SEARCH · Engineering Papers

Results for “community detection”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13

All sources NASA NTRS DOE OSTI

DOE OSTI2023

Quantum Devices for Neutrino and Rare Particle Detection (Final Report)

This report summarizes the technical work completed under the DOE-QuanISED program. This grant allowed the MIT group --and associated groups on the project, including UMass Amherst and Argonne National Laboratory -- to explore the use of superconducting bolometers for the purpose of neutrino and low mass dark matter detection. These low threshold bolometric detection techniques have wide applicability in nuclear and particle physics, including searching for low mass weakly-interacting dark matter; detecting neutrinos via coherent neutrino scattering; searching for axions and nuclear non-proliferation. In addition, we explored how using techniques readily used by the quantum computing community–such as multiplexed microwave resonators and quantum amplifiers—could be used as a low noise readout scheme for an array of superconducting detectors. We briefly summarized the accomplishments under the project, and what future research directions have spawned from this research.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2024

Laser-induced graphene gas sensors for environmental monitoring

Artemesia tridentatais a foundational plant taxon in western North America and an important medicinal plant threatened by climate change. Low-cost fabrication of sensors is critical for developing large-area sensor networks for understanding and monitoring a range of environmental conditions. However, the availability of materials and manufacturing processes is still in the early stages, limiting the capacity to develop cost-effective sensors at a large scale. In this study, we demonstrate the fabrication of low-cost flexible sensors using laser-induced graphene (LIG); a graphitic material synthesized using a 450-nm wavelength bench top laser patterned onto polyimide substrates. We demonstrate the effect of the intensity and focus of the incident beam on the morphology and electrical properties of the synthesized material. Raman analyses of the synthesized LIG show a defect-rich graphene with a crystallite size in the tens of nanometers. This shows that the high level of disorder within the LIG structure, along with the porous nature of the material provide a good surface for gas adsorption. The initial characterization of the material has shown an analyte response represented by a change in resistance of up to 5% in the presence of volatile organic compounds (VOCs) that are emitted and detected byArtemisiaspecies. Bend testing up to 100 cycles provides evidence that these sensors will remain resilient when deployed across the landscapes to assess VOC signaling in plant communities. The versatile low-cost laser writing technique highlights the promise of low-cost and scalable fabrication of LIG sensors for gas sensor monitoring.

Chemistry↗

DOE OSTI2021

Dissecting the dominant hot spring microbial populations based on community-wide sampling at single-cell genomic resolution

With advances in DNA sequencing and miniaturized molecular biology workflows, rapid and affordable sequencing of single-cell genomes has become a reality. Compared to 16S rRNA gene surveys and shotgun metagenomics, large-scale application of single-cell genomics to whole microbial communities provides an integrated snapshot of community composition and function, directly links mobile elements to their hosts, and enables analysis of population heterogeneity of the dominant community members. To that end, we sequenced nearly 500 single-cell genomes from a low diversity hot spring sediment sample from Dewar Creek, British Columbia, and compared this approach to 16S rRNA gene amplicon and shotgun metagenomics applied to the same sample. We found that the broad taxonomic profiles were similar across the three sequencing approaches, though several lineages were missing from the 16S rRNA gene amplicon dataset, likely the result of primer mismatches. At the functional level, we detected a large array of mobile genetic elements present in the single-cell genomes but absent from the corresponding same species metagenome-assembled genomes. Moreover, we performed a single-cell population genomic analysis of the three most abundant community members, revealing differences in population structure based on mutation and recombination profiles. While the average pairwise nucleotide identities were similar across the dominant species-level lineages, we observed differences in the extent of recombination between these dominant populations. Most intriguingly, the creek's Hydrogenobacter sp. population appeared to be so recombinogenic that it more closely resembled a sexual species than a clonally evolving microbe. Together, this work demonstrates that a randomized single-cell approach can be useful for the exploration of previously uncultivated microbes from community composition to population structure.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

AI-accelerated protein-ligand docking for SARS-CoV-2 is 100-fold faster with no significant change in detection

Protein-ligand docking is a computational method for identifying drug leads. The method is capable of narrowing a vast library of compounds down to a tractable size for downstream simulation or experimental testing and is widely used in drug discovery. While there has been progress in accelerating scoring of compounds with artificial intelligence, few works have bridged these successes back to the virtual screening community in terms of utility and forward-looking development. We demonstrate the power of high-speed ML models by scoring 1 billion molecules in under a day (50 k predictions per GPU seconds). We showcase a workflow for docking utilizing surrogate AI-based models as a pre-filter to a standard docking workflow. Our workflow is ten times faster at screening a library of compounds than the standard technique, with an error rate less than 0.01% of detecting the underlying best scoring 0.1% of compounds. Our analysis of the speedup explains that another order of magnitude speedup must come from model accuracy rather than computing speed. In order to drive another order of magnitude of acceleration, we share a benchmark dataset consisting of 200 million 3D complex structures and 2D structure scores across a consistent set of 13 million “in-stock” molecules over 15 receptors, or binding sites, across the SARS-CoV-2 proteome. We believe this is strong evidence for the community to begin focusing on improving the accuracy of surrogate models to improve the ability to screen massive compound libraries 100 × or even 1000 × faster than current techniques and reduce missing top hits. The technique outlined aims to be a fast drop-in replacement for docking for screening billion-scale molecular libraries.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2022

Capturing the microbial volatilome: an oft overlooked 'ome'

Among the diverse metabolites produced by microbial communities, some are volatile. Volatile organic compounds (VOCs) are vigorously cycled by microbes as metabolic substrates and products and as signaling molecules. Yet, current microbial metabolomic studies predominantly focus on nonvolatile metabolites and overlook VOCs, which therefore represent a missing component of the metabolome. Advances in VOC detection now allow simultaneous observation of the numerous VOCs constituting the ‘volatilome’ of microbial systems. Here, we present a roadmap for integrating and advancing VOC and other ‘omics approaches and highlight the potential for realtime VOC measurements to help overcome limitations in discrete ‘omics sampling. Including volatile metabolites in metabolomics, both conceptually and in practice, will build a more comprehensive understanding of microbial processes across ecological communities.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2022

Assessing the Impact of Lightning NOx Emissions in CMAQ Using Lightning Flash Data from WWLLN over the Contiguous United States

Comparison of lightning flash data from the National Lightning Detection Network (NLDN) and from the World Wide Lightning Location Network (WWLLN) over the contiguous United States (CONUS) for the 2016–2018 period reveals temporally and spatially varying flash rates that would influence lightning NO x (LNO x ) production due to known detection efficiency differences especially during summer months over land (versus over ocean). However, the lightning flash density differences between the two networks show persistent seasonal patterns over geographical regions (e.g., land versus ocean). Since the NLDN data are considered to have higher accuracy (lightning detection with >95% efficiency), we developed scaling factors for the WWLLN flash data based on the ratios of WWLLN to NLDN flash data over time (months of year) and space. In this study, sensitivity simulations using the Community Multiscale Air Quality (CMAQ) model are performed utilizing the original data sets (both NLDN and WWLLN) and the scaled WWLLN flash data for LNO x production over the CONUS. The model performance of using the different lightning flash datasets for ambient O 3 and NO x mixing ratios that are directly impacted by LNO x emissions and the wet and dry deposition of oxidized nitrogen species that are indirectly impacted by LNO x emissions is assessed based on comparisons with ground-based observations, vertical profile measurements, and satellite products. During summer months, the original WWLLN data produced less LNO x emissions (due to its lower lightning detection efficiency) compared to the NLDN data, which resulted in less improvement in model performance than the simulation using NLDN data as compared to the simulation without any LNO x emissions. However, the scaled WWLLN data produced LNO x estimates and model performance comparable with the NLDN data, suggesting that scaled WWLLN may be used as a substitute for the NLDN data to provide LNO x estimates in air quality models when the NLDN data are not available (e.g., due to prohibitive cost or lack of spatial coverage).

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

A red algal polysaccharide influences the multicellular development of the choanoflagellate Salpingoeca rosetta

The choanoflagellate Salpingoeca rosetta exemplifies the capacity of marine microeukaryotes to integrate environmental cues into their life histories. Some of the cues that S. rosetta detects are lipids from their bacterial prey that induce the development of multicellular colonies called rosettes. In the aquatic environments where S. rosetta would encounter bacterial cues, microbial communities gain refuge and food from algae. For example, microbes in coastal environments benefit from macroalgae that annually release ∼10 14 g of carbon, primarily in the form of polysaccharides that comprise ∼80% of dissolved organic carbon in surface waters. Despite the ecological impact of algae and historical descriptions of choanoflagellates attached to algae, the influence of algae on choanoflagellate life history transitions has remained unexplored. Here, we report that porphyran, a polysaccharide produced by the red macroalga Porphyra umbilicalis, induces multicellular development in S. rosetta. We first noticed this response when S. rosetta formed multicellular rosette colonies while growing in media prepared from P. umbilicalis, even though rosette-inducing bacteria were absent. By biochemically purifying extracts of P. umbilicalis, we identified porphyran as the rosette-inducing factor. This response provides a biochemical insight that indicates algal polysaccharides may serve as cues to mediate associations between choanoflagellates and algae. Moreover, this observation complements evidence from environmental and geochemical studies that show the impact that algae have exerted on the ecology and evolution of eukaryotes, including a rise in algal productivity during the origin of animals, the closest living relatives of choanoflagellates.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

Amazonian fog harbors viable microbes

Fog formation over tropical forests remains poorly characterized, despite its potential role in bioaerosol dispersion and ecosystem processes. Here, we analyzed fog samples collected at the Amazon Tall Tower Observatory using flow cytometry and culture-based techniques to characterize viable microbial communities. Microbial cell concentrations varied over an order of magnitude across 13 fog events, reaching up to 8 × 104 cells per ml of fog water. Flow cytometry consistently detected metabolically active cells, while culturing and mass spectrometry-based identification yielded eight viable bacterial species and seven fungal taxa. The bacteria Serratia marcescens, Ralstonia pickettii and Sphingomonas paucimobilis exhibited seasonal variations in prevalence. The fungal species identified were primarily mesophilic saprophytes and endophytes, commonly associated with soil and plant surfaces. Our findings indicate that fog harbors viable microbes, including Serratia marcescens and Ralstonia pickettii, which may imply a relevance of fog for microbial dispersal, colonization and nutrient cycling in the Amazon rainforest.

Godoi, Ricardo H. (ORCID:0000000247744870)↗

DOE OSTI2022

Practical galaxy morphology tools from deep supervised representation learning

Astronomers have typically set out to solve supervised machine learning problems by creating their own representations from scratch. We show that deep learning models trained to answer every Galaxy Zoo DECaLS question learn meaningful semantic representations of galaxies that are useful for new tasks on which the models were never trained. We exploit these representations to outperform several recent approaches at practical tasks crucial for investigating large galaxy samples. The first task is identifying galaxies of similar morphology to a query galaxy. Given a single galaxy assigned a free text tag by humans (e.g. ‘#diffuse’), we can find galaxies matching that tag for most tags. The second task is identifying the most interesting anomalies to a particular researcher. Our approach is 100 per cent accurate at identifying the most interesting 100 anomalies (as judged by Galaxy Zoo 2 volunteers). The third task is adapting a model to solve a new task using only a small number of newly labelled galaxies. Models fine-tuned from our representation are better able to identify ring galaxies than models fine-tuned from terrestrial images (ImageNet) or trained from scratch. We solve each task with very few new labels; either one (for the similarity search) or several hundred (for anomaly detection or fine-tuning). This challenges the longstanding view that deep supervised methods require new large labelled data sets for practical use in astronomy. To help the community benefit from our pretrained models, we release our fine-tuning code zoobot. Zoobot is accessible to researchers with no prior experience in deep learning.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2021

LAMP Diagnostics at the Point-of-Care: Emerging Trends and Perspectives for the Developer Community

Over the past decade, loop-mediated isothermal amplification (LAMP) technology has played an important role in molecular diagnostics. Amongst numerous nucleic acid amplification assays, LAMP stands out in terms of sample-to-answer time, sensitivity, specificity, cost, robustness, and accessibility, making it ideal for field-deployable diagnostics in resource-limited regions. In this review, we outline the front-end LAMP design practices for point-of-care (POC) applications, including sample handling and various signal readout methodologies. Next, we explore existing LAMP technologies that have been validated with clinical samples in the field. We summarize recent work that utilizes reverse transcription (RT) LAMP to rapidly detect SARS-CoV-2 as an alternative to standard PCR protocols. Finally, we describe challenges in translating LAMP from the benchtop to the field and opportunities for future LAMP assay development and performance reporting. Despite the popularity of LAMP in the academic research community and a recent surge in interest in LAMP due to the COVID-19 pandemic, there are numerous areas for improvement in the fundamental understanding of LAMP, which are needed to elevate the field of LAMP assay development and characterization.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2023

Assessing residential activity in a home plumbing system simulator: monitoring the occurrence and relationship of major opportunistic pathogens and phagocytic amoebas

Opportunistic premise plumbing pathogens (OPPPs) have been detected in buildings’ plumbing systems causing waterborne disease outbreaks in the United States. In this study, we monitored the occurrence of OPPPs along with free-living amoeba (FLA) and investigated the effects of residential activities in a simulated home plumbing system (HPS). Water samples were collected from various locations in the HPS and analyzed for three major OPPPs: Legionella pneumophila, nontuberculous mycobacterial species (e.g., Mycobacterium avium, M. intracellulare, and M. abscessus), and Pseudomonas aeruginosa along with two groups of amoebas (Acanthamoeba and Vermamoeba vermiformis). A metagenomic approach was also used to further characterize the microbial communities. Results show that the microbial community is highly diverse with evidence of spatial and temporal structuring influenced by environmental conditions. L. pneumophila was the most prevalent pathogen (86% of samples), followed by M. intracellulare (66%) and P. aeruginosa (21%). Interestingly, M. avium and M. abscessus were not detected in any samples. The data revealed a relatively low prevalence of Acanthamoeba spp. (4%), while V. vermiformis was widely detected (81%) across all the sampling locations within the HPS. Locations with a high concentration of L. pneumophila and M. intracellulare coincided with the highest detection of V. vermiformis, suggesting the potential growth of both populations within FLA and additional protection in drinking water. After a period of stagnation lasting at least 2-weeks, the concentrations of OPPPs and amoeba immediately increased and then decreased gradually back to the baseline. Furthermore, monitoring the microbial population after drainage of the hot water tank and partial drainage of the entire HPS demonstrated no significant mitigation of the selected OPPPs. This study demonstrates that these organisms can adjust to their environment during such events and may survive in biofilms and/or grow within FLA, protecting them from stressors in the supplied water.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2021

Assessing Microbial Communities Related to Mercury Transformations in Contaminated Streambank Soils

In nature, the bioaccumulative potent neurotoxin methylmercury (MeHg) is produced from inorganic mercury (Hg) predominantly by anaerobic microorganisms. Hg-contaminated soils are a potential source of MeHg due to microbial activity. We examine streambank soils collected from the contaminated East Fork Poplar Creek (EFPC) in Tennessee, USA, where seasonal variations in MeHg levels have been observed throughout the year, suggesting active microbial Hg methylation. In this study, we characterized the microbial community in contaminated bank soil samples collected from two locations over a period of one year and compared the results to soil samples from an uncontaminated reference site with similar geochemistry (n = 12). Microbial community composition and diversity were assessed by 16S rRNA gene amplicon sequencing. Furthermore, to isolate potential methylators from soils, enrichment cultures were prepared using selective media. A set of three clade-specific primers targeting the gene hgcA were used to detect Hg methylators among the δ-Proteobacteria in EFPC bank soils across all seasons. Two families among the δ-Proteobacteria that have been previously associated with Hg methylation, Geobacteraceae and Syntrophobacteraceae, were found to be predominant with relative abundances of 0.13% and 4.0%, respectively. However, in soil enrichment cultures, Firmicutes were predominant among families associated with Hg methylation. Specifically, Clostridiaceae and Peptococcaceae and their genera Clostridium and Desulfosporosinus were among the ten most abundant genera with relative abundances of 2.6% and 1.7%, respectively. These results offer insights into the role of microbial communities on Hg transformation processes in contaminated bank soils in EFPC. Identifying the biogeochemical drivers of MeHg production is critical for future remediation efforts.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

PRIMED for the Future: Purposing Raw Intake for Machine Learning-Enabled Detection

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2023

Fine-scale evaluation of two standard 16S rRNA gene amplicon primer pairs for analysis of total prokaryotes and archaeal nitrifiers in differently managed soils

The advance of high-throughput molecular biology tools allows in-depth profiling of microbial communities in soils, which possess a high diversity of prokaryotic microorganisms. Amplicon-based sequencing of 16S rRNA genes is the most common approach to studying the richness and composition of soil prokaryotes. To reliably detect different taxonomic lineages of microorganisms in a single soil sample, an adequate pipeline including DNA isolation, primer selection, PCR amplification, library preparation, DNA sequencing, and bioinformatic post-processing is required. Besides DNA sequencing quality and depth, the selection of PCR primers and PCR amplification reactions arguably have the largest influence on the results. This study tested the performance and potential bias of two primer pairs, i.e., 515F (Parada)-806R (Apprill) and 515F (Parada)-926R (Quince) in the standard pipelines of 16S rRNA gene Illumina amplicon sequencing protocol developed by the Earth Microbiome Project (EMP), against shotgun metagenome-based 16S rRNA gene reads. The evaluation was conducted using five differently managed soils. We observed a higher richness of soil total prokaryotes by using reverse primer 806R compared to 926R, contradicting to in silico evaluation results. Both primer pairs revealed various degrees of taxon-specific bias compared to metagenome-derived 16S rRNA gene reads. Nonetheless, we found consistent patterns of microbial community variation associated with different land uses, irrespective of primers used. Total microbial communities, as well as ammonia oxidizing archaea (AOA), the predominant ammonia oxidizers in these soils, shifted along with increased soil pH due to agricultural management. In the unmanaged low pH plot abundance of AOA was dominated by the acid-tolerant NS-Gamma clade, whereas limed agricultural plots were dominated by neutral-alkaliphilic NS-Delta/NS-Alpha clades. This study stresses how primer selection influences community composition and highlights the importance of primer selection for comparative and integrative studies, and that conclusions must be drawn with caution if data from different sequencing pipelines are to be compared.

16S rRNA gene amplicon Illumina sequencing↗

DOE OSTI2023

Keeping LANL at the forefront of list mode capability for safeguards and MC&A [Slides]

Neutron based detection forms key capability to enable and support international and domestic nuclear safeguards and non-proliferation missions. Key stakeholders include International Atomic Energy Agency (IAEA), European Atomic Energy Community (EURATOM), regional safeguards inspectorates, domestic stakeholders. LANL has been historically a leader in development of Non-destructive Assay (NDA) instruments and associated data acquisition hardware and software.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗

DOE OSTI2023

BatAnalysis - A Comprehensive Python Pipeline for Swift BAT Survey Analysis

The Swift Burst Alert Telescope (BAT) is a coded-aperture gamma-ray instrument with a large field of view that primarily operates in survey mode when it is not triggering on transient events. The survey data consist of 80- channel detector plane histograms that accumulate photon counts over periods of at least 5 minutes. These histograms are processed on the ground and are used to produce the survey data set between 14 and 195 keV. Survey data comprise >90% of all BAT data by volume and allow for the tracking of long-term light curves and spectral properties of cataloged and uncataloged hard X-ray sources. Until now, the survey data set has not been used to its full potential due to the complexity associated with its analysis and the lack of easily usable pipelines. Here, we introduce the BatAnalysis Python package, a wrapper for HEASoftpy, which provides a modern, opensource pipeline to process and analyze BAT survey data. BatAnalysis allows members of the community to use BAT survey data in more advanced analyses of astrophysical sources, including pulsars, pulsar wind nebula, active galactic nuclei, and other known/unknown transient events that may be detected in the hard X-ray band. We outline the steps taken by the Python code and exemplify its usefulness and accuracy by analyzing survey data of the Crab Nebula, NGC 2992, and a previously uncataloged MAXI transient. The BatAnalysis package allows for ~18 yr of BAT survey data to be used in a systematic way to study a large variety of astrophysical sources.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2023

BatAnalysis - A Comprehensive Python Pipeline for Swift BAT Survey Analysis

The Swift Burst Alert Telescope (BAT) is a coded-aperture gamma-ray instrument with a large field of view that primarily operates in survey mode when it is not triggering on transient events. The survey data consist of 80-channel detector plane histograms that accumulate photon counts over periods of at least 5 minutes. These histograms are processed on the ground and are used to produce the survey data set between 14 and 195 keV. Survey data comprise >90% of all BAT data by volume and allow for the tracking of long-term light curves and spectral properties of cataloged and uncataloged hard X-ray sources. Until now, the survey data set has not been used to its full potential due to the complexity associated with its analysis and the lack of easily usable pipelines. Here, we introduce the BatAnalysis Python package, a wrapper for HEASoftpy, which provides a modern, open-source pipeline to process and analyze BAT survey data. BatAnalysis allows members of the community to use BAT survey data in more advanced analyses of astrophysical sources, including pulsars, pulsar wind nebula, active galactic nuclei, and other known/unknown transient events that may be detected in the hard X-ray band. We outline the steps taken by the Python code and exemplify its usefulness and accuracy by analyzing survey data of the Crab Nebula, NGC 2992, and a previously uncataloged MAXI transient. The BatAnalysis package allows for ~18 yr of BAT survey data to be used in a systematic way to study a large variety of astrophysical sources.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2023

PRIMED for the Future: Purposing Raw Intake for Machine Learning-Enabled Detection (Final Report)

The COVID-19 pandemic demonstrated how a novel, elusive, and diffuse biological threat can engender uncertainty and misinformation, and it underscored the need for flexible analytical modalities agnostic to the identity of biological material. Yet even before the pandemic, recognition of the limitations of the current, list-based approach, which focuses on known pathogens and biotoxins, and of the importance of agent-agnostic biodetection was growing within the biosecurity community. In a 2018 report on “Biodefense in the Age of Synthetic Biology,” for example, the National Academy of Sciences stated that “an overreliance on the Select Agent List is a systematic weakness affecting many aspects of the United States’ current biodefense mitigation capability”. More recently, a group of biodefense researchers proposed the identification and adoption of “bioagent-agnostic signatures (BASs)” as a way of detecting and characterizing not only existing agents but also novel ones, an approach they believe will “enable a more flexible and resilient biodefense posture”. Indeed, the future of biodetection requires us to begin developing novel analytics that can identify anomalies and/or characteristics that indicate a potential threat, whether known or unknown, without looking for a specific signature that has been identified previously. To assess potential threats more rapidly, it is critical to develop agnostic artificial intelligence (AI)/machine learning (ML) systems that can be employed for real-time assessment of the nature and source of a perturbation. Such systems should be multi scale and multi-dimensional, integrating sensor data from a range of biological, chemical, and physical application spaces. Emerging deep learning (DL) models demonstrate exceptional promise for identification of discriminatory features within multi-dimensional datasets. DL models have the capacity to recognize and encode highly complex patterns in a wide range of input data modalities, including images, text, and biological/chemical/physical spectra. As such, they can execute a wide range of assessments and determinations that have traditionally required a human operator. The promise of advances in DL is apparent in the realm of human health and medicine. DL models have been validated for evaluating a variety of clinical threats to human health in a range of contexts, including infection and cancer, and they demonstrated improved performance in predicting stroke relative to human neurologists in some categories of data. Continuously evolving advances in AI/ML are expected to support more efficient evaluation of raw sequence, spectroscopy, and spectrometry data. For instance, recent advances and deployment of large language models (LLM) such as Generative Pre training Transformer (GPT) and Bidirectional Encoder Representations from Transformers (BERT) have already motivated application of these models for biological function prediction. As frameworks such as LLMs become larger and more complex in their representations, their capacity to serve as pre-trained models that can be fine-tuned for biological/biodetection purposes will similarly be amplified. While existing and emerging AI/ML have found broad applicability and use cases in the clinical sciences, development for environmental evaluation and biodetection has been limited. Functionalizing such capabilities for this purpose requires an understanding of the existing technical landscape and how the respective tools and algorithms are currently being employed. This landscape awareness then allows an assessment of the current practical capabilities of existing models and the anticipated requirements and development efforts that will be needed to adapt available algorithms for biodetection applications relevant to DHS. Leveraging expertise in biodetection, ML, and operational biodetection, the effort described in this report is comprised of a systematic landscape assessment (Subtask 2.1), comparative evaluation (Subtask 2.2), and formulation of a value proposition (Subtask 2.3) for the prospect of ML-enabled, agnostic biodetection from raw, or minimally-processed, datasets.

59 BASIC BIOLOGICAL SCIENCES↗