Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Decoder”

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 1,189 records · Page 66

HydraGNN v4.0

The new version of HydraGNN v4.0 provides additional core capabilities, such as: Inclusion of multi-body atomistic cluster expansion MACE, polarizable atom interaction neural network PAINN, and equivariant principal neighborhood aggregation (PNAEq) among the message passing layers supported -Inclusion of graph transformers to directly model long-range interactions between nodes that are distant in the graph topology Integration of graph transformers with message passing layers by combining the graph embedding generated by the two mechanisms, which allows for an improved expressivity of the HydraGNN architecture Improved re-implementation of multi-task learning (MTL) to allow its use for stabilized training across imbalanced, multi-source, multi-fidelity data Introduction of multi-task parallelism, a newly proposed type of model parallelism specifically for MTL architectures, which allows to dispatch different output decoding heads to different GPU devices Integration of multi-task parallelism with pre-existing distributed data parallelism to enable a 2D parallelization for distributed training Improved portability of the distributed training across Intel GPUs, which has been testes on ALCF exascale supercomputer Aurora Inclusion of 2-level fine-grained energy profilers portable across NVIDIA, AMD, and Intel GPUs to monitor the power and energy consumption associated with different functions executed by the HydraGNN code during data pre-load and training Restructuring of previous examples and inclusion of new sets of examples to illustrate the download, preprocess, and training of HydraGNN models on new large-scale open-source datasets for atomistic materials modeling (e.g., Alexandria, Transition1x, OMat24, OMol25)

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗

Assessing Tropical Pacific-induced Predictability of Southern California Precipitation Using a Novel Multi-input Multi-output Autoencoder

We construct a novel Multi-Input Multi-Output Autoencoder-decoder (MIMO-AE) to capture the non-linear relationship of Southern California precipitation and tropical Pacific Ocean sea surface temperature. The MIMO-AE is trained on both monthly TP-SST and SC-PRECIP anomalies simultaneously. The co-variability of the two fields in the MIMO-AE shared nonlinear latent space can be condensed into an index, termed the MIMO-AE index. We use a transfer learning approach to train a MIMO-AE on the combined dataset of 100 years of output from a historical simulation with the Energy Exascale Earth Systems Model version 1 and a segment of observational data. We further use Long Short-Term Memory networks to assess sub-seasonal predictability of SC-PRECIP using the MIMO-AE index. We find that the MIMO-AE index provides enhanced predictability of SC-PRECIP for a lead-time of up-to four months as compared to Niño 3.4 index and the El Niño Southern Oscillation Longitudinal Index.

54 ENVIRONMENTAL SCIENCES↗

An explainable variational autoencoder model for three-dimensional acoustic emission source localization in hollow cylindrical structures

We introduce an explainable variational autoencoder for three-dimensional (3D) localization of acoustic emission sources in hollow cylindrical structures, with an unsupervised approach. This research capitalizes on multi-arrival waveforms generated by helical path propagation in cylindrical geometries to enable efficient two-receiver localization. By integrating the modal characteristics of Lamb modes under multi-path conditions, we demonstrate that two sets of time-of-arrival differences and peak amplitudes extracted from one receiver can serve as effective localization features. This initial approach identifies four potential source locations, highlighting the feasibility of two-receiver source localization using traditional feature extraction methods. However, direct extraction can be challenging when mode overlaps occur, complicating the localization process. To address this, our work proposes a novel waveform-based method. This method leverages the consistent dispersion characteristics within isotropic materials, where each unique combination of mode arrival times and peak amplitudes constructs a distinct waveform. This distinctiveness overcomes the ambiguities associated with mode overlaps, significantly enhancing the method’s precision and robustness. Our approach adopts a data-driven strategy for waveform-based localization using variational autoencoder (VAE). VAE discerns waveform patterns for localization, while also addressing data uncertainties. The VAE’s encoder and decoder networks capture the localization process and the source’s influence on waveform generation, respectively, guiding latent variables to segregate waveforms by source in the latent space. The design of the learning process focuses on specific localization characteristics to enhance result explainability. Localization predictions are generated by projecting test waveforms, not included in the training set, onto a trained latent space. The prediction is determined using a nearest-neighbor approach based on the closest latent representation of a source. Validation with pencil-lead-break tests on a metallic pipe confirmed our method’s effectiveness, achieving an averaged 3D localization accuracy of 0.84.

Lee, Guan-Wei↗

Harnessing redox proteomics to study metabolic regulation and stress response in lignin-fed Rhodococci

Abstract Background Rhodococci are studied for their bacterial ligninolytic capabilities and proclivity to accumulate lipids. Lignin utilization is a resource intensive process requiring a variety of redox active enzymes and cofactors for degradation as well as defense against the resulting toxic byproducts and oxidative conditions. Studying enzyme expression and regulation between carbon sources will help decode the metabolic rewiring that stymies lignin to lipid conversion in these bacteria. Herein, a redox proteomics approach was applied to investigate a fundamental driver of carbon catabolism and lipid anabolism: redox balance. Results A consortium of Rhodococcus strains was employed in this study given its higher capacity for lignin degradation compared to monocultures. This consortium was grown on glucose vs. lignin under nitrogen limitation to study the importance of redox balance as it relates to nutrient availability. A modified bottom–up proteomics workflow was harnessed to acquire a general relationship between protein abundance and protein redox states. Global proteomics results affirm differential expression of enzymes involved in sugar metabolism vs. those involved in lignin degradation and aromatics metabolism. As reported previously, several enzymes in the lipid biosynthetic pathways were downregulated, whereas many involved in β-oxidation were upregulated. Interestingly, proteins involved in oxidative stress response were also upregulated perhaps in response to lignin degradation and aromatics catabolism, which require oxygen and reactive oxygen species and generate toxic byproducts. Enzymes displaying little-to-no change in abundance but differences in redox state were observed in various pathways for carbon utilization (e.g., β‑ketoadipate pathway), lipid metabolism, as well as nitrogen metabolism (e.g., purine scavenging/synthesis), suggesting potential mechanisms of redox-dependent regulation of metabolism. Conclusions Efficient lipid production requires a steady carbon and energy flux while balancing fundamental requirements for enzyme production and cell maintenance. For lignin, we theorize that this balance is difficult to establish due to resource expenditure for enzyme production and stress response. This is supported by significant changes to protein abundances and protein cysteine oxidation in various metabolic pathways and redox processes.

09 BIOMASS FUELS↗

Hidden Markov Model: a shortest unique representative approach to detect the protein toxins, virulence factors and antibiotic resistance genes

Objective: Currently, next generation sequencing (NGS) is widely used to decode potential novel or variant pathogens both in emergent outbreaks and in routine clinical practice. However, the efficient identification of novel or diverged pathogenomic compositions remains a big challenge. It is especially true for short DNA sequence fragments from NGS, since sequence similarity searching is vulnerable to false negatives or false positives, as is mismatching or matching with unrelated proteins. Therefore, this study aimed to establish a bioinformatics approach that can generate unique motif sequences for profiling searching, resulting in high specificity and sensitivity. Results: In this study, we introduced a Shortest Unique Representative Hidden Markov Model (HMM) approach to identify bacterial toxin, virulence factor (VF), and antimicrobial resistance (AR) in short sequence reads. We first construct unique representative domain sequences of toxin genes, VFs, and ARs to avoid potential false positives, and then to use HMM models to accurately identify potential toxin, VF, and AR fragments. The benchmark shows this approach can achieve relatively high specificity and sensitivity if the appropriate cutoff value is applied. Our approach can be used to recognize the protein sequences of known toxins and pathogens, identifies their common characteristics and then searches for similar sequences in other organisms.

59 BASIC BIOLOGICAL SCIENCES↗

Time-series metagenomics reveals changing protistan ecology of a temperate dimictic lake

Abstract Background Protists, single-celled eukaryotic organisms, are critical to food web ecology, contributing to primary productivity and connecting small bacteria and archaea to higher trophic levels. Lake Mendota is a large, eutrophic natural lake that is a Long-Term Ecological Research site and among the world’s best-studied freshwater systems. Metagenomic samples have been collected and shotgun sequenced from Lake Mendota for the last 20 years. Here, we analyze this comprehensive time series to infer changes to the structure and function of the protistan community and to hypothesize about their interactions with bacteria. Results Based on small subunit rRNA genes extracted from the metagenomes and metagenome-assembled genomes of microeukaryotes, we identify shifts in the eukaryotic phytoplankton community over time, which we predict to be a consequence of reduced zooplankton grazing pressures after the invasion of a invasive predator (the spiny water flea) to the lake. The metagenomic data also reveal the presence of the spiny water flea and the zebra mussel, a second invasive species to Lake Mendota, prior to their visual identification during routine monitoring. Furthermore, we use species co-occurrence and co-abundance analysis to connect the protistan community with bacterial taxa. Correlation analysis suggests that protists and bacteria may interact or respond similarly to environmental conditions. Cryptophytes declined in the second decade of the timeseries, while many alveolate groups (e.g., ciliates and dinoflagellates) and diatoms increased in abundance, changes that have implications for food web efficiency in Lake Mendota. Conclusions We demonstrate that metagenomic sequence-based community analysis can complement existing efforts to monitor protists in Lake Mendota based on microscopy-based count surveys. We observed patterns of seasonal abundance in microeukaryotes in Lake Mendota that corroborated expectations from other systems, including high abundance of cryptophytes in winter and diatoms in fall and spring, but with much higher resolution than previous surveys. Our study identified long-term changes in the abundance of eukaryotic microbes and provided context for the known establishment of an invasive species that catalyzes a trophic cascade involving protists. Our findings are important for decoding potential long-term consequences of human interventions, including invasive species introduction.

59 BASIC BIOLOGICAL SCIENCES↗

Multi-site-specific 16S rRNA Methyltransferase RsmF from Thermus thermophilus

Cells devote a significant effort toward the production of multiple modified nucleotides in rRNAs, which fine tune the ribosome function. Here, we report that two methyltransferases, RsmB and RsmF, are responsible for all four 5-methylcytidine (m{sup 5}C) modifications in 16S rRNA of Thermus thermophilus. Like Escherichia coli RsmB, T. thermophilus RsmB produces m{sup 5}C967. In contrast to E. coli RsmF, which introduces a single m{sup 5}C1407 modification, T. thermophilus RsmF modifies three positions, generating m{sup 5}C1400 and m{sup 5}C1404 in addition to m{sup 5}C1407. These three residues are clustered near the decoding site of the ribosome, but are situated in distinct structural contexts, suggesting a requirement for flexibility in the RsmF active site that is absent from the E. coli enzyme. Two of these residues, C1400 and C1404, are sufficiently buried in the mature ribosome structure so as to require extensive unfolding of the rRNA to be accessible to RsmF. In vitro, T. thermophilus RsmF methylates C1400, C1404, and C1407 in a 30S subunit substrate, but only C1400 and C1404 when naked 16S rRNA is the substrate. The multispecificity of T. thermophilus RsmF is potentially explained by three crystal structures of the enzyme in a complex with cofactor S-adenosyl-methionine at up to 1.3 {angstrom} resolution. In addition to confirming the overall structural similarity to E. coli RsmF, these structures also reveal that key segments in the active site are likely to be dynamic in solution, thereby expanding substrate recognition by T. thermophilus RsmF.

59 BASIC BIOLOGICAL SCIENCES↗

Five-degrees-of-freedom diffractive laser encoder

Linear laser encoders have been widely used for precision positioning control of a linear stage. We develop a five-degrees-of-freedom (5-DOF) laser linear encoder to simultaneously measure the position, straightness, pitch, roll, and yaw errors along one moving axis. This study integrates the circular polarized interferometric technique with the three-dimensional diffracted ray-tracing method to develop a novel laser encoder with 5-DOF. The phases encoded within the +1 and -1 order diffraction lights reflected from the diffraction grating are decoded by the circular polarized interferometric technique to measure the linear displacement when the diffraction grating moves. The three-dimensional diffracted ray tracing of the +1- and -1-order diffraction lights induced by the motion errors of the moved grating were analyzed to calculate the other motion errors based on the detection of light spots on two quadrant photodiode detectors. The period of the grating is 0.83 {mu}m and the experimental results show that the measurement accuracy was better than {+-}0.3 {mu}m/{+-}41 {mu}m for straightness, {+-}1 arc sec.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Linear diffraction grating interferometer with high alignment tolerance and high accuracy

We present an innovative structure of a linear diffraction grating interferometer as a long stroke and nanometer resolution displacement sensor for any linear stage. The principle of this diffractive interferometer is based on the phase information encoded by the {+-}1st order beams diffracted by a holographic grating. Properly interfering these two beams leads to modulation similar to a Doppler frequency shift that can be translated to displacement measurements via phase decoding. A self-compensation structure is developed to improve the alignment tolerance. LightTool analysis shows that this new structure is completely immune to alignment errors of offset, standoff, yaw, and roll. The tolerance of the pitch is also acceptable for most installation conditions. In order to compact the structure and improve the signal quality, a new optical bonding technology by mechanical fixture is presented so that the miniature optics can be permanently bonded together without an air gap in between. For the output waveform signals, a software module is developed for fast real-time pulse counting and phase subdivision. A laser interferometer HP5529A is employed to test the repeatability of the whole system. Experimental data show that within 15 mm travel length, the repeatability is within 15 nm.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Orthogonal spatial coding with stimulated parametric down-conversion

Orthogonal optical coding is widely used in classical multi-user communication networks. Using the phase conjugation property of stimulated parametric down-conversion, we extend the current time-domain orthogonal optical coding scheme to the spatial domain to encode and decode image information. In this process, the idler beam inherits the complex conjugate of the field information encoded in the seed beam. An encoding phase mask introduced onto the input seed beam blurs the image transferred to the idler. The original image is restored by passing the coded transferred image through a corrective phase mask placed in the momentum space of the idler beam. We expect that this scheme can also inspire new techniques in secure image transmission, aberration cancellation, and frequency conversion imaging.

Xu, Yang↗

Real-time multiplexed digital cavity-enhanced spectroscopy

Cavity-enhanced spectroscopy is a sensitive optical absorption technique but one where the practical applications have been limited to studying small wavelength ranges. In addition, this Letter shows that wideband operation can be achieved by combining techniques usually reserved for the communications community with that of cavity-enhanced spectroscopy, producing a multiplexed real-time cavity-enhanced spectrometer. We use multiple collinear laser sources operating asynchronously and simultaneously while being detected on a single photodetector. This is synonymous with radio frequency (RF) cellular systems in which signals are detected on a single antenna but decoded uniquely. Here, we demonstrate results with spectra of methyl salicylate and show parts-per-billion per root hertz sensitivity measured in real-time.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Image translation for single-shot focal tomography

Focus and depth of field are conventionally addressed by adjusting longitudinal lens position. More recently, combinations of deliberate blur and computational processing have been used to extend depth of field. Here we show that dynamic control of transverse and longitudinal lens position can be used to decode focus and extend depth of field without degrading static resolution. Our results suggest that optical image stabilization systems may be used for autofocus, extended depth of field, and 3D imaging.

47 OTHER INSTRUMENTATION↗

High-dimensional quantum key distribution using orbital angular momentum of single photons from a colloidal quantum dot at room temperature

High-dimensional quantum key distribution (HDQKD) is a promising avenue to address the inherent limitations of basic quantum key distribution (QKD) protocols. However, experimental realizations of HDQKD to date have relied on indeterministic photon sources that limit the achievable key rate. In this paper, we demonstrate a full emulation of a HDQKD system using a single colloidal giant quantum dot (gQD) as a deterministic, compact, and room-temperature single-photon source (SPS). We demonstrate a practical protocol by encoding information in a high-dimensional space (d = 3) of the orbital angular momentum of the photons. Our experimental configuration incorporates two spatial light modulators for encoding and decoding the spatial information carried by individual photons. Our experimental demonstration establishes the feasibility of utilizing high radiative quantum yield gQDs as practical SPSs for HDQKD. We also experimentally demonstrate surpassing the traditional d = 2 QKD capacity with comparable error rates, indicating a significant improvement in performance while maintaining reliability.

36 MATERIALS SCIENCE↗

Quantitative Analysis of the Drosophila Segmentation Regulatory Network Using Pattern Generating Potentials

Cis-regulatory modules that drive precise spatial-temporal patterns of gene expression are central to the process of metazoan development. We describe a new computational strategy to annotate genomic sequences based on their ‘‘pattern generating potential’’ and to produce quantitative descriptions of transcriptional regulatory networks at the level of individual protein-module interactions. We use this approach to convert the qualitative understanding of interactions that regulate Drosophila segmentation into a network model in which a confidence value is associated with each transcription factor-module interaction. Sequence information from multiple Drosophila species is integrated with transcription factor binding specificities to determine conserved binding site frequencies across the genome. These binding site profiles are combined with transcription factor expression information to create a model to predict module activity patterns. This model is used to scan genomic sequences for the potential to generate all or part of the expression pattern of a nearby gene, obtained from available gene expression databases. Interactions between individual transcription factors and modules are inferred by a statistical method to quantify a factor’s contribution to the module’s pattern generating potential. We use these pattern generating potentials to systematically describe the location and function of known and novel cis-regulatory modules in the segmentation network, identifying many examples of modules predicted to have overlapping expression activities. Surprisingly, conserved transcription factor binding site frequencies were as effective as experimental measurements of occupancy in predicting module expression patterns or factor-module interactions. Thus, unlike previous module prediction methods, this method predicts not only the location of modules but also their spatial activity pattern and the factors that directly determine this pattern. As databases of transcription factor specificities and in vivo gene expression patterns grow, analysis of pattern generating potentials provides a general method to decode transcriptional regulatory sequences and networks.

59 BASIC BIOLOGICAL SCIENCES↗

Deep learning as a tool for neural data analysis: Speech classification and cross-frequency coupling in human sensorimotor cortex

A fundamental challenge in neuroscience is to understand what structure in the world is represented in spatially distributed patterns of neural activity from multiple single-trial measurements. This is often accomplished by learning a simple, linear transformations between neural features and features of the sensory stimuli or motor task. While successful in some early sensory processing areas, linear mappings are unlikely to be ideal tools for elucidating nonlinear, hierarchical representations of higher-order brain areas during complex tasks, such as the production of speech by humans. Here, we apply deep networks to predict produced speech syllables from a dataset of high gamma cortical surface electric potentials recorded from human sensorimotor cortex. We find that deep networks had higher decoding prediction accuracy compared to baseline models. Having established that deep networks extract more task relevant information from neural data sets relative to linear models (i.e., higher predictive accuracy), we next sought to demonstrate their utility as a data analysis tool for neuroscience. We first show that deep network's confusions revealed hierarchical latent structure in the neural data, which recapitulated the underlying articulatory nature of speech motor control. We next broadened the frequency features beyond high-gamma and identified a novel high-gamma-to-beta coupling during speech production. Finally, we used deep networks to compare task-relevant information in different neural frequency bands, and found that the high-gamma band contains the vast majority of information relevant for the speech prediction task, with little-to-no additional contribution from lower-frequency amplitudes. Together, these results demonstrate the utility of deep networks as a data analysis tool for basic and applied neuroscience.

60 APPLIED LIFE SCIENCES↗

Bayesian inference of structured latent spaces from neural population activity with the orthogonal stochastic linear mixing model

The brain produces diverse functions, from perceiving sounds to producing arm reaches, through the collective activity of populations of many neurons. Determining if and how the features of these exogenous variables (e.g., sound frequency, reach angle) are reflected in population neural activity is important for understanding how the brain operates. Often, high-dimensional neural population activity is confined to low-dimensional latent spaces. However, many current methods fail to extract latent spaces that are clearly structured by exogenous variables. This has contributed to a debate about whether or not brains should be thought of as dynamical systems or representational systems. Here, we developed a new latent process Bayesian regression framework, the orthogonal stochastic linear mixing model (OSLMM) which introduces an orthogonality constraint amongst time-varying mixture coefficients, and provide Markov chain Monte Carlo inference procedures. We demonstrate superior performance of OSLMM on latent trajectory recovery in synthetic experiments and show superior computational efficiency and prediction performance on several real-world benchmark data sets. We primarily focus on demonstrating the utility of OSLMM in two neural data sets: μ ECoG recordings from rat auditory cortex during presentation of pure tones and multi-single unit recordings form monkey motor cortex during complex arm reaching. We show that OSLMM achieves superior or comparable predictive accuracy of neural data and decoding of external variables (e.g., reach velocity). Most importantly, in both experimental contexts, we demonstrate that OSLMM latent trajectories directly reflect features of the sounds and reaches, demonstrating that neural dynamics are structured by neural representations. Together, these results demonstrate that OSLMM will be useful for the analysis of diverse, large-scale biological time-series datasets.

59 BASIC BIOLOGICAL SCIENCES↗

Surface-Enhanced Raman Imaging of Intracellular Bioreduction of Chromate in Shewanella oneidensis

This proposed research aims to use novel nanoparticle sensors and spectroscopic tools constituting surface-enhanced Raman spectroscopy (SERS) and Fluorescence Lifetime imaging (FLIM) to study intracellular chemical activities within single bioremediating microorganism. The grand challenge is to develop a mechanistic understanding of chromate reduction and localization by the remediating bacterium Shewanella oneidensis MR-1 by chemical and lifetime imaging. MR-1 has attracted wide interest from the research community because of its potential in reducing multiple chemical and metallic electron acceptors. While several biomolecular approaches to decode microbial reduction mechanisms exist, there is a considerable gap in the availability of sensor platforms to advance research from population-based studies to the single cell level. This study is one of the first attempts to incorporate SERS imaging to address this gap. First, we demonstrate that chromate-decorated nanoparticles can be taken up by cells using TEM and Fluorescence Lifetime imaging to confirm the internalization of gold nanoprobes. Second, we demonstrate the utility of a Raman chemical imaging platform to monitor chromate reduction and localization within single cells. Distinctive differences in Raman signatures of Cr(VI) and Cr(III) enabled their spatial identification within single cells from the Raman images. A comprehensive evaluation of toxicity and cellular interference experiments conducted revealed the inert nature of these probes and that they are non-toxic. Our results strongly suggest the existence of internal reductive machinery and that reduction occurs at specific sites within cells instead of at disperse reductive sites throughout the cell as previously reported. While chromate-decorated gold nanosensors used in this study provide an improved means for the tracking of specific chromate interactions within the cell and on the cell surface, we expect our single cell imaging tools to be extended to monitor the interaction of other toxic metal species.

59 BASIC BIOLOGICAL SCIENCES↗

High-Resolution, Non-Invasive Imaging of Upper Vocal Tract Articulators Compatible with Human Brain Recordings

A complete neurobiological understanding of speech motor control requires determination of the relationship between simultaneously recorded neural activity and the kinematics of the lips, jaw, tongue, and larynx. Many speech articulators are internal to the vocal tract, and therefore simultaneously tracking the kinematics of all articulators is nontrivial-especially in the context of human electrophysiology recordings. Here, we describe a noninvasive, multi-modal imaging system to monitor vocal tract kinematics, demonstrate this system in six speakers during production of nine American English vowels, and provide new analysis of such data. Classification and regression analysis revealed considerable variability in the articulator-to-acoustic relationship across speakers. Non-negative matrix factorization extracted basis sets capturing vocal tract shapes allowing for higher vowel classification accuracy than traditional methods. Statistical speech synthesis generated speech from vocal tract measurements, and we demonstrate perceptual identification. We demonstrate the capacity to predict lip kinematics from ventral sensorimotor cortical activity. These results demonstrate a multi-modal system to non-invasively monitor articulator kinematics during speech production, describe novel analytic methods for relating kinematic data to speech acoustics, and provide the first decoding of speech kinematics from electrocorticography. These advances will be critical for understanding the cortical basis of speech production and the creation of vocal prosthetics.

60 APPLIED LIFE SCIENCES↗