Search NASA⌕ Search

SEARCH · Search NASA

Results for “LABELLING”

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 685 records · Page 38

Chemical Imaging of the Cell Membrane by NanoSIMS

The existence of lipid microdomains and their role in cell membrane organization are currently topics of great interest and controversy. The cell membrane is composed of a lipid bilayer with embedded proteins that can flow along the two-dimensional surface defined by the membrane. Microdomains, known as lipid rafts, are believed to play a central role in organizing this fluid system, enabling the cell membrane to carry out essential cellular processes, including protein recruitment and signal transduction. Lipid rafts are also implicated in cell invasion by pathogens, as in the case of the HIV. Therefore, understanding the role of lipid rafts in cell membrane organization not only has broad scientific implications, but also has practical implications for medical therapies. One of the major limitations on lipid organization research has been the inability to directly analyze lipid composition without introducing artifacts and at the relevant length-scales of tens to hundreds of nanometers. Fluorescence microscopy is widely used due to its sensitivity and specificity to the labeled species, but only the labeled components can be observed, fluorophores can alter the behavior of the lipids they label, and the length scales relevant to imaging cell membrane domains are between that probed by fluorescence resonance energy transfer (FRET) imaging (<10 nm) and the diffraction limit of light. Topographical features can be imaged on this length scale by atomic force microscopy (AFM), but the chemical composition of the observed structures cannot be determined. Immuno-labeling can be used to study the distribution of membrane proteins at high resolution, but not lipid composition. We are using imaging mass spectrometry by secondary ion mass spectrometry (SIMS) in concert with other high resolution imaging methods to overcome these limitations. The experimental approach of this project is to combine molecule-specific stable isotope labeling with high-resolution SIMS using a Cameca NanoSIMS 50 to probe membrane organization and test microdomain hypotheses. The NanoSIMS is an imaging secondary ion mass spectrometer with an unprecedented combination of spatial resolution, sensitivity and mass specificity. It has 50 nm lateral resolution and is capable of detecting 1 in 20 nitrogen atoms while excluding near-neighbor isobaric interferences. The tightly focused cesium ion beam is rastered across the sample to produce simultaneous, quantitative digital images of up to five different masses. By labeling each specific components of a membrane with a unique rare stable isotope or element and mapping the location of the labels with the NanoSIMS, the location of the each labeled component can be determined and quantified. This new approach to membrane composition analysis allows molecular interactions of biological membranes to be probed at length-scales relevant to lipid rafts (10s to 100s of nm) that were not previously possible. Results from our most recent experiments analyzing whole cells will be presented.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Advancements and Application of Microsecond Synchrotron X-ray Footprinting at the Advanced Light Source

The method of synchrotron X-ray protein footprinting (XF-MS) is used to determine protein conformational changes, folding, protein-protein and protein-ligand interactions, providing information which is often difficult to obtain using X-ray crystallography and other common structural biology methods [1–3]. The technique uses comparative in situ labeling of solvent-accessible side chains by highly reactive hydroxyl radicals (•OH) in buffered aqueous solution under different assay conditions. In regions where a protein is folded or binds a partner, these •OH susceptible sites are inaccessible to solvent, and therefore protected from labeling. The •OH are generated by the ionization of water using high-flux-density X-rays. High-flux density is a key factor for XF-MS labeling because obtaining an adequate steady-state concentration of hydroxyl radical within a short irradiation time is necessary to minimize radiation-induced secondary damage and also to overcome various scavenging reactions that reduce the yield of labeled side chains.

47 OTHER INSTRUMENTATION↗

Large fluxes of fatty acids from membranes to triacylglycerol and back during N-deprivation and recovery in Chlamydomonas

Abstract Microalgae accumulate triacylglycerol (TAG) during nutrient deprivation and break it down after nutrient resupply, and these processes involve dramatic shifts in cellular carbon allocation. Due to the importance of algae in the global carbon cycle, and the potential of algal lipids as feedstock for chemical and fuel production, these processes are of both ecophysiological and biotechnological importance. However, the metabolism of TAG is not well understood, particularly the contributions of fatty acids (FAs) from different membrane lipids to TAG accumulation and the fate of TAG FAs during degradation. Here, we used isotopic labeling time course experiments on Chlamydomonas reinhardtii to track FA synthesis and transfer between lipid pools during nitrogen (N)-deprivation and resupply. When cells were labeled before N-deprivation, total levels of label in cellular FAs were unchanged during subsequent N-deprivation and later resupply, despite large fluxes into and out of TAG and membrane lipid pools. Detailed analyses of FA levels and labeling revealed that about one-third of acyl chains accumulating in TAG during N-deprivation derive from preexisting membrane lipids, and in total, at least 45% of TAG FAs passed through membrane lipids at one point. Notably, most acyl chains in membrane lipids during recovery after N-resupply come from TAG. Fluxes of polyunsaturated FAs from plastidic membranes into TAG during N-deprivation were particularly noteworthy. These findings demonstrate a high degree of integration of TAG and membrane lipid metabolism and highlight a role for TAG in storage and supply of membrane lipid components.

Plant Sciences↗

Multiatlas whole heart segmentation of CT data using conditional entropy for atlas ranking and selection

Purpose: Cardiac computed tomography (CT) is widely used in clinical diagnosis of cardiovascular diseases. Whole heart segmentation (WHS) plays a vital role in developing new clinical applications of cardiac CT. However, the shape and appearance of the heart can vary greatly across different scans, making the automatic segmentation particularly challenging. The objective of this work is to develop and evaluate a multiatlas segmentation (MAS) scheme using a new atlas ranking and selection algorithm for automatic WHS of CT data. Research on different MAS strategies and their influence on WHS performance are limited. This work provides a detailed comparison study evaluating the impacts of label fusion, atlas ranking, and sizes of the atlas database on the segmentation performance. Methods: Atlases in a database were registered to the target image using a hierarchical registration scheme specifically designed for cardiac images. A subset of the atlases were selected for label fusion, according to the authors’ proposed atlas ranking criterion which evaluated the performance of each atlas by computing the conditional entropy of the target image given the propagated atlas labeling. Joint label fusion was used to combine multiple label estimates to obtain the final segmentation. The authors used 30 clinical cardiac CT angiography (CTA) images to evaluate the proposed MAS scheme and to investigate different segmentation strategies. Results: The mean WHS Dice score of the proposed MAS method was 0.918 ± 0.021, and the mean runtime for one case was 13.2 min on a workstation. This MAS scheme using joint label fusion generated significantly better Dice scores than the other label fusion strategies, including majority voting (0.901 ± 0.276, p < 0.01), locally weighted voting (0.905 ± 0.0247, p < 0.01), and probabilistic patch-based fusion (0.909 ± 0.0249, p < 0.01). In the atlas ranking study, the proposed criterion based on conditional entropy yielded a performance curve with higher WHS Dice scores compared to the conventional schemes (p < 0.03). In the atlas database study, the authors showed that the MAS using larger atlas databases generated better performance curves than the MAS using smaller ones, indicating larger atlas databases could produce more accurate segmentation. Conclusions: The authors have developed a new MAS framework for automatic WHS of CTA and investigated alternative implementations of MAS. With the proposed atlas ranking algorithm and joint label fusion, the MAS scheme is able to generate accurate segmentation within practically acceptable computation time. This method can be useful for the development of new clinical applications of cardiac CT.

60 APPLIED LIFE SCIENCES↗

Deep active learning for classifying cancer pathology reports

Abstract Background Automated text classification has many important applications in the clinical setting; however, obtaining labelled data for training machine learning and deep learning models is often difficult and expensive. Active learning techniques may mitigate this challenge by reducing the amount of labelled data required to effectively train a model. In this study, we analyze the effectiveness of 11 active learning algorithms on classifying subsite and histology from cancer pathology reports using a Convolutional Neural Network as the text classification model. Results We compare the performance of each active learning strategy using two differently sized datasets and two different classification tasks. Our results show that on all tasks and dataset sizes, all active learning strategies except diversity-sampling strategies outperformed random sampling, i.e., no active learning. On our large dataset (15K initial labelled samples, adding 15K additional labelled samples each iteration of active learning), there was no clear winner between the different active learning strategies. On our small dataset (1K initial labelled samples, adding 1K additional labelled samples each iteration of active learning), marginal and ratio uncertainty sampling performed better than all other active learning techniques. We found that compared to random sampling, active learning strongly helps performance on rare classes by focusing on underrepresented classes. Conclusions Active learning can save annotation cost by helping human annotators efficiently and intelligently select which samples to label. Our results show that a dataset constructed using effective active learning techniques requires less than half the amount of labelled data to achieve the same performance as a dataset constructed using random sampling.

59 BASIC BIOLOGICAL SCIENCES↗

Developing an Active Learning algorithm for learning Bayesian classifiers under the Multiple Instance Learning scenario

In the Multiple Instance Learning scenario, the training data consists of instances grouped into bags, and each bag is labelled with whether it is positive, i.e. contains at least one positive instance. First, Active Learning, in which additional labels can be iteratively requested, has the potential to allow more accurate classifiers to be learned with less labels. Active Learning has been applied to the Multiple Instance Learning under two settings: when bag labels of unlabelled bags can be requested, and when instance labels within bags known to be positive can be requested. Second, Bayesian Active learning methods have the potential to learn accurate classifiers with few labels, because they explicitly track the classifier uncertainty and can thus address its knowledge gaps. Yet, there does not exist any Bayesian Active Learning method for the Multiple Instance Learning Scenario. In this work, we develop the first such method. We develop a Bayesian classifier for the Multiple Instance Learning scenario, show how it can be efficiently used for Bayesian Active Learning, and perform experiments assessing its performance. While its performance exceeds that when no Active Learning is used, it is sometimes better, sometimes worse than the naive baseline of uncertainty sampling, depending on the situation. This suggests future work: building more customizable Bayesian Active Learning methods for the Multiple Instance Scenario, customizable to whether bag or instance label accuracy is targeted, and the labeling budget.

97 MATHEMATICS AND COMPUTING↗

MindSynchro

This report presents the developments and results of MindSynchro project as part of DOE OE FOA 1861. DOE and Pacific Northwest National Laboratory (PNNL) have made available to FOA awardees datasets containing years of real historical data recorded from various phasor measurement units (PMUs) which are installed in three large US interconnections: Texas (IC A), Western (IC B), and Eastern (IC C). The main goal of the project, which was successfully achieved, was to develop methods for detection and identification of events which are relevant for power grid operation. Tasks performed for achieving the project goals included data exploration and pre-processing, the development and application of physics-based features, data analysis and labeling based on unsupervised learning approaches, training and testing of DSSL models for classification of events which are relevant for power grid operation, and deployment of solutions to cloud environments. The methods developed in the project can potentially provide relevant benefits to power grid asset owners/operators in general in terms of situational awareness. Two main types of outcomes can be provided by these tools: Identification of specific relevant power grid event types: Semi-supervised ML methods developed in the project can adequately employ not only the relatively scarce labeled data but also the large amount of available unlabeled data to train models for detection of specific event types. Such methods enable the application of trained models for the detection of events in a population of PMUs much larger than that associated to the labeled events. Support in data labeling / label validation: Labels are critical for training of models for identification of specific types of events. However, labeling large amounts of data is a manual and tedious process. This means that such process is error prone and is not scalable. Methods developed in the project, based on ensembles of clustering models, have been successfully employed for turning manual labeling into a scalable process. Accurate identification of specific relevant events can provide the operators with immediate situational awareness that could otherwise require hours or days of analysis from domain experts. We envision that such methods could be initially employed in support of post-mortem analysis of events and, as confidence is gained, they could be employed for online/real-time support, providing, among other benefits, insights for avoiding major events which could happen due to a combination of smaller ones. On the longer term, related methods could potentially be employed to improve protection and control.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Mass spectrometry identifies multiple organophosphorylated sites on tubulin

Acute toxicity of organophosphorus poisons (OP) is explained by inhibition of acetylcholinesterase in nerve synapses. Low-dose effects are hypothesized to result from modification of other proteins, whose identity is not yet established. The goal of the present work was to obtain information that would make it possible to identify tubulin as a target of OP exposure. Tubulin was selected for study because live mice injected with a nontoxic dose of a biotinylated organophosphorus agent appeared to have OP-labeled tubulin in brain as determined by binding to avidin beads and mass spectrometry. The experiments with live mice were not conclusive because binding to avidin beads could be nonspecific. To be convincing, it is necessary to find and characterize the OP-labeled tubulin peptide. The search for OP-labeled tubulin peptides was begun by identifying residues capable of making a covalent bond with OP. Pure bovine tubulin (0.012 mM) was treated with 0.01-0.5 mM chlorpyrifos oxon for 24 h at 37 {sup o}C in pH 8.3 buffer. The identity of labeled amino acids and percent labeling was determined by mass spectrometry. Chlorpyrifos oxon bound covalently to tyrosines 83, 103, 108, 161, 224, 262, 272, 357, and 399 in bovine alpha tubulin, and to tyrosines 50, 51, 59, 106, 159, 281, 310, and 340 in bovine beta tubulin. The most reactive were tyrosine 83 in alpha and tyrosine 281 in beta tubulin. In the presence of 1 mM GTP, percent labeling increased 2-fold. Based on the crystal structure of the tubulin heterodimer (PDB 1jff) tyrosines 83 and 281 are well exposed to solvent. In conclusion seventeen tyrosines in tubulin have the potential to covalently bind chlorpyrifos oxon. These results will be useful when searching for OP-labeled tubulin in live animals.

60 APPLIED LIFE SCIENCES↗

Automated Grain Boundary (GB) Segmentation and Microstructural Analysis in 347H Stainless Steel Using Deep Learning and Multimodal Microscopy

Austenitic 347H stainless steel offers superior mechanical properties and corrosion resistance required for extreme operating conditions such as high temperature. The change in microstructure due to composition and process variations is expected to impact material properties. Identifying microstructural features such as grain boundaries thus becomes an important task in the process-microstructure-properties loop. Applying convolutional neural network (CNN)-based deep learning models is a powerful technique to detect features from material micrographs in an automated manner. In contrast to microstructural classification, supervised CNN models for segmentation tasks require pixel-wise annotation labels. However, manual labeling of the images for the segmentation task poses a major bottleneck for generating training data and labels in a reliable and reproducible way within a reasonable timeframe. Microstructural characterization especially needs to be expedited for faster material discovery by changing alloy compositions. Here, in this study, we attempt to overcome such limitations by utilizing multimodal microscopy to generate labels directly instead of manual labeling. We combine scanning electron microscopy images of 347H stainless steel as training data and electron backscatter diffraction micrographs as pixel-wise labels for grain boundary detection as a semantic segmentation task. The viability of our method is evaluated by considering a set of deep CNN architectures. We demonstrate that despite producing instrumentation drift during data collection between two modes of microscopy, this method performs comparably to similar segmentation tasks that used manual labeling. Additionally, we find that naïve pixel-wise segmentation results in small gaps and missing boundaries in the predicted grain boundary map. By incorporating topological information during model training, the connectivity of the grain boundary network and segmentation performance is improved. Finally, our approach is validated by accurate computation on downstream tasks of predicting the underlying grain morphology distributions which are the ultimate quantities of interest for microstructural characterization.

36 MATERIALS SCIENCE↗

Single-molecule imaging at high fluorophore concentrations by local activation of dye

Single-molecule fluorescence microscopy is a powerful approach to observe biomolecular interactions with high spatial and temporal resolution. Detecting fluorescent signals from individual, labeled proteins above high levels of background fluorescence remains challenging, however. For this reason, the concentrations of labeled proteins in in vitro assays are often kept low compared to their in vivo concentrations. Here, we present a new fluorescence imaging technique by which single fluorescent molecules can be observed in real time at high, physiologically relevant concentrations. The technique requires a protein and its macromolecular substrate to be labeled each with a different fluorophore. Then, making use of short-distance energy-transfer mechanisms, the fluorescence from only those proteins bound to their substrate are selectively activated. This approach is demonstrated by labeling a DNA substrate with an intercalating stain, exciting the stain, and using energy transfer from the stain to activate the fluorescence of only those labeled DNA-binding proteins bound to the DNA. Such an experimental design allowed us to observe the sequence-independent interaction of Cy5-labeled interferon-inducible protein 16 (IFI16) with DNA and the sliding via one-dimensional diffusion of Cy5-labeled adenovirus protease (pVIc-AVP) on DNA in the presence of a background of hundreds of nM Cy5 fluorophore.

59 BASIC BIOLOGICAL SCIENCES↗

Soil organic matter is principally root derived in an Ultisol under oak forest

To a large degree, the sources and stability of soil organic carbon remain poorly constrained. A clear understanding of links among the components of the soil C cycle is hampered by the complexity of the system as well as challenges associated with partitioning bulk soil C into meaningful fractions. A large accidental 14 CO 2 release at the Oak Ridge Reservation in Tennessee, USA provided a strong label pulse into adjacent, well-studied oak forests, resulting in highly elevated Δ 14 C values in leaf litter (~1000‰) and roots (~260–450‰). A four-year manipulative study was conducted to determine the relative contribution of litter versus roots to the bulk mineral soil C pool, as well as to free light, occluded light and heavy fractions. The heavy fraction was further split into fractions with densities of 1.7–2.4 g cm –3 and >2.4 g cm –3 to test the homogeneity of the mineral-associated fraction of C. Substantial concentrations of label were detected in all soil fractions within a year of the 14 CO 2 release, indicating rapid incorporation of newly fixed photosynthates in all fractions of soil organic C, regardless of differences in stability inferred by previous work. This rapid incorporation of label occurred only in treatments where roots were labeled, indicating that roots are the major source of inputs to mineral soil C stocks at these sites. Separation of the heavy fraction into subfractions of intermediate (1.7–2.4 g cm –3 ) and high (>2.4 g cm –3 ) density indicated that both subfractions incorporated label at similar rates, despite significant differences in degree of microbial processing. In general, the rate of label incorporation suggested a much faster turnover for all fractions than indicated by natural radiocarbon abundance values. This study suggests that within each soil fraction there are portions of slow-cycling and fast-cycling materials, and the determination of an average turnover time or mean age is dependent on experimental approach. The rapid incorporation of label into all fractions within a year regardless of inferred stability implies a high degree of heterogeneity in all fractions regardless of how finely the soils are partitioned. Further refinement of the nature and drivers of this heterogeneity could yield important insights into the soil C cycle.

58 GEOSCIENCES↗

Assessing fungal contributions to cellulose degradation in soil by using high-throughput stable isotope probing

Soils represent one of the largest and most active pools of C in the biosphere, and soil respiration represents a major component of global C flux. Fungi are essential to soil carbon cycling due to their propensity for decomposing organic polymers such as cellulose. We performed high throughput sequencing enabled stable isotope probing (HTS-SIP) with 13 C-cellulose to characterize the dynamics of fungi and bacteria during cellulose degradation in an agricultural soil. Here, a total of 1900 fungal taxa were observed and 190 of these assimilated 13 C-cellulose during a 30-day incubation. A majority of 13 C-labeled fungi belonged to Ascomycota, Basidiomycota, and Mucoromycota. However, most 13 C-labeled fungi could not be annotated at the species (71%, n = 134), or genus (49%, n = 93) level. Mucoromycota were 13 C-labeled early, and by day 3 the most abundant 13 C-labeled organism belonged to Mortierella. In contrast, 13 C-labeled Ascomycota increased in diversity through day 14 and their relative abundance comprised more than 40% of fungal ITS sequences by day 30. These results show that: i) the majority of fungal taxa that assimilated 13 C from 13 C-cellulose are uncultivated and poorly characterized, ii) the beta-diversity of 13 C-labeled fungi changed significantly over time during cellulose degradation, iii) a relatively small number of the 13 C-labeled taxa dominated the community response to cellulose, and iv) fungi incorporated cellulose into DNA more rapidly and in greater numbers than did bacteria. Furthermore, these results show that fungi in a tilled agricultural field respond rapidly to new cellulose inputs, exhibiting complex temporal dynamics that likely drive carbon flow into diverse taxa within the soil community.

59 BASIC BIOLOGICAL SCIENCES↗

The ionic charge of Copper-64 complexes conjugated to an engineered antibody effects biodistribution

The development of biomolecules as imaging probes requires radiolabeling methods that do not significantly influence their biodistribution. Sarcophagine (Sar) chelators form extremely stable complexes with copper, and are therefore a promising option for labeling proteins with ⁶⁴Cu. However, initial studies using the first-generation sarcophagine bifunctional chelator SarAr to label the engineered antibody fragment ch14.18-ΔC H 2 (MW 120 kDa) with ⁶⁴Cu showed high tracer retention in the kidneys,(>38% injected dose per gram (ID/g) 48 h post-injection), presumably because the high local positive charge on the Cu II -SarAr moiety resulted in increased binding of the labeled protein to the negatively charged basal cells of the glomerulus. To test this hypothesis, ch14.18-ΔC H 2 was conjugated with a series of Sar derivatives of decreasing positive charge and three commonly used macrocyclic polyaza polycarboxylate (PAC) BFCs. The immunoconjugates were labeled with ⁶⁴Cu and injected into mice, and PET/CT images were obtained at 24 and 48 h post injection (p.i.). At 48 h p.i., ex vivo biodistribution was carried out. In addition, to demonstrate the potential of metastasis detection using ⁶⁴Cu-labeled ch14.18-ΔC H 2, a preclinical imaging study of intrahepatic neuroblastoma tumors was performed carried out. Reducing the positive charge on the Sar chelators decreased kidney uptake of Cu-labeled ch14.18-ΔC H 2 by more than 6-fold, from >45 ID/g to <6% ID/g, while the uptake in most other tissues, including liver, was relatively unchanged. However, despite this dramatic decrease, the renal uptake of the PAC BFCs was generally lower than that of the Sar derivatives, as was the liver uptake. Uptake of ⁶⁴Cu-labeled ch14.18-ΔC H 2 in neuroblastoma hepatic metastases was detected using PET.

43 PARTICLE ACCELERATORS↗

Source of 12 C in Calvin–Benson cycle intermediates and isoprene emitted from plant leaves fed with 13 CO 2

Feeding 14 CO 2 was crucial to uncovering the path of carbon in photosynthesis. Feeding 13 CO 2 to photosynthesizing leaves emitting isoprene has been used to develop hypotheses about the sources of carbon for the methylerythritol 4-phosphate pathway, which makes the precursors for terpene synthesis in chloroplasts and bacteria. Both photosynthesis and isoprene studies found that products label very quickly (<10 min) up to 80–90% but the last 10–20% of labeling requires hours indicating a source of 12 C during photosynthesis and isoprene emission. Furthermore, studies with isoprene showed that the proportion of slow label could vary significantly. This was interpreted as a variable contribution of carbon from sources other than the Calvin–Benson cycle (CBC) feeding the methylerythritol 4-phosphate pathway. Here, we measured the degree of label in isoprene and photosynthetic metabolites 20 min after beginning to feed 13 CO 2 . Isoprene labeling was the same as labeling of photosynthesis intermediates. High temperature reduced the label in isoprene and photosynthesis intermediates by the same amount indicating no role for alternative carbon sources for isoprene. A model assuming glucose, fructose, and/or sucrose reenters the CBC as ribulose 5-phosphate through a cytosolic shunt involving glucose 6-phosphate dehydrogenase was consistent with the observations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Automated annotation of scientific texts for ML-based keyphrase extraction and validation

Advanced omics technologies and facilities generate a wealth of valuable data daily; however, the data often lack the essential metadata required for researchers to find, curate, and search them effectively. The lack of metadata poses a significant challenge in the utilization of these data sets. Machine learning (ML)–based metadata extraction techniques have emerged as a potentially viable approach to automatically annotating scientific data sets with the metadata necessary for enabling effective search. Text labeling, usually performed manually, plays a crucial role in validating machine-extracted metadata. However, manual labeling is time-consuming and not always feasible; thus, there is a need to develop automated text labeling techniques in order to accelerate the process of scientific innovation. This need is particularly urgent in fields such as environmental genomics and microbiome science, which have historically received less attention in terms of metadata curation and creation of gold-standard text mining data sets. In this paper, we present two novel automated text labeling approaches for the validation of ML-generated metadata for unlabeled texts, with specific applications in environmental genomics. Our techniques show the potential of two new ways to leverage existing information that is only available for select documents within a corpus to validate ML models, which can then be used to describe the remaining documents in the corpus. The first technique exploits relationships between different types of data sources related to the same research study, such as publications and proposals. The second technique takes advantage of domain-specific controlled vocabularies or ontologies. In this paper, we detail applying these approaches in the context of environmental genomics research for ML-generated metadata validation. Our results show that the proposed label assignment approaches can generate both generic and highly specific text labels for the unlabeled texts, with up to 44% of the labels matching with those suggested by a ML keyword extraction algorithm.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

A Peptide-Based Method for 13C Metabolic Flux Analysis in Microbial Communities

The study of intracellular metabolic fluxes and inter-species metabolite exchange for microbial communities is of crucial importance to understand and predict their behaviour. The most authoritative method of measuring intracellular fluxes, 13C Metabolic Flux Analysis (13C MFA), uses the labeling pattern obtained from metabolites (typically amino acids) during 13C labeling experiments to derive intracellular fluxes. However, these metabolite labeling patterns cannot easily be obtained for each of the members of the community. Here we propose a new type of 13C MFA that infers fluxes based on peptide labeling, instead of amino acid labeling. The advantage of this method resides in the fact that the peptide sequence can be used to identify the microbial species it originates from and, simultaneously, the peptide labeling can be used to infer intracellular metabolic fluxes. Peptide identity and labeling patterns can be obtained in a high-throughput manner from modern proteomics techniques. We show that, using this method, it is theoretically possible to recover intracellular metabolic fluxes in the same way as through the standard amino acid based 13C MFA, and quantify the amount of information lost as a consequence of using peptides instead of amino acids. We show that by using a relatively small number of peptides we can counter this information loss. We computationally tested this method with a well-characterized simple microbial community consisting of two species.

59 BASIC BIOLOGICAL SCIENCES↗

Advanced Semi-Supervised Learning with Uncertainty Estimation for Phase Identification in Distribution Systems

The integration of advanced metering infrastructure (AMI) into power distribution networks generates valuable data for tasks such as phase identification; however, the limited and unreliable availability of labeled data in the form of customer phase connectivity presents challenges. To address this issue, we propose a semi-supervised learning (SSL) framework that effectively leverages labeled and unlabeled data. Our approach incorporates self-training, label spreading, and Bayesian neural networks (BNNs) to enhance phase identification with AMI data. Our method uses an ensemble of multilayer perceptron classifiers in a self-training setup, iteratively adding high-confidence pseudo-labels to improve robustness. We also apply label spread to propagate labels based on data similarity, which enhances generalization across diverse distributions. In addition, we employ a BNNs with uncertainty estimation, boosting confidence in predictions and reducing phase identification errors. In our case study, we achieved approximately 98% +/- 0.08 accuracy with uncertainty using minimal and unreliable labeled data from a real U.S. utility, Duquesne Light Company. Our SSL approach, combined with uncertainty estimation, provides an efficient solution for phase identification in AMI data, ultimately improving the reliability of smart grid applications.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Advanced Semi-Supervised Learning With Uncertainty Estimation for Phase Identification in Distribution Systems

The integration of advanced metering infrastructure (AMI) into power distribution networks generates valuable data for tasks such as phase identification; however, the limited and unreliable availability of labeled data in the form of customer phase connectivity presents challenges. To address this issue, we propose a semi-supervised learning (SSL) framework that effectively leverages labeled and unlabeled data. Our approach incorporates self-training, label spreading, and Bayesian neural networks (BNNs) to enhance phase identification with AMI data. Our method uses an ensemble of multilayer perceptron classifiers in a self-training setup, iteratively adding high-confidence pseudo-labels to improve robustness. We also apply label spread to propagate labels based on data similarity, which enhances generalization across diverse distributions. In addition, we employ a BNNs with uncertainty estimation, boosting confidence in predictions and reducing phase identification errors. In our case study, we achieved approximately 98% +/- 0.08 accuracy with uncertainty using minimal and unreliable labeled data from a real U.S. utility, Duquesne Light Company. Our SSL approach, combined with uncertainty estimation, provides an efficient solution for phase identification in AMI data, ultimately improving the reliability of smart grid applications.

24 POWER TRANSMISSION AND DISTRIBUTION↗