Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “combinatorial data”

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 217 records · Page 12

Experimental determinations of carbon and hydrogen isotope fractionations and methane clumped isotope compositions associated with ethane pyrolysis from 550 to 600°C

Methane clumped isotope compositions signify the relative natural abundances of rare, doubly substituted isotopic species of methane ( 13 CH 3 D and 12 CH 2 D 2 ) and have emerged as a new isotopic tool to trace the sources, sinks, and lifecycles of methane in the environment. Such measurements can identify equilibration (or reequilibration) temperatures if found to be in isotopic equilibrium or non-equilibrium processes (e.g., kinetically controlled reactions or mixing) if not in isotopic equilibrium. Naturally occurring thermogenic methane—formed by the thermally activated breakdown of larger organic molecules—has been found to have clumped isotope compositions consistent with equilibrium at reasonable gas formation temperatures in some settings and non-equilibrium processes occurring during either formation, migration, storage, or extraction in others. To explore the potential controls on the isotopic composition of thermogenic methane, we conducted isothermal time-series ethane pyrolysis experiments at 550 and 600 °C to measure methane and ethane 13 C/ 12 C and D/H fractionations and methane clumped isotope compositions (resolved 13 CH 3 D and 12 CH 2 D 2 ). We explore the effects of modifying the initial clumped isotope composition of ethane and the addition of water vapor to pyrolysis experiments. We observe that ethane and methane 13 C/ 12 C are controlled by kinetic isotope effects and Rayleigh distillation processes. In contrast, ethane and methane D/H and methane clumped isotope compositions appear to be controlled by a combination of these processes and hydrogen isotope exchange. The hydrogen isotope exchange processes lead to isotopic equilibrium as reaction completion is approached for both D/H (ethane/methane) and methane clumped isotope compositions. Here, we develop a chemical model based on a mass balance approach that accounts for inheritance vs. hydrogen-abstraction formation pathways for singly and doubly substituted isotopologues of ethane and methane that is compared to the experimental data. The model allows the determination of carbon and hydrogen kinetic isotope effects associated with ethane cracking and hydrogen abstraction reactions that, where applicable, we compare to prior theoretical constraints. From the comparison of the model to the experimental data, we infer that the kinetically controlled ethane and methane bulk isotope compositions and methane clumped isotope compositions are controlled by kinetic isotope effects (both primary and secondary) associated with both C–C bond and C–H bond cleavage reactions. Specifically, the methane clumped isotope compositions likely result from a combination of clumped isotope effects associated with ethane breakdown and/or assembly of methane isotopologues (expressed in terms of γ-factor parameters ≠ 1) and combinatorial effects that arise probabilistically. We discuss our experimental results in the context of recent pyrolysis experiments and observations of naturally occurring thermogenic methane. We consider a proposal consistent with observations from nature that the hydrogen isotope exchange reactions that promote equilibration of methane isotopic molecules at or near formation temperature may be facilitated by free radicals generated by pyrolysis reactions. In this framework, isotope exchange effectively ceases when pyrolysis effectively ceases locking in compositions that can be consistent with peak formation temperatures.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A systems level predictive model for global gene regulation of methanogenesis in a hydrogenotrophic methanogen

Methanogens catalyze the critical methane-producing step (called methanogenesis) in the anaerobic decomposition of organic matter. Here, we present the first predictive model of global gene regulation of methanogenesis in a hydrogenotrophic methanogen, Methanococcus maripaludis. We generated a comprehensive list of genes (protein-coding and noncoding) for M. maripaludis through integrated analysis of the transcriptome structure and a newly constructed Peptide Atlas. The environment and gene-regulatory influence network (EGRIN) model of the strain was constructed from a compendium of transcriptome data that was collected over 58 different steady-state and time-course experiments that were performed in chemostats or batch cultures under a spectrum of environmental perturbations that modulated methanogenesis. Analyses of the EGRIN model have revealed novel components of methanogenesis that included at least three additional protein-coding genes of previously unknown function as well as one noncoding RNA. We discovered that at least five regulatory mechanisms act in a combinatorial scheme to intercoordinate key steps of methanogenesis with different processes such as motility, ATP biosynthesis, and carbon assimilation. Through a combination of genetic and environmental perturbation experiments we have validated the EGRIN-predicted role of two novel transcription factors in the regulation of phosphate-dependent repression of formate dehydrogenase—a key enzyme in the methanogenesis pathway. The EGRIN model demonstrates regulatory affiliations within methanogenesis as well as between methanogenesis and other cellular functions.

Biochemistry & Molecular Biology↗

Automated exploitation of the big configuration space of large adsorbates on transition metals reveals chemistry feasibility

Mechanistic understanding of large molecule conversion and the discovery of suitable heterogeneous catalysts have been lagging due to the combinatorial inventory of intermediates and the inability of humans to enumerate all structures. Here, we introduce an automated framework to predict stable configurations on transition metal surfaces and demonstrate its validity for adsorbates with up to 6 carbon and oxygen atoms on 11 metals, enabling the exploration of ~10 8 potential configurations. It combines a graph enumeration platform, force field, multi-fidelity DFT calculations, and first-principles trained machine learning. Clusters in the data reveal groups of catalysts stabilizing different structures and expose selective catalysts for showcase transformations, such as the ethylene epoxidation on Ag and Cu and the lack of C-C scission chemistry on Au. Deviations from the commonly assumed atom valency rule of small adsorbates are also manifested. This library can be leveraged to identify catalysts for converting large molecules computationally.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Hydrogen Regulation and Global Responses to Electron, Carbon and Nitrogen Sources of Methanococcus Maripaludis

Methanogens catalyze the critical, methane-producing step (called methanogenesis) in the anaerobic decomposition of organic matter. This project has generated the first predictive model of global gene regulation of methanogenesis in a hydrogenotrophic methanogen, Methanococcus maripaludis. We generated a comprehensive list of genes (protein-coding and non-coding) for M. maripaludis through integrated analysis of the transcriptome structure and a newly constructed Peptide Atlas. The environment and gene-regulatory influence network (EGRIN) model of the strain was constructed from a compendium of transcriptome data that was collected over 58 different steady-state and time course experiments that were performed in chemostats or batch cultures, under a spectrum of environmental perturbations that modulated methanogenesis. Analyses of the EGRIN model have revealed novel components of methanogenesis that included at least three additional protein-coding genes of previously unknown function as well as one non-coding RNA. We discovered that at least five regulatory mechanisms act in a combinatorial scheme to inter-coordinate key steps of methanogenesis with different processes such as motility, ATP biosynthesis, and carbon assimilation. Through a combination of genetic and environmental perturbation experiments we have validated the EGRIN-predicted role of two novel transcription factors in the regulation of phosphate-dependent repression of formate dehydrogenase a key enzyme in the methanogenesis pathway. The EGRIN model demonstrates regulatory affiliations within methanogenesis as well as between methanogenesis and other cellular functions. In addition, we have published an analysis of transcriptome architecture in M. maripaludis and an analysis of the effects of H 2 and formate on growth yield and regulation of methanogenesis in M. maripaludis.

59 BASIC BIOLOGICAL SCIENCES↗

Accelerated Search for BaTiO 3 -Based Ceramics with Large Energy Storage at Low Fields Using Machine Learning and Experimental Design

The problem that is considered is that of maximizing the energy storage density of Pb-free BaTiO3-based dielectrics at low electric fields. It is demonstrated that how varying the size of the combinatorial search space influences the efficiency of material discovery by comparing the performance of two machine learning based approaches where different levels of physical insights are involved. It is started with physics intuition to provide guiding principles to find better performers lying in the crossover region in the composition–temperature phase diagram between the ferroelectric phase and relaxor ferroelectric phase. Such an approach is limiting for multidopant solid solutions and motivates the use of two data-driven machine learning and design strategies with a feedback loop to experiments. Strategy I considers learning and property prediction on all the compounds, and strategy II learns to preselect compounds in the crossover region on which prediction is carried out. By performing only two active learning loops via strategy II, the compound (Ba0.86Ca0.14)(Ti0.79Zr0.11Hf0.10)O3 is synthesized with the largest energy storage density ≈73 mJ cm-3 at a field of 20 kV cm-1 , and an insight into the relative performance of the strategies using varying levels of knowledge is provided.

36 MATERIALS SCIENCE↗

Supporting data for climatic clustering and longitudinal analysis with impacts on food, bioenergy, and pandemics

This data supports the conclusions found in climatic clustering and longitudinal analysis with impacts on food, bioenergy, and pandemics. Included here are (i) the binarized geolocation vectors used for exhaustive vector comparisons, (ii) the resulting climatic networks, (iii) the results of applying Markov clustering to the climatic networks, and (iv) the results of applying Correlation-of-Correlations (cor-cor) to the climatic networks. The set of binarized geolocation vectors that are used as inputs for the Combinatorial Metrics library (CoMet) are of the form comet-UUUUUxVVVVV-XXXX-YYYY.shuffled.tped where UUUUU is the number of vectors, VVVVV is the length of each vector, XXXX is the starting year, and YYYY is the ending year. Each line corresponds to a geolocation vector of binary elements A (i.e., 0) and T (i.e., 1). The set of climatic networks that are used for downstream network analysis are of the form network-U-way-XXXX-YYYY.parsed.txt where U is the order of the comparison (2-way or 3-way), XXXX is the starting year, and YYYY is the ending year. Each line corresponds to an edge linking two geolocations (defined by latitude and longitude) with its corresponding edge weight (i.e., DUO score). The set of cluster results are of the form clusters-U-way-XXXX-YYYY-thresh-VVVV-inflation-WWW.clustered.txt where U is the order of the comparison (2-way or 3-way), XXXX is the starting year, YYYY is the ending year, VVVV is the similarity threshold, and WWW is the Markov clustering inflation rate. Each line corresponds to a single cluster and is composed of a number of corresponding geolocations (defined by latitude and longitude). The set of cor-cor results are of the form corcor-U-way-XXXX-YYYY.cumulative.txt where U is the order of the comparison (2-way or 3-way), XXXX is the starting year, and YYYY is the ending year. Each line corresponds to a single geolocation with it's corresponding cor-cor value.

54 ENVIRONMENTAL SCIENCES↗

SPARTAN (Scalable Probabilistic Application Reconfigurable Tensor Autonomous Network)

The technical founder of Ludwig Computing Inc has been competitively selected for support by Cyclotron Road, a U.S. Department of Energy (DOE) Advanced Manufacturing Office (AMO) Lab-Embedded Entrepreneurship Program (LEEP) through an approved merit review process. Ludwig Computing Inc, supported by the U.S. Department of Energy's Advanced Manufacturing Office through the Cyclotron Road program, has investigated the advantages of probabilistic computing for real-world compute-intensive applications. This research adds to the understanding of alternative computing paradigms by exploring a unique hardware-software co-design that integrates quantum computing methods with nature-inspired problem-solving techniques. The project's focus on areas such as combinatorial optimization, graph analytics, and machine learning demonstrates the potential for significant advancements in computational efficiency and performance. By harnessing natural randomness to streamline large circuits into fewer devices, Ludwig's approach enables massive parallelism, potentially offering higher throughput, speed, and energy efficiency compared to conventional hardware solutions. This work benefits the public by paving the way for more efficient computing solutions that could address complex real-world problems while potentially reducing energy consumption in data-intensive industries.

97 MATHEMATICS AND COMPUTING↗

The HEP.TrkX Project: Deep Learning for Particle Tracking

Charged particle reconstruction in dense environments, such as the detectors of the High Luminosity Large Hadron Collider (HL-LHC) is a challenging pattern recognition problem. Traditional tracking algorithms, such as the combinatorial Kalman Filter, have been used with great success in HEP experiments for years. However, these state-of-the-art techniques are inherently sequential and scale quadratically or worse with increased detector occupancy. The HEP.TrkX project is a pilot project with the aim to identify and develop cross-experiment solutions based on machine learning algorithms for track reconstruction. Machine learning algorithms bring a lot of potential to this problem thanks to their capability to model complex non-linear data dependencies, to learn effective representations of high-dimensional data through training, and to parallelize easily on high-throughput architectures such as FPGAs or GPUs. In this paper we present the evolution and performance of our recurrent (LSTM) and convolutional neural networks moving from basic 2D models to more complex models and the challenges of scaling up to realistic dimensionality/sparsity.

Tsaris, Aristeidis↗

Recognition of Multivalent Histone States Associated with Heterochromatin by UHRF1 Protein

Histone modifications and DNA methylation represent two layers of heritable epigenetic information that regulate eukaryotic chromatin structure and gene activity. UHRF1 is a unique factor that bridges these two layers; it is required for maintenance DNA methylation at hemimethylated CpG sites, which are specifically recognized through its SRA domain and also interacts with histone H3 trimethylated on lysine 9 (H3K9me3) in an unspecified manner. Here we show that UHRF1 contains a tandem Tudor domain (TTD) that recognizes H3 tail peptides with the heterochromatin-associated modification state of trimethylated lysine 9 and unmodified lysine 4 (H3K4me0/K9me3). Solution NMR and crystallographic data reveal the TTD simultaneously recognizes H3K9me3 through a conserved aromatic cage in the first Tudor subdomain and unmodified H3K4 within a groove between the tandem subdomains. The subdomains undergo a conformational adjustment upon peptide binding, distinct from previously reported mechanisms for dual histone mark recognition. Mutant UHRF1 protein deficient for H3K4me0/K9me3 binding shows altered localization to heterochromatic chromocenters and fails to reduce expression of a target gene, p16{sup INK4A}, when overexpressed. Our results demonstrate a novel recognition mechanism for the combinatorial readout of histone modification states associated with gene silencing and add to the growing evidence for coordination of, and cross-talk between, the modification states of H3K4 and H3K9 in regulation of gene expression.

59 BASIC BIOLOGICAL SCIENCES↗

Massive Associative K-biclustering (MAK) v1.0

Massive Associative K-biclustering (MAK) is a method to address the key limitations in other biclustering approaches. In a formal evaluation against new simulated data sets with more realistic properties MAK improves recovery of biclusters exhibiting common behavior across conditions, with the highest F1 scores, a measure of a test's accuracy, and lowest false positive rates, the probability of falsely rejecting the null hypothesis, compared to established methods. The developers show how MAK can be applied to a popular data set from Saccharomyces cerevisiae to uncover new properties not captured by other methods. MAK biclusters showed increased statistical enrichments for known regulatory associations and gene functions, including for nearly 30% of the novel MAK biclusters not found by other methods. Compared to other methods, MAK results had at least two-fold higher coverage of differential expression in the overall dataset, an overlapping bicluster network with distinct modules for protein expression and energy metabolism, and biologically relevant overlapping biclusters including cases of combinatorial regulation. MAK is suited for large datasets with many potentially noisy and overlapping patterns of varying sizes.

JOACHIMIAK, MARCIN↗

Selecting a Classification Ensemble and Detecting Process Drift in an Evolving Data Stream

We characterize the commercial behavior of a group of companies in a common line of business using a small ensemble of classifiers on a stream of records containing commercial activity information. This approach is able to effectively find a subset of classifiers that can be used to predict company labels with reasonable accuracy. Performance of the ensemble, its error rate under stable conditions, can be characterized using an exponentially weighted moving average (EWMA) statistic. The behavior of the EWMA statistic can be used to monitor a record stream from the commercial network and determine when significant changes have occurred. Results indicate that larger classification ensembles may not necessarily be optimal, pointing to the need to search the combinatorial classifier space in a systematic way. Results also show that current and past performance of an ensemble can be used to detect when statistically significant changes in the activity of the network have occurred. The dataset used in this work contains tens of thousands of high level commercial activity records with continuous and categorical variables and hundreds of labels, making classification challenging.

Ensemble classifiers↗

Using Bacteria to Determine Protein Kinase Specificity and Predict Target Substrates

The identification of protein kinase targets remains a significant bottleneck for our understanding of signal transduction in normal and diseased cellular states. Kinases recognize their substrates in part through sequence motifs on substrate proteins, which, to date, have most effectively been elucidated using combinatorial peptide library approaches. Here, we present and demonstrate the ProPeL method for easy and accurate discovery of kinase specificity motifs through the use of native bacterial proteomes that serve as in vivo libraries for thousands of simultaneous phosphorylation reactions. Using recombinant kinases expressed in E. coli followed by mass spectrometry, the approach accurately recapitulated the well established motif preferences of human basophilic (Protein Kinase A) and acidophilic (Casein Kinase II) kinases. These motifs, derived for PKA and CK II using only bacterial sequence data, were then further validated by utilizing them in conjunction with the scan-x software program to computationally predict known human phosphorylation sites with high confidence.

59 BASIC BIOLOGICAL SCIENCES↗

Rapid Chemometric Filtering of Spectral Data

A method of rapid, programmable filtering of spectral transmittance, reflectance, or fluorescence data to measure the concentrations of chemical species has been proposed. By programmable is meant that a variety of spectral analyses can readily be performed and modified in software, firmware, and/or electronic hardware, without need to change optical filters or other optical hardware of the associated spectrometers. The method is intended to enable real-time identification of single or multiple target chemical species in applications that involve high-throughput screening of multiple samples. Examples of such applications include (but are not limited to) combinatorial chemistry, flow cytometry, bead assays, testing drugs, remote sensing, and identification of targets. The basic concept of the proposed method is to perform real-time crosscorrelations of a measured spectrum with one or more analytical function(s) of wavelength that could be, for example, the known spectra of target species. Assuming that measured spectral intensities are proportional to concentrations of target species plus background spectral intensities, then after subtraction of background levels, it should be possible to determine target species concentrations from cross-correlation values. Of course, the problem of determining the concentrations is more complex when spectra of different species overlap, but the problem can be solved by use of multiple analytical functions in combination with computational techniques that have been developed previously for analyses of this type. The method is applicable to the design and operation of a spectrometer in which spectrally dispersed light is measured by means of an active-pixel sensor (APS) array. The row or column dimension of such an array is generally chosen to be aligned along the spectral-dispersion dimension, so that each pixel intercepts light in a narrow spectral band centered on a wavelength that is a known function of the pixel position. The proposed method admits of two hardware implementations for computing cross-correlations in real time.

Beaman, Gregory↗

Chemoproteogenomic stratification of the missense variant cysteinome

Abstract Cancer genomes are rife with genetic variants; one key outcome of this variation is widespread gain-of-cysteine mutations. These acquired cysteines can be both driver mutations and sites targeted by precision therapies. However, despite their ubiquity, nearly all acquired cysteines remain unidentified via chemoproteomics; identification is a critical step to enable functional analysis, including assessment of potential druggability and susceptibility to oxidation. Here, we pair cysteine chemoproteomics—a technique that enables proteome-wide pinpointing of functional, redox sensitive, and potentially druggable residues—with genomics to reveal the hidden landscape of cysteine genetic variation. Our chemoproteogenomics platform integrates chemoproteomic, whole exome, and RNA-seq data, with a customized two-stage false discovery rate (FDR) error controlled proteomic search, which is further enhanced with a user-friendly FragPipe interface. Chemoproteogenomics analysis reveals that cysteine acquisition is a ubiquitous feature of both healthy and cancer genomes that is further elevated in the context of decreased DNA repair. Reference cysteines proximal to missense variants are also found to be pervasive, supporting heretofore untapped opportunities for variant-specific chemical probe development campaigns. As chemoproteogenomics is further distinguished by sample-matched combinatorial variant databases and is compatible with redox proteomics and small molecule screening, we expect widespread utility in guiding proteoform-specific biology and therapeutic discovery.

Desai, Heta (ORCID:0000000343621707)↗

The Materials Provenance Store

Abstract We present a database resulting from high throughput experimentation, primarily on metal oxide solid state materials. The central relational database, the Materials Provenance Store (MPS), manages the metadata and experimental provenance from acquisition of raw materials, through synthesis, to a broad range of materials characterization techniques. Given the primary research goal of materials discovery of solar fuels materials, many of the characterization experiments involve electrochemistry, along with optical, structural, and compositional characterizations. The MPS is populated with all information required for executing common data queries, which typically do not involve direct query of raw data. The result is a database file that can be distributed to users so that they can independently execute queries and subsequently download the data of interest. We propose this strategy as an approach to manage the highly heterogeneous and distributed data that arises from materials science experiments, as demonstrated by the management of over 30 million experiments run on over 12 million samples in the present MPS release.

36 MATERIALS SCIENCE↗

Acrolein cytotoxicity in hepatocytes involves endoplasmic reticulum stress, mitochondrial dysfunction and oxidative stress

Acrolein is a common environmental, food and water pollutant and a major component of cigarette smoke. Also, it is produced endogenously via lipid peroxidation and cellular metabolism of certain amino acids and drugs. Acrolein is cytotoxic to many cell types including hepatocytes; however the mechanisms are not fully understood. We examined the molecular mechanisms underlying acrolein hepatotoxicity in primary human hepatocytes and hepatoma cells. Acrolein, at pathophysiological concentrations, caused a dose-dependent loss of viability of hepatocytes. The death was apoptotic at moderate and necrotic at high concentrations of acrolein. Acrolein exposure rapidly and dramatically decreased intracellular glutathione and overall antioxidant capacity, and activated the stress-signaling MAP-kinases JNK, p42/44 and p38. Our data demonstrate for the first time in human hepatocytes, that acrolein triggered endoplasmic reticulum (ER) stress and activated eIF2α, ATF-3 and -4, and Gadd153/CHOP, resulting in cell death. Notably, the protective/adaptive component of ER stress was not activated, and acrolein failed to up-regulate the protective ER-chaperones, GRP78 and GRP94. Additionally, exposure to acrolein disrupted mitochondrial integrity/function, and led to the release of pro-apoptotic proteins and ATP depletion. Acrolein-induced cell death was attenuated by N-acetyl cysteine, phenyl-butyric acid, and caspase and JNK inhibitors. Our data demonstrate that exposure to acrolein induces a variety of stress responses in hepatocytes, including GSH depletion, oxidative stress, mitochondrial dysfunction and ER stress (without ER-protective responses) which together contribute to acrolein toxicity. Our study defines basic mechanisms underlying liver injury caused by reactive aldehyde pollutants such as acrolein. -- Highlights: ► Human primary hepatocytes and cultured cell lines are used. ► Multiple cell death signaling pathways are activated by acrolein. ► Novel finding of acrolein-induced ER stress ► Acrolein fails to activate ER stress-induced protective responses. ► Combinatorial therapies may be needed for preventing acrolein hepatotoxicity.

60 APPLIED LIFE SCIENCES↗

Film Thickness and Composition Effects in Symmetric Ternary Block Copolymer/Homopolymer Blend Films: Domain Spacing and Orientation

We report thickness and homopolymer molar mass dependencies in thin film blends of a compositionally symmetric 75 kg/mol lamellar polystyrene-block-poly(methyl methacrylate) diblock copolymer (PS-b-PMMA) with near-equal volume fractions of PS and PMMA homopolymers on “neutral” surfaces that promote a vertical domain orientation. PS and PMMA homopolymers with equal molar masses of 1, 3, and 22 kg/mol are blended in a 1:1 mass ratio. To efficiently explore this parameter space, we prepared combinatorial samples with gradient homopolymer concentrations using electrospray deposition (ESD) and gradient film thicknesses using flow coating; these samples were subsequently characterized by grazing-incidence small-angle X-ray scattering and scanning electron microscopy, respectively. Our results demonstrate that the added homopolymers increase or reduce domain spacings with respect to the neat (unblended) block copolymer depending on their molar masses, in line with previous reports; a simple heuristic to estimate the domain scaling in ternary blends based on the homopolymer molar mass agrees well with data for the thinnest blend films studied here. More surprisingly, we observe that the measured domain spacing in blends also depends on film thickness, with thicker films exhibiting larger domain spacings than thinner films at the same homopolymer concentration. Furthermore, thickness gradients at fixed homopolymer concentrations reveal a change in lamellae orientation from vertical to horizontal as film thickness is increased beyond the nominal lamellar domain spacing. Coarse-grained molecular dynamics simulations indicate that this change in orientation is induced by homopolymer segregation to film surfaces. Finally, tying these effects together, we hypothesize that the observed dependence of domain spacing on film thickness is a consequence of changes in the vertical homopolymer composition profile in thicker films during lamellae reorientation and the conformational asymmetry between PMMA and PS segments.

36 MATERIALS SCIENCE↗

The HEP.TrkX Project: deep neural networks for HL-LHC online and offline tracking

Particle track reconstruction in dense environments such as the detectors of the High Luminosity Large Hadron Collider (HL-LHC) is a challenging pattern recognition problem. Traditional tracking algorithms such as the combinatorial Kalman Filter have been used with great success in LHC experiments for years. However, these state-of-the-art techniques are inherently sequential and scale poorly with the expected increases in detector occupancy in the HL-LHC conditions. The HEP.TrkX project is a pilot project with the aim to identify and develop cross-experiment solutions based on machine learning algorithms for track reconstruction. Machine learning algorithms bring a lot of potential to this problem thanks to their capability to model complex non-linear data dependencies, to learn effective representations of high-dimensional data through training, and to parallelize easily on high-throughput architectures such as GPUs. This contribution will describe our initial explorations into this relatively unexplored idea space. Furthermore, we will discuss the use of recurrent (LSTM) and convolutional neural networks to find and fit tracks in toy detector data.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗