 282 and 625 and application-specific operating conditions (max. temperature of 730 °C and stress of 150 MPa) were identified for the heat exchanger. An extensive corrosion and creep dataset was assimilated for the relevant operating conditions and was supported by detailed characterization of about 100 metallographic cross-sections. The corrosion dataset consisted of scanning electron microscopy images (secondary electron and backscatter electron), measured concentration profiles of alloying elements using energy dispersive X-ray spectroscopy (EDS), widths of denuded zones (dissolution of strengthening phases) and depths of attack in molten KCl-MgCl 2 mixtures using image analyses. The creep dataset comprised of creep rupture data and creep strain curves (for 740H and 282). Coupled thermodynamic-kinetic microstructure-based models were employed to predict the stress-corrosion induced compositional and phase evolutions in the alloy during operation under the identified operating conditions. Reduced order models were developed from advanced physics-based models and were integrated in a user-friendly alloy selection tool. The corrosion model was able to predict the time to a critical Cr concentration at the oxide/alloy interface (chemical lifetime) within ±10% (1 standard deviation) of typical statistical variation in corrosion tests and EDS measurement errors (±0.5 wt%). The initial scope of the project was limited to predict creep rupture times (Larson-Miller parameter). Based on the input provided by industry, the mechanical lifetime of the heat exchanger is governed by accumulated creep strains (2%) rather than creep rupture. To be able to predict the times to specific creep strains, a more extensive creep model development was undertaken largely beyond the initial scope of the project. The continuum damage mechanics creep model was able to predict times to 2% creep strain, t 2% with an accuracy of ±500h. Ultimately, a screening protocol for SiC was generated to demonstrate the pathway for integration of one of the currently immature materials from a commercial adoption standpoint in the current material evaluation tool. The modeling tool developed here is accessible to the science community and stakeholders and lays the foundation for methods that will enable a rapid evaluation of optimum materials for CSP applications and reliable prediction of material degradation thereby considerably reducing operational costs, improving reliability and increasing overhaul intervals. However, the complete potential of such a tool to include a wider range of materials and test conditions can only be realized with a more concentrated combined experimental-characterization-computation effort.

14 SOLAR ENERGY↗
DOE OSTI2022

Improved precision in As speciation analysis with HERFD-XANES at the As K -edge: the case of As speciation in mine waste

High-energy-resolution fluorescence-detected (HERFD) X-ray absorption near-edge spectroscopy (XANES) is a spectroscopic method that allows for increased spectral feature resolution, and greater selectivity to decrease complex matrix effects compared with conventional XANES. XANES is an ideal tool for speciation of elements in solid-phase environmental samples. Accurate speciation of As in mine waste materials is important for understanding the mobility and toxicity of As in near-surface environments. In this study, linear combination fitting (LCF) was performed on synthetic spectra generated from mixtures of eight measured reference compounds for both HERFD-XANES and transmission-detected XANES to evaluate the improvement in quantitative speciation with HERFD-XANES spectra. The reference compounds arsenolite (As 2 O 3 ), orpiment (As 2 S 3 ), getchellite (AsSbS 3 ), arsenopyrite (FeAsS), kaňkite (FeAsO 4 ·3.5H 2 O), scorodite (FeAsO 4 ·2H 2 O), sodium arsenate (Na 3 AsO 4 ), and realgar (As 4 S 4 ) were selected for their importance in mine waste systems. Statistical methods of principal component analysis and target transformation were employed to determine whether HERFD improves identification of the components in a dataset of mixtures of reference compounds. LCF was performed on HERFD- and total fluorescence yield (TFY)-XANES spectra collected from mine waste samples. Arsenopyrite, arsenolite, orpiment, and sodium arsenate were more accurately identified in the synthetic HERFD-XANES spectra compared with the transmission-XANES spectra. In mine waste samples containing arsenopyrite and either scorodite or kaňkite, LCF with HERFD-XANES measurements resulted in fits with smaller R -factors than concurrently collected TFY measurements. The improved accuracy of HERFD-XANES analysis may provide enhanced delineation of As phases controlling biogeochemical reactions in mine wastes, contaminated soils, and remediation systems.

58 GEOSCIENCES↗
DOE OSTI2023

Machine-Learning Assisted Identification of Battery Life Models

Predictive battery life models are commonly utilized to extrapolate degradation trends observed during accelerated aging tests for simulation of degradation in real-world applications. Thus, fitting accelerated aging data as accurately as possible and with low uncertainty is crucial for making believable projections of battery lifetime, but it is challenging to identify algebraic expressions that accurately fit multivariate degradation trends. A review of models published in literature reveal some common expressions for fitting calendar aging data, which is only dependent on temperature and state-of-charge, but no consistency across many models for fitting cycle aging data, indicating the need for a statistically rigorous data driven approach for developing empirical models. This talk will describe a machine-learning assisted method for identification of predictive battery life models utilizing bilevel optimization and symbolic regression. Bilevel optimization with cross-validation is used to statistically determine cell- and stress-dependent model parameters, while symbolic regression identifies both linear and multiplicative candidate expressions to predict stress-dependent degradation rates by selecting low-order subsets of features from a generated feature library. Because model expressions are identified empirically, it is crucial to ensure resulting models behave according to physical expectations, so the stability of models for interpolation or extrapolation is interrogated qualitatively through simulation and quantitatively through cross-validation and uncertainty quantification via bootstrap resampling. This model identification approach substantially improves upon models identified purely using expert judgement in terms of both accuracy and uncertainty. Model simulation and validation is then conducted by deriving a state-equation form of the predictive model, enabling simulation of battery aging under dynamic stresses. This enables validation of the predictive battery model on lab-based tests with varying conditions or on drive-cycle or application-cycle testing protocols. Parameter uncertainty can be carried forward into model simulation, giving lifetime estimates and confidence windows for cell- or system-level lifetime. The financial impact of battery model uncertainty can be estimated by incorporating uncertainty into a technoeconomic model.

battery↗
DOE OSTI2023

Effects of finite sampling on fatigue damage estimation of wind turbine components: A statistical study

The variability of the wind turbine loads complicates fatigue assessment in the design phase, as performing simulations covering the entire lifetime is computationally expensive. The current work provides important information for assessing the uncertainty in fatigue damage estimation due to finite data. Here we study the sample size effect on mean, variance, and skewness of damage in each wind bin, identify the important wind bins, and study the uncertainty propagation from each wind bin to the lifetime damage using 3600 aeroelastic simulations and bootstrapping. To achieve less than 1% error in the damage estimation across all load channels in the current case study, at least 100 turbulence seeds are needed. Damage in different wind bins follows a lognormal distribution when using the conventional approach of six seeds. The provided insights and information allow the designer to achieve a specific level of accuracy for a given computational cost using strategic bin sampling.

17 WIND ENERGY↗
DOE OSTI2020

Detect and correct bias in multi-site neuroimaging datasets

The desire to train complex machine learning algorithms and to increase the statistical power in association studies drives neuroimaging research to use ever-larger datasets. The most obvious way to increase sample size is by pooling scans from independent studies. However, simple pooling is often ill-advised as selection, measurement, and confounding biases may creep in and yield spurious correlations. In this work, we combine 35,320 magnetic resonance images of the brain from 17 studies to examine bias in neuroimaging. In the first experiment, Name That Dataset, we provide empirical evidence for the presence of bias by showing that scans can be correctly assigned to their respective dataset with 71.5% accuracy. Given such evidence, we take a closer look at confounding bias, which is often viewed as the main shortcoming in observational studies. In practice, we neither know all potential confounders nor do we have data on them. Hence, we model confounders as unknown, latent variables. Kolmogorov complexity is then used to decide whether the confounded or the causal model provides the simplest factorization of the graphical model. Finally, we present methods for dataset harmonization and study their ability to remove bias in imaging features. In particular, we propose an extension of the recently introduced ComBat algorithm to control for global variation across image features, inspired by adjusting for unknown population stratification in genetics. Overall, our results demonstrate that harmonization can reduce dataset-specific information in image features. Further, confounding bias can be reduced and even turned into a causal relationship. However, harmonization also requires caution as it can easily remove relevant subject-specific information. Code is available at https://github.com/ai-med/Dataset-Bias.

42 ENGINEERING↗
DOE OSTI2024

Combining compositional data sets introduces error in covariance network reconstruction

Microbial communities are diverse biological systems that include taxa from across multiple kingdoms of life. Notably, interactions between bacteria and fungi play a significant role in determining community structure. However, these statistical associations across kingdoms are more difficult to infer than intra-kingdom associations due to the nature of the data involved using standard network inference techniques. We quantify the challenges of cross-kingdom network inference from both theoretical and practical points of view using synthetic and real-world microbiome data. We detail the theoretical issue presented by combining compositional data sets drawn from the same environment, e.g. 16S and ITS sequencing of a single set of samples, and we survey common network inference techniques for their ability to handle this error. We then test these techniques for the accuracy and usefulness of their intra- and inter-kingdom associations by inferring networks from a set of simulated samples for which a ground-truth set of associations is known. We show that while the two methods mitigate the error of cross-kingdom inference, there is little difference between techniques for key practical applications including identification of strong correlations and identification of possible keystone taxa (i.e. hub nodes in the network). Furthermore, we identify a signature of the error caused by transkingdom network inference and demonstrate that it appears in networks constructed using real-world environmental microbiome data.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Combining compositional data sets introduces error in covariance network reconstruction

Microbial communities are diverse biological systems that include taxa from across multiple kingdoms of life. Notably, interactions between bacteria and fungi play a significant role in determining community structure. However, these statistical associations across kingdoms are more difficult to infer than intra-kingdom associations due to the nature of the data involved using standard network inference techniques. We quantify the challenges of cross-kingdom network inference from both theoretical and practical points of view using synthetic and real-world microbiome data. We detail the theoretical issue presented by combining compositional data sets drawn from the same environment, e.g. 16S and ITS sequencing of a single set of samples, and we survey common network inference techniques for their ability to handle this error. We then test these techniques for the accuracy and usefulness of their intra- and interkingdom associations by inferring networks from a set of simulated samples for which a ground-truth set of associations is known. We show that while the two methods mitigate the error of cross-kingdom inference, there is little difference between techniques for key practical applications including identification of strong correlations and identification of possible keystone taxa (i.e. hub nodes in the network). Furthermore, we identify a signature of the error caused by transkingdom network inference and demonstrate that it appears in networks constructed using real-world environmental microbiome data.

59 BASIC BIOLOGICAL SCIENCES↗