Search NASA⌕ Search

SEARCH · Search NASA

Results for “data pipelines”

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 937 records · Page 52
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

End-to-end learning of multiple sequence alignments with differentiable Smith–Waterman

Abstract Motivation Multiple sequence alignments (MSAs) of homologous sequences contain information on structural and functional constraints and their evolutionary histories. Despite their importance for many downstream tasks, such as structure prediction, MSA generation is often treated as a separate pre-processing step, without any guidance from the application it will be used for. Results Here, we implement a smooth and differentiable version of the Smith–Waterman pairwise alignment algorithm that enables jointly learning an MSA and a downstream machine learning system in an end-to-end fashion. To demonstrate its utility, we introduce SMURF (Smooth Markov Unaligned Random Field), a new method that jointly learns an alignment and the parameters of a Markov Random Field for unsupervised contact prediction. We find that SMURF learns MSAs that mildly improve contact prediction on a diverse set of protein and RNA families. As a proof of concept, we demonstrate that by connecting our differentiable alignment module to AlphaFold2 and maximizing predicted confidence, we can learn MSAs that improve structure predictions over the initial MSAs. Interestingly, the alignments that improve AlphaFold predictions are self-inconsistent and can be viewed as adversarial. This work highlights the potential of differentiable dynamic programming to improve neural network pipelines that rely on an alignment and the potential dangers of optimizing predictions of protein sequences with methods that are not fully understood. Availability and implementation Our code and examples are available at: https://github.com/spetti/SMURF. Supplementary information Supplementary data are available at Bioinformatics online.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Strangers in a foreign land: ‘Yeastizing’ plant enzymes

Abstract Expressing plant metabolic pathways in microbial platforms is an efficient, cost‐effective solution for producing many desired plant compounds. As eukaryotic organisms, yeasts are often the preferred platform. However, expression of plant enzymes in a yeast frequently leads to failure because the enzymes are poorly adapted to the foreign yeast cellular environment. Here, we first summarize the current engineering approaches for optimizing performance of plant enzymes in yeast. A critical limitation of these approaches is that they are labour‐intensive and must be customized for each individual enzyme, which significantly hinders the establishment of plant pathways in cellular factories. In response to this challenge, we propose the development of a cost‐effective computational pipeline to redesign plant enzymes for better adaptation to the yeast cellular milieu. This proposition is underpinned by compelling evidence that plant and yeast enzymes exhibit distinct sequence features that are generalizable across enzyme families. Consequently, we introduce a data‐driven machine learning framework designed to extract ‘yeastizing’ rules from natural protein sequence variations, which can be broadly applied to all enzymes. Additionally, we discuss the potential to integrate the machine learning model into a full design‐build‐test cycle.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

SRF Cavity Fault Classification and Prediction at Jefferson Lab

Over the last few years several machine learning projects at Jefferson Lab have had a common focus to optimize operation of superconducting RF (SRF) cavities in the Continuous Electron Beam Accelerator Facility (CEBAF). In this talk we highlight work to identify and classify types of faults from C100-type cavities and then to extend those capabilities to provide real-time fault prediction. Early prediction may enable mitigation strategies to prevent some types of faults. In our approach we apply a two-step fault prediction pipeline. In the first step, a model distinguishes between faulty and normal signals. In the second step, signals flagged as faulty by the first model are classified into one of seven fault types based on learned signatures in the data. Initial results show that our model can successfully predict most fault types 200 ms before onset. In additional to model performance, we also highlight challenges in working with real-world data and challenges for deploying models.

Tennant, C.↗
DOE OSTI2024

DiffLense: a conditional diffusion model for super-resolution of gravitational lensing data

Abstract Gravitational lensing data is frequently collected at low resolution due to instrumental limitations and observing conditions. Machine learning-based super-resolution techniques offer a method to enhance the resolution of these images, enabling more precise measurements of lensing effects and a better understanding of the matter distribution in the lensing system. This enhancement can significantly improve our knowledge of the distribution of mass within the lensing galaxy and its environment, as well as the properties of the background source being lensed. Traditional super-resolution techniques typically learn a mapping function from lower-resolution to higher-resolution samples. However, these methods are often constrained by their dependence on optimizing a fixed distance function, which can result in the loss of intricate details crucial for astrophysical analysis. In this work, we introduce DiffLense , a novel super-resolution pipeline based on a conditional diffusion model specifically designed to enhance the resolution of gravitational lensing images obtained from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). Our approach adopts a generative model, leveraging the detailed structural information present in Hubble space telescope (HST) counterparts. The diffusion model, trained to generate HST data, is conditioned on HSC data pre-processed with denoising techniques and thresholding to significantly reduce noise and background interference. This process leads to a more distinct and less overlapping conditional distribution during the model’s training phase. We demonstrate that DiffLense outperforms existing state-of-the-art single-image super-resolution techniques, particularly in retaining the fine details necessary for astrophysical analyses.

Computer Science↗
DOE OSTI2024

Automating the Analysis of Large Language Models Responses through Zero-Shot Question Answering

Recent advancements in Large Language Models (LLMs) have shown significant potential in various applications, yet their evaluation, particularly in zero-shot question answering scenarios, remains a challenging task. In this study, our objective was to explore precision metrics for Large Language Models (LLM) and design and implement a software pipeline to automatically evaluate LLMs' outputs under zero-shot question answering. Zero-shot question answering involves a model providing answers to questions about topics it hasn't seen during training. It leverages the principles of zero-shot learning by relying on semantic understanding and generalization from related knowledge. The data used was metadata from medical databases on congenital heart disease. We explored eleven LLM metrics and selected three for our evaluation: BLEU, BERTScore, and MoverScore. BLEU calculates a score based on the overlap of n-grams (contiguous sequences of n items, typically words) between the machine-generated translation and the reference translations. Higher BLEU scores indicate better correspondence between the machine-generated and human-generated translations. BERTScore is a metric used to evaluate the quality of machine-generated text by measuring the similarity of token embeddings produced by BERT (Bidirectional Encoder Representations from Transformers) between the generated text and reference text. MoverScore is a metric that quantifies the dissimilarity between the distributions of word embeddings from machine-generated text and reference text, emphasizing semantic similarity over exact token overlap. We also introduced HBKI, a composite metric summarizing these approaches. We tested five models —GPT-3, Llama-2, Gemini 1.5 Pro, Solar 10.7B, and Mixtral-8x7b. Our software pipeline, designed and implemented using Object-Oriented Programming principles, allows users to customize the selection and extraction of features for topics of interest in their own research. Our results show that MoverScore delivered the most precise evaluation of the LLM's outputs, while Mixtral-8x7b achieved the best overall performance in extracting metadata from the databases.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Providing Geospatial Intelligence through a Scalable Imagery Pipeline

This chapter describes ORNL’s (Oak Ridge National Laboratory’s) contributions to imagery preprocessing for geospatial intelligence research and development (R&D) in four sections. First, we discuss challenges involved in building an effective imagery preprocessing workflow and the world-class high-performance computing (HPC) resources at ORNL available to process petabytes of imagery data. Second, we highlight how we developed imagery preprocessing tools over three decades while paving the way for our current cutting-edge machine learning and computer vision algorithms that are impacting humanitarian and disaster response efforts. Third, we discuss how PIPE modules work together to turn raw images into analysis-ready datasets. Fourth, we look toward the future and discuss planned advancements to PIPE and computing trends that will affect geospatial intelligence R&D.

Reith, Andrew↗
DOE OSTI2026

Automatic Lane-Level Road Network Extraction from Aerial Imagery for Transportation Digital Twins

Accurate road networks are essential for credible traffic microsimulation and transportation digital twins, yet high-definition maps are often difficult to obtain due to limited availability, high cost, or proprietary restrictions. Some build networks from crowdsourced data, such as OpenStreetMap, but these sources often contain geometric and semantic inconsistencies. Others create networks manually, a process that is labor-intensive and difficult to scale. To address these limitations, this work presents an end-to-end pipeline that automatically extracts georeferenced, lane-level road networks from publicly available high-resolution satellite imagery and converts them into simulation-ready assets. The developed end-to-end pipeline has three primary modules: (1) A computer-vision-based module first detects directed lane geometries and intersection layouts. (2) A heuristic-based topology construction module then identifies approach and exit legs and establishes conflict-free lane-to-lane connections. (3) Finally, an automatic simulation-building module converts the extracted network into standard formats, e.g., OpenDRIVE, and generates routable SUMO networks. The framework supports both complete network construction from scratch and local-scale refinement of existing networks through lane-count correction, transition recovery, and geometric regularization. The proposed pipeline provides a practical pathway to generate traffic simulation networks from satellite imagery, significantly reducing manual reconstruction effort and enabling scalable, continuously updated transportation digital twins.

Guo, Hetian [University of Georgia, Athens] (ORCID↗
DOE OSTI2023

Sky subtraction in an era of low surface brightness astronomy

ABSTRACT The Vera C. Rubin Observatory Wide-Fast Deep sky survey will reach unprecedented surface brightness depths over tens of thousands of square degrees. Surface brightness photometry has traditionally been a challenge. Current algorithms which combine object detection with sky estimation systematically oversubtract the sky, biasing surface brightness measurements at the faint end and destroying or severely compromising low surface brightness light. While it has recently been shown that properly accounting for undetected faint galaxies and the wings of brighter objects can in principle recover a more accurate sky estimate, this has not yet been demonstrated in practice. Obtaining a consistent spatially smooth underlying sky estimate is particularly challenging in the presence of representative distributions of bright and faint objects. In this paper, we use simulations of crowded and uncrowded fields designed to mimic Hyper Suprime-Cam data to perform a series of tests on the accuracy of the recovered sky. Dependence on field density, galaxy type, and limiting flux for detection are all considered. Several photometry packages are utilized: source extractor, gnuastro, and the LSST science pipelines. Each is configured in various modes, and their performance at extreme low surface brightness analysed. We find that the combination of the source extractor software package with novel source model masking techniques consistently produce extremely faint output sky estimates, by up to an order of magnitude, as well as returning high fidelity output science catalogues.

Astronomy & Astrophysics↗
DOE OSTI2021

CodeFlow: A Code Generation System for Flash-X Orchestration Runtime

We propose the CodeFlow toolchain for Flash-X that realizes the “recipe-to-source” code transformation for Flash-X simulations and that is necessary to achive performance portability. We design a high-level language to express operations of simulations in so-called recipes, which are given as input to the toolchain. The tools of the CodeFlow pipeline include code transformation with tree-based source code representation techniques and code orchestration and generation based on control flow graphs. The generated source code utilizes a new runtime, developed for Flash-X, that orchestrates dynamic and asynchronous data movement and task execution. The functionality of CodeFlow is demonstrated using a hydrodynamic problem with a strong shock.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Performance of methods for SARS-CoV-2 variant detection and abundance estimation within mixed population samples

The accurate identification of SARS-CoV-2 (SC2) variants and estimation of their abundance in mixed population samples (e.g., air or wastewater) is imperative for successful surveillance of community level trends. Assessing the performance of SC2 variant composition estimators (VCEs) should improve our confidence in public health decision making. Here, we introduce a linear regression based VCE and compare its performance to four other VCEs: two re-purposed DNA sequence read classifiers (Kallisto and Kraken2), a maximum-likelihood based method (Lineage deComposition for Sars-Cov-2 pooled samples (LCS)), and a regression based method (Freyja). We simulated DNA sequence datasets of known variant composition from both Illumina and Oxford Nanopore Technologies (ONT) platforms and assessed the performance of each VCE. We also evaluated VCEs performance using publicly available empirical wastewater samples collected for SC2 surveillance efforts. Bioinformatic analyses were performed with a custom NextFlow workflow (C-WAP, CFSAN Wastewater Analysis Pipeline). Relative root mean squared error (RRMSE) was used as a measure of performance with respect to the known abundance and concordance correlation coefficient (CCC) was used to measure agreement between pairs of estimators. Based on our results from simulated data, Kallisto was the most accurate estimator as it had the lowest RRMSE, followed by Freyja. Kallisto and Freyja had the most similar predictions, reflected by the highest CCC metrics. We also found that accuracy was platform and amplicon panel dependent. For example, the accuracy of Freyja was significantly higher with Illumina data compared to ONT data; performance of Kallisto was best with ARTICv4. However, when analyzing empirical data there was poor agreement among methods and variations in the number of variants detected (e.g., Freyja ARTICv4 had a mean of 2.2 variants while Kallisto ARTICv4 had a mean of 10.1 variants). This work provides an understanding of the differences in performance of a number of VCEs and how accurate they are in capturing the relative abundance of SC2 variants within a mixed sample (e.g., wastewater). Such information should help officials gauge the confidence they can have in such data for informing public health decisions.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

PTArcade

This is a lightweight manual for PTArcade, a wrapper of ENTERPRISE and ceffyl that allows for easy implementation of new-physics searches in PTA data. In this manual, we describe how to get PTArcade installed (either on your local machine or an HPC cluster). We discuss how to define a stochastic or deterministic signal and how PTArcade implements these signals in PTA-analysis pipelines. Finally, we show how to handle and analyze the PTArcade output using a series of utility functions that come together with PTArcade.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Workflow for High-throughput Screening of Enzyme Mutant Libraries Using Matrix-assisted Laser Desorption/Ionization Mass Spectrometry Analysis of Escherichia coli Colonies

High-throughput molecular screening of microbial colonies and DNA libraries are critical procedures that enable applications such as directed evolution, functional genomics, microbial identification, and creation of engineered microbial strains to produce high-value molecules. A promising chemical screening approach is the measurement of products directly from microbial colonies via optically guided matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). Measuring the compounds from microbial colonies bypasses liquid culture with a screen that takes approximately 5 s per sample. We describe a protocol combining a dedicated informatics pipeline and sample preparation method that can prepare up to 3,000 colonies in under 3 h. The screening protocol starts from colonies grown on Petri dishes and then transferred onto MALDI plates via imprinting. The target plate with the colonies is imaged by a flatbed scanner and the colonies are located via custom software. The target plate is coated with MALDI matrix, MALDI-MS analyzes the colony locations, and data analysis enables the determination of colonies with the desired biochemical properties. This workflow screens thousands of colonies per day without requiring additional automation. The wide chemical coverage and the high sensitivity of MALDI-MS enable diverse screening projects such as modifying enzymes and functional genomics surveys of gene activation/inhibition libraries.

Choe, Kisurb↗
DOE OSTI2021

Interpretable artificial intelligence and exascale molecular dynamics simulations to reveal kinetics: Applications to Alzheimer's disease

The rapid increase in computing power, especially with the integration of graphics processing units, has dramatically increased the capabilities of molecular dynamics simulations. To date, these capabilities extend from running very long simulations (tens to hundreds of microseconds) to thousands of short simulations. However, the expansive data generated in these simulations must be made interpretable not only by the investigator who performs them but also by others as well. Here, we demonstrate how integrating learning techniques, such as artificial intelligence, machine learning, and neural networks, into analysis pipelines can reveal the kinetics of Alzheimer's disease (AD) protein aggregation. Finally, we review select AD targets, describe current simulation methods, and introduce learning concepts and their application in AD, highlighting limitations and potential solutions.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Neural network accelerator for quantum control

Efficient quantum control is necessary for practical quantum computing implementations with current technologies. Conventional algorithms for determining optimal control parameters are computationally expensive, largely excluding them from use outside of the simulation. Existing hardware solutions structured as lookup tables are imprecise and costly. By designing a machine learning model to approximate the results of traditional tools, a more efficient method can be produced. Such a model can then be synthesized into a hardware accelerator for use in quantum systems. In this study, we demonstrate a machine learning algorithm for predicting optimal pulse parameters. This algorithm is lightweight enough to fit on a low-resource FPGA and perform inference with a latency of 175 ns and pipeline interval of 5 ns with > 0.99 gate fidelity. In the long term, such an accelerator could be used near quantum computing hardware where traditional computers cannot operate, enabling quantum control at a reasonable cost at low latencies without incurring large data bandwidths outside of the cryogenic environment.

43 PARTICLE ACCELERATORS↗
DOE OSTI2022

U.S. Hydropower Workforce: Challenges and Opportunities

Over the past four years, the National Renewable Energy Laboratory (NREL) has engaged with the hydropower industry, academia, and students to understand the perspectives and challenges of the U.S. hydropower workforce. This engagement and research were sponsored by the U.S. Department of Energy's (DOE's) Water Power Technologies Office as part of an NREL-led project—Water Power Science, Technology, Engineering, and Mathematics (STEM) to Workforce—which has the goal of developing tools and programs to strengthen the water power workforce pipeline. This report shares findings from that research and provides an update on U.S. hydropower workforce trends as a follow-on to NREL's 2019 report, Workforce Development for U.S. Hydropower: Key Trends and Findings (Keyser and Tegen 2019). Recent data from NREL, the Hydropower Foundation's survey efforts, and the hydropower industry are presented.

13 HYDRO ENERGY↗
DOE OSTI2025

A Preferences Corpus and Annotation Scheme for Human-Guided Alignment of Time-Series GPTs

The process of time-series forecasting such as predicting trajectories of silicon content in blast furnaces is a difficult task. Most time-series approaches today focus on scalar-type MSE loss optimization. This optimization approach, while widely common, could benefit from the use of human expert or process-level preferences. In this paper, we introduce a novel alignment and fine-tuning approach that involves learning from a corpus of preferred and dis-preferred time-series prediction trajectories. Our contributions include (1) a preference annotation pipeline for time-series forecasts, (2) the application of Score-based Preference Optimization (SPO) to train decoder-only transformers from preferences, and (3) results showing improvements in forecast quality. The approach is validated on both proprietary blast furnace data and the UCI Appliances Energy dataset. The proposed preference corpus and training strategy offer a new option for fine-tuning sequence models in industrial settings.

DPO↗