Showing papers by "Lehigh University published in 2021"
01 Sep 2021
TL;DR: The methodology presented within this work is a result of years of interactions between many junior and senior X-ray Photoelectron Spectroscopy (XPS) users operating within the CasaXPS spectral processing and interpretation program framework.
Abstract: The methodology presented within this work is a result of years of interactions between many junior and senior X-ray Photoelectron Spectroscopy (XPS) users operating within the CasaXPS spectral processing and interpretation program framework. In particular, discussions arising from a series of workshops have been a significant source for developing the overall XPS data processing concept and are the motivation for creating this work. These workshops organized by the Institut des Materiaux Jean Rouxel (IMN), Nantes gather both experienced and novice users of XPS for a week of discourse in conceptual experiment design and the resulting data processing. However, the framework constructed and utilized within these workshops encouraged the dissemination of knowledge beyond XPS data analysis and emphasized the importance of a multi-disciplinary collaborative approach to surface analysis problem-solving. The material presented here embodies data treatment originating from data made available to the first CNRS Thematic Workshop presented at Roscoff 2013. The methodology described here has evolved over the subsequent workshops in 2016 and 2019 and currently represents the philosophy used in CasaXPS spectral data processing paradigm.
290 citations
TL;DR: In this paper, a liquid medium annealing (LMA) technology is used to create a robust chemical environment and constant heating field to modulate crystal growth over the entire film, which produces films with high crystallinity, fewer defects, desired stoichiometry, and overall film homogeneity.
Abstract: Solution processing of semiconductors is highly promising for the high-throughput production of cost-effective electronics and optoelectronics. Although hybrid perovskites have potential in various device applications, challenges remain in the development of high-quality materials with simultaneously improved processing reproducibility and scalability. Here, we report a liquid medium annealing (LMA) technology that creates a robust chemical environment and constant heating field to modulate crystal growth over the entire film. Our method produces films with high crystallinity, fewer defects, desired stoichiometry, and overall film homogeneity. The resulting perovskite solar cells (PSCs) yield a stabilized power output of 24.04% (certified 23.7%, 0.08 cm2) and maintain 95% of their initial power conversion efficiency (PCE) after 2000 hours of operation. In addition, the 1-cm2 PSCs exhibit a stabilized power output of 23.15% (certified PCE 22.3%) and keep 90% of their initial PCE after 1120 hours of operation, which illustrates their feasibility for scalable fabrication. LMA is less climate dependent and produces devices in-house with negligible performance variance year round. This method thus opens a new and effective avenue to improving the quality of perovskite films and photovoltaic devices in a scalable and reproducible manner.
164 citations
TL;DR: PURPOSENivolumab as discussed by the authors received US Food and Drug Administration approval as a single agent or in combination with ipilimumab in patients with microsatellite instability-high/mismatch repair-deficient (MS...
Abstract: PURPOSENivolumab received US Food and Drug Administration approval as a single agent or in combination with ipilimumab in patients with microsatellite instability-high/mismatch repair-deficient (MS...
141 citations
Texas A&M University1, University of Exeter2, University of Helsinki3, Université du Québec à Montréal4, Tanjungpura University5, University of Hawaii at Manoa6, University of Bristol7, Bowdoin College8, Chulalongkorn University9, University of California, Los Angeles10, Max Planck Society11, University of Nottingham12, University of Magallanes13, Oeschger Centre for Climate Change Research14, Université de Montréal15, Lehigh University16, Northeast Normal University17, Mount Holyoke College18, McGill University19, Stockholm University20, University of Leicester21, Katholieke Universiteit Leuven22, University of St Andrews23, Florida State University24, Aarhus University25, University of Toronto26, University of New Hampshire27, University of Łódź28, Centre national de la recherche scientifique29, Cranfield University30, University of Alberta31, Stockholm Environment Institute32, Lawrence Berkeley National Laboratory33, United States Geological Survey34, Texas A&M University at Galveston35, University of Victoria36, Adam Mickiewicz University in Poznań37, Finnish Meteorological Institute38, Royal Holloway, University of London39, University of Queensland40, Lamont–Doherty Earth Observatory41, National Park Service42, University of York43, Hope College44, University of Reading45, Uva Wellassa University46, Queen's University Belfast47, University of California, Berkeley48, Memorial University of Newfoundland49
TL;DR: In this article, the authors define and quantify the leading drivers of change that have impacted peatland carbon stocks during the Holocene and predict their effect during this century and in the far future.
Abstract: The carbon balance of peatlands is predicted to shift from a sink to a source this century. However, peatland ecosystems are still omitted from the main Earth system models that are used for future climate change projections, and they are not considered in integrated assessment models that are used in impact and mitigation studies. By using evidence synthesized from the literature and an expert elicitation, we define and quantify the leading drivers of change that have impacted peatland carbon stocks during the Holocene and predict their effect during this century and in the far future. We also identify uncertainties and knowledge gaps in the scientific community and provide insight towards better integration of peatlands into modelling frameworks. Given the importance of the contribution by peatlands to the global carbon cycle, this study shows that peatland science is a critical research area and that we still have a long way to go to fully understand the peatland–carbon–climate nexus. Peatlands are impacted by climate and land-use changes, with feedback to warming by acting as either sources or sinks of carbon. Expert elicitation combined with literature review reveals key drivers of change that alter peatland carbon dynamics, with implications for improving models.
141 citations
Massachusetts Institute of Technology1, University of Texas at Austin2, University of Wisconsin-Madison3, Dartmouth College4, Smithsonian Institution5, California Institute of Technology6, University of New Mexico7, University of Bern8, Lehigh University9, Open University10, Columbia University11, University of Florida12, University of California, Santa Cruz13, University of Maryland, College Park14, University of Cambridge15, Space Telescope Science Institute16, Goddard Space Flight Center17, University of Maryland, Baltimore County18, Ames Research Center19, Search for extraterrestrial intelligence20, Princeton University21, University of Chicago22, University of Colorado Boulder23, Technical University of Denmark24, Arizona State University25, Cornell University26, University of California, Riverside27, Aarhus University28, Max Planck Society29, Spanish National Research Council30, Western Connecticut State University31, INAF32, Fisk University33, Vanderbilt University34, University of Geneva35
TL;DR: In this article, the authors presented 2241 exoplanet candidates identified with data from the Transiting Exoplanet Survey Satellite (TESS) during its 2-year Prime Mission.
Abstract: We present 2241 exoplanet candidates identified with data from the Transiting Exoplanet Survey Satellite (TESS) during its 2 yr Prime Mission. We list these candidates in the TESS Objects of Interest (TOI) Catalog, which includes both new planet candidates found by TESS and previously known planets recovered by TESS observations. We describe the process used to identify TOIs, investigate the characteristics of the new planet candidates, and discuss some notable TESS planet discoveries. The TOI catalog includes an unprecedented number of small planet candidates around nearby bright stars, which are well suited for detailed follow-up observations. The TESS data products for the Prime Mission (sectors 1-26), including the TOI catalog, light curves, full-frame images, and target pixel files, are publicly available at the Mikulski Archive for Space Telescopes.
140 citations
TL;DR: Odom et al. as mentioned in this paper described a set of practices that have evidence of positive effects with autistic children and youth, and reviewed 972 articles from which they found 28 focused intervention practices that met the criteria for evidence-based practice (EBP).
Abstract: This systematic review describes a set of practices that have evidence of positive effects with autistic children and youth. This is the third iteration of a review of the intervention literature (Odom et al. in J Autism Dev Disorders 40(4):425-436, 2010a; Prevent School Fail 54(4):275-282, 2010b; Wong et al. in https://autismpdc.fpg.unc.edu/sites/autismpdc.fpg.unc.edu/files/imce/documents/2014-EBP-Report.pdf ; J Autism Dev Disorders 45(7):1951-1966, 2015), extending coverage to articles published between 1990 and 2017. A search initially yielded 31,779 articles, and the subsequent screening and evaluation process found 567 studies to include. Combined with the previous review, 972 articles were synthesized, from which the authors found 28 focused intervention practices that met the criteria for evidence-based practice (EBP). Former EBPs were recategorized and some manualized interventions were distinguished as meeting EBP criteria. The authors discuss implications for current practices and future research.
135 citations
TL;DR: A new deep learning algorithm is proposed for the automated diagnosis of COVID-19, which only requires a few samples for training and uses contrastive learning to train an encoder which can capture expressive feature representations on large and publicly available lung datasets and adopt the prototypical network for classification.
122 citations
TL;DR: For the first time, these sets were compiled using a data-driven selection process supported by the solution of a sequence of mixed integer optimization problems, which encode requirements on diversity and balancedness with respect to instance features and performance data.
Abstract: We report on the selection process leading to the sixth version of the Mixed Integer Programming Library, MIPLIB 2017. Selected from an initial pool of 5721 instances, the new MIPLIB 2017 collection consists of 1065 instances. A subset of 240 instances was specially selected for benchmarking solver performance. For the first time, these sets were compiled using a data-driven selection process supported by the solution of a sequence of mixed integer optimization problems, which encode requirements on diversity and balancedness with respect to instance features and performance data.
116 citations
TL;DR: This paper proposes to use graph convolutional policy network based on reinforcement learning to generate dynamic graphs when the dynamic graphs are incomplete due to the data sparsity, and demonstrates that the model can achieve stable and effective long-term predictions of traffic flow, and can reduce the impact of data defects on prediction results.
108 citations
TL;DR: In this paper, the first evidence of a non-monotonic variation in the kurtosis times variance of the net-proton number (proxy for net-baryon number) distribution as a function of collision energy was reported.
Abstract: Nonmonotonic variation with collision energy (sqrt[s_{NN}]) of the moments of the net-baryon number distribution in heavy-ion collisions, related to the correlation length and the susceptibilities of the system, is suggested as a signature for the quantum chromodynamics critical point. We report the first evidence of a nonmonotonic variation in the kurtosis times variance of the net-proton number (proxy for net-baryon number) distribution as a function of sqrt[s_{NN}] with 3.1 σ significance for head-on (central) gold-on-gold (Au+Au) collisions measured solenoidal tracker at Relativistic Heavy Ion Collider. Data in noncentral Au+Au collisions and models of heavy-ion collisions without a critical point show a monotonic variation as a function of sqrt[s_{NN}].
101 citations
TL;DR: Zhang et al. as discussed by the authors proposed a hierarchical taxonomy-aware and attentional graph capsule recurrent CNNs framework for large-scale multi-label text classification, where each document was modeled as a word order preserved graph-of-words and normalized it as a corresponding word matrix representation preserving both the non-consecutive, long-distance and local sequential semantics.
Abstract: CNNs, RNNs, GCNs, and CapsNets have shown significant insights in representation learning and are widely used in various text mining tasks such as large-scale multi-label text classification. Most existing deep models for multi-label text classification consider either the non-consecutive and long-distance semantics or the sequential semantics. However, how to coherently take them into account is still far from studied. In addition, most existing methods treat output labels as independent medoids, ignoring the hierarchical relationships among them, which leads to a substantial loss of useful semantic information. In this paper, we propose a novel hierarchical taxonomy-aware and attentional graph capsule recurrent CNNs framework for large-scale multi-label text classification. Specifically, we first propose to model each document as a word order preserved graph-of-words and normalize it as a corresponding word matrix representation preserving both the non-consecutive, long-distance and local sequential semantics. Then the word matrix is input to the proposed attentional graph capsule recurrent CNNs for effectively learning the semantic features. To leverage the hierarchical relations among the class labels, we propose a hierarchical taxonomy embedding method to learn their representations, and define a novel weighted margin loss by incorporating the label representation similarity. Extensive evaluations on three datasets show that our model significantly improves the performance of large-scale multi-label text classification by comparing with state-of-the-art approaches.
11 Jul 2021
TL;DR: Zhang et al. as discussed by the authors proposed a new framework, UPFD, which simultaneously captures various signals from user preferences by joint content and graph modeling, and evaluated the effectiveness of the proposed framework on real-world datasets.
Abstract: Disinformation and fake news have posed detrimental effects on individuals and society in recent years, attracting broad attention to fake news detection. The majority of existing fake news detection algorithms focus on mining news content and/or the surrounding exogenous context for discovering deceptive signals; while the endogenous preference of a user when he/she decides to spread a piece of fake news or not is ignored. The confirmation bias theory has indicated that a user is more likely to spread a piece of fake news when it confirms his/her existing beliefs/preferences. Users' historical, social engagements such as posts provide rich information about users' preferences toward news and have great potentials to advance fake news detection. However, the work on exploring user preference for fake news detection is somewhat limited. Therefore, in this paper, we study the novel problem of exploiting user preference for fake news detection. We propose a new framework, UPFD, which simultaneously captures various signals from user preferences by joint content and graph modeling. Experimental results on real-world datasets demonstrate the effectiveness of the proposed framework. We release our code and data as a benchmark for GNN-based fake news detection: https://github.com/safe-graph/GNN-FakeNews.
TL;DR: The identification of two clinically relevant subtypes of post-traumatic stress disorder (PTSD) and major depressive disorder (MDD) on the basis of robust and distinct functional-connectivity patterns, prominently within the frontoparietal-control and default-mode networks are reported.
Abstract: The understanding and treatment of psychiatric disorders, which are known to be neurobiologically and clinically heterogeneous, could benefit from the data-driven identification of disease subtypes. Here, we report the identification of two clinically relevant subtypes of post-traumatic stress disorder (PTSD) and major depressive disorder (MDD) on the basis of robust and distinct functional connectivity patterns, prominently within the frontoparietal control network and the default mode network. We identified the disease subtypes by analysing, via unsupervised and supervised machine learning, the power-envelope-based connectivity of signals reconstructed from high-density resting-state electroencephalography in four datasets of patients with PTSD and MDD, and show that the subtypes are transferable across independent datasets recorded under different conditions. The subtype whose functional connectivity differed most from those of healthy controls was less responsive to psychotherapy treatment for PTSD and failed to respond to an antidepressant medication for MDD. By contrast, both subtypes responded equally well to two different forms of repetitive transcranial magnetic stimulation therapy for MDD. Our data-driven approach may constitute a generalizable solution for connectome-based diagnosis.
TL;DR: In this article, the authors examine five forms of impact, namely, scholarly, practical, societal, policy, and educational, outlining how scholars can systematically extend or enlarge their research agenda or projects to amplify their impact on the challenges societies face.
Abstract: The world is undergoing dramatic transformations. Many of the grand societal challenges we currently face underscore the need for scholarly research – including management studies – that can help us best sort out and solve them. Yet, management scholars struggle to produce concrete solutions or to communicate how their research can help to tackle these grand societal challenges. With this editorial, we want to help scholars seeking to ‘make a difference’ by broadening our understanding of what constitutes impactful research. We examine five forms of impact – scholarly, practical, societal, policy, and educational – outlining how scholars can systematically extend or enlarge their research agenda or projects to amplify their impact on the challenges societies face. We suggest that each of these forms of impact has intrinsic value in advancing the scientific enterprise and, together, can help to address key societal problems that reach beyond the immediate and traditional context of business management. With concrete suggestions for getting started on these forms of impact, and possible outputs for each, we hope to stimulate management and organization scholars to think more broadly about the opportunities for making an impact with their research and to begin doing so more often.
19 Apr 2021
TL;DR: Wang et al. as discussed by the authors proposed a hierarchical subgraph-level selection and embedding-based graph neural network for graph classification, which can learn more discriminative subgraph representations and respond in an explanatory way.
Abstract: Graph representation learning has attracted increasing research attention. However, most existing studies fuse all structural features and node attributes to provide an overarching view of graphs, neglecting finer substructures’ semantics, and suffering from interpretation enigmas. This paper presents a novel hierarchical subgraph-level selection and embedding-based graph neural network for graph classification, namely SUGAR, to learn more discriminative subgraph representations and respond in an explanatory way. SUGAR reconstructs a sketched graph by extracting striking subgraphs as the representative part of the original graph to reveal subgraph-level patterns. To adaptively select striking subgraphs without prior knowledge, we develop a reinforcement pooling mechanism, which improves the generalization ability of the model. To differentiate subgraph representations among graphs, we present a self-supervised mutual information mechanism to encourage subgraph embedding to be mindful of the global graph structural properties by maximizing their mutual information. Extensive experiments on six typical bioinformatics datasets demonstrate a significant and consistent improvement in model quality with competitive performance and interpretability.
TL;DR: In this article, the authors analyzed several methods for approximating gradients of noisy functions using only function values, including finite differences, linear interpolation, Gaussian smoothing, and smoothing on a sphere.
Abstract: In this paper, we analyze several methods for approximating gradients of noisy functions using only function values. These methods include finite differences, linear interpolation, Gaussian smoothing, and smoothing on a sphere. The methods differ in the number of functions sampled, the choice of the sample points, and the way in which the gradient approximations are derived. For each method, we derive bounds on the number of samples and the sampling radius which guarantee favorable convergence properties for a line search or fixed step size descent method. To this end, we use the results in Berahas et al. (Global convergence rate analysis of a generic line search algorithm with noise, arXiv:1910.04055
, 2019) and show how each method can satisfy the sufficient conditions, possibly only with some sufficiently large probability at each iteration, as happens to be the case with Gaussian smoothing and smoothing on a sphere. Finally, we present numerical results evaluating the quality of the gradient approximations as well as their performance in conjunction with a line search derivative-free optimization algorithm.
TL;DR: The TAPUR Study as discussed by the authors is a phase II basket trial that aims to identify signals of antitumor activity of commercially available targeted agents in patients with advanced cancers harboring genomic a...
Abstract: PURPOSEThe TAPUR Study is a phase II basket trial that aims to identify signals of antitumor activity of commercially available targeted agents in patients with advanced cancers harboring genomic a...
TL;DR: Recent progress is highlighted in developing new computational and theoretical approaches to study the structure and dynamics of monomeric and order higher assemblies of IDPs, with a particular emphasis on their phase separation into protein-rich condensates.
TL;DR: In this paper, Zhao et al. used force spectroscopy and steered molecular dynamics (SMD) simulations to quantify the specific interactions between SARS-CoV-2 and ACE2.
TL;DR: In this article, the use of soft oxidants, such as CO2, N2O, S-conta, and S-ContaO 2, was used for the O2-assisted dehydrogenation of propane to propylene.
Abstract: Oxidative dehydrogenation of propane to propylene can be achieved using conventional, oxygen-assisted dehydrogenation of propane (O2–ODHP) or via the use of soft oxidants, such as CO2, N2O, S-conta...
TL;DR: In this article, Bimetallic Pd-Fe catalysts supported on TiO2 are shown to be highly effective toward the selective oxidation of benzyl alcohol to benzaldehyde via the in situ production of H2O2 from molecular H2 and O2, under conditions where no reaction is observed with molecular O2 alone.
Abstract: Bimetallic Pd-Fe catalysts supported on TiO2 are shown to be highly effective toward the selective oxidation of benzyl alcohol to benzaldehyde via the in situ production of H2O2 from molecular H2 and O2, under conditions where no reaction is observed with molecular O2 alone. The rate of benzyl alcohol oxidation observed over supported Pd-Fe nanoparticles is significantly higher than those of either Pd-Au or Pd-only analogues. This enhanced activity can be attributed to the bifunctionality of the Pd-Fe catalyst to both synthesize H2O2 and catalyze the production of oxygen-based radical specie,s as indicated by an electron paramagnetic resonance analysis. Further studies also reveal the noninnocent nature of the solvent, resulting in the propagation of radical generation pathways.
TL;DR: This review discusses the anchoring sites of the supported MOx species on the ZSM-5 support, molecular structures of the initial dispersed surface MOx sites, nature of the active sites during MDA, reaction mechanisms, rate-determining step, kinetics and catalyst activity of the MDA reaction.
Abstract: This review focuses on recent fundamental insights about methane dehydroaromatization (MDA) to benzene over ZSM-5-supported transition metal oxide-based catalysts (MOx/ZSM-5, where M = V, Cr, Mo, W, Re, Fe). Benzene is an important organic intermediate, used for the synthesis of chemicals like ethylbenzene, cumene, cyclohexane, nitrobenzene and alkylbenzene. Current production of benzene is primarily from crude oil processing, but due to the abundant availability of natural gas, there is much recent interest in developing direct processes to convert CH4 to liquid chemicals. Among the various gas-to-liquid methods, the thermodynamically-limited Methane DehydroAromatization (MDA) to benzene under non-oxidative conditions appears very promising as it circumvents deep oxidation of CH4 to CO2 and does not require the use of a co-reactant. The findings from the MDA catalysis literature is critically analyzed with emphasis on in situ and operando spectroscopic characterization to understand the molecular level details regarding the catalytic sites before and during the MDA reaction. Specifically, this review discusses the anchoring sites of the supported MOx species on the ZSM-5 support, molecular structures of the initial dispersed surface MOx sites, nature of the active sites during MDA, reaction mechanisms, rate-determining step, kinetics and catalyst activity of the MDA reaction. Finally, suggestions are given regarding future experimental investigations to fill the information gaps currently found in the literature.
Texas Tech University1, United States Department of the Army2, French Institute for Research in Computer Science and Automation3, University of Wisconsin-Madison4, Johns Hopkins University5, United States Army Research Laboratory6, National Institute for Occupational Safety and Health7, Carnegie Mellon University8, California Institute of Technology9, Max Planck Society10, New York University11, Massachusetts Institute of Technology12, Google13, Lehigh University14
TL;DR: It is posited that well-validated computer simulation can provide a virtual proving ground that in many cases is instrumental in understanding safely, faster, at lower costs, and more thoroughly how the robots of the future should be designed and controlled for safe operation and improved performance.
Abstract: The last five years marked a surge in interest for and use of smart robots, which operate in dynamic and unstructured environments and might interact with humans. We posit that well-validated computer simulation can provide a virtual proving ground that in many cases is instrumental in understanding safely, faster, at lower costs, and more thoroughly how the robots of the future should be designed and controlled for safe operation and improved performance. Against this backdrop, we discuss how simulation can help in robotics, barriers that currently prevent its broad adoption, and potential steps that can eliminate some of these barriers. The points and recommendations made concern the following simulation-in-robotics aspects: simulation of the dynamics of the robot; simulation of the virtual world; simulation of the sensing of this virtual world; simulation of the interaction between the human and the robot; and, in less depth, simulation of the communication between robots. This Perspectives contribution summarizes the points of view that coalesced during a 2018 National Science Foundation/Department of Defense/National Institute for Standards and Technology workshop dedicated to the topic at hand. The meeting brought together participants from a range of organizations, disciplines, and application fields, with expertise at the intersection of robotics, machine learning, and physics-based simulation.
TL;DR: Cawley et al. as discussed by the authors conducted a study to estimate the causal effect of obesity on direct medical care costs at the national and state levels and found that obesity increased costs in every category of care: inpatient, outpatient, and prescription drugs.
Abstract: BACKGROUND: After a dramatic increase in prevalence over several decades, obesity has become a major public health crisis in the United States. Research to date has consistently demonstrated a correlation between obesity and higher medical costs for a variety of U.S. subpopulations and specific categories of care. However, by examining associations rather than causal effects, previous studies likely underestimated the effect of obesity on medical expenditures. OBJECTIVE: To estimate the causal effect of obesity on direct medical care costs at the national and state levels. METHODS: This study is a pooled cross-sectional analysis of retrospective data from the 2001-2016 Medical Expenditure Panel Surveys. Adults aged 20-65 years with a biological child living in the household were included in the study sample. Primary outcomes were individual-level medical expenditures due to obesity, overall, as well as separately by type of payer and category of medical care. Results were reported at the national level and separately for the 20 most populous states. The expenditure estimates were obtained from 2-part models of instrumental variables in which the respondent's body mass index (BMI) was instrumented using the BMI of their biological child. RESULTS: Adults with obesity in the United States compared with those with normal weight experienced higher annual medical care costs by $2,505 or 100%, with costs increasing significantly with class of obesity, from 68.4% for class 1 to 233.6% for class 3. The effects of obesity raised costs in every category of care: inpatient, outpatient, and prescription drugs. Increases in medical expenditures due to obesity were higher for adults covered by public health insurance programs ($2,868) than for those having private health insurance ($2,058). In 2016, the aggregate medical cost due to obesity among adults in the United States was $260.6 billion. The increase in individual-level expenditures due to obesity varied considerably by state (e.g., 24.0% in Florida, 66.4% in New York, and 104.9% in Texas). CONCLUSIONS: The 2-part models of instrumental variables, which estimate the causal effects of obesity on direct medical costs, showed that the effect of obesity is greater than suggested by previous studies, which estimated only correlations. Much of the aggregate national cost of obesity-$260.6 billion-represents external costs, providing a rationale for interventions to prevent and reduce obesity. DISCLOSURES: Novo Nordisk financed the development of the study design, analysis, and interpretation of data, as well as writing support of the manuscript. Cawley, Biener, and Meyerhoefer received financial support from Novo Nordisk to conduct the research study on which this manuscript is based. Smolarz and Ramasamy are employees of Novo Nordisk. Ding and Zvenyach have no conflicts to declare. Our research has been presented as a poster at the 2020 Academy Health Annual Research Meeting (Virtual), July 28-August 6, 2020.
TL;DR: Wang et al. as mentioned in this paper designed a novel event-based meta-schema to characterize the semantic relatedness of social events and then built an event based heterogeneous information network (HIN) integrating information from external knowledge base.
Abstract: Events are happening in real world and real time, which can be planned and organized for occasions, such as social gatherings, festival celebrations, influential meetings, or sports activities. Social media platforms generate a lot of real-time text information regarding public events with different topics. However, mining social events is challenging because events typically exhibit heterogeneous texture and metadata are often ambiguous. In this article, we first design a novel event-based meta-schema to characterize the semantic relatedness of social events and then build an event-based heterogeneous information network (HIN) integrating information from external knowledge base. Second, we propose a novel Pairwise Popularity Graph Convolutional Network, named as PP-GCN, based on weighted meta-path instance similarity and textual semantic representation as inputs, to perform fine-grained social event categorization and learn the optimal weights of meta-paths in different tasks. Third, we propose a streaming social event detection and evolution discovery framework for HINs based on meta-path similarity search, historical information about meta-paths, and heterogeneous DBSCAN clustering method. Comprehensive experiments on real-world streaming social text data are conducted to compare various social event detection and evolution discovery algorithms. Experimental results demonstrate that our proposed framework outperforms other alternative social event detection and evolution discovery techniques.
TL;DR: The beer game is widely used in supply chain management classes to demonstrate the bullwhip effect and the importance of supply chain coordination.
Abstract: Problem definition: The beer game is widely used in supply chain management classes to demonstrate the bullwhip effect and the importance of supply chain coordination. The game is a decentralized, ...
TL;DR: In this article, a full-length SARS-CoV-2 S protein was modeled in a viral membrane including both open and closed conformations of the receptor-binding domain (RBD) and different templates for the stalk region.
Abstract: The spike (S) protein of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) mediates host cell entry by binding to angiotensin-converting enzyme 2 (ACE2) and is considered the major target for drug and vaccine development. We previously built fully glycosylated full-length SARS-CoV-2 S protein models in a viral membrane including both open and closed conformations of the receptor-binding domain (RBD) and different templates for the stalk region. In this work, multiple μs-long all-atom molecular dynamics simulations were performed to provide deeper insights into the structure and dynamics of S protein and glycan functions. Our simulations reveal that the highly flexible stalk is composed of two independent joints and most probable S protein orientations are competent for ACE2 binding. We identify multiple glycans stabilizing the open and/or closed states of the RBD and demonstrate that the exposure of antibody epitopes can be captured by detailed antibody-glycan clash analysis instead of commonly used accessible surface area analysis that tends to overestimate the impact of glycan shielding and neglect possible detailed interactions between glycan and antibodies. Overall, our observations offer structural and dynamic insights into the SARS-CoV-2 S protein and potentialize for guiding the design of effective antiviral therapeutics.
TL;DR: A mixed integer programming formulation to construct optimal decision trees of a prespecified size that takes the special structure of categorical features into account and allow combinatorial decisions (based on subsets of values of features) at each node.
Abstract: Decision trees have been a very popular class of predictive models for decades due to their interpretability and good performance on categorical features. However, they are not always robust and tend to overfit the data. Additionally, if allowed to grow large, they lose interpretability. In this paper, we present a mixed integer programming formulation to construct optimal decision trees of a prespecified size. We take the special structure of categorical features into account and allow combinatorial decisions (based on subsets of values of features) at each node. Our approach can also handle numerical features via thresholding. We show that very good accuracy can be achieved with small trees using moderately-sized training sets. The optimization problems we solve are tractable with modern solvers.
Nanjing University1, Chinese Academy of Sciences2, National Oceanography Centre, Southampton3, Australian National University4, Ohio State University5, Sun Yat-sen University6, Nanjing University of Information Science and Technology7, University of Hong Kong8, Lehigh University9, Northeast Normal University10, Jiangsu Normal University11
TL;DR: Wang et al. as mentioned in this paper reported paired mean annual air temperature and monsoon intensity proxy records over the past 88,000 years from Lake Tengchongqinghai in southwestern China.
Abstract: Orbital-scale global climatic changes during the late Quaternary are dominated by high-latitude influenced ~100,000-year global ice-age cycles and monsoon influenced ~23,000-year low-latitude hydroclimate variations. However, the shortage of highly-resolved land temperature records remains a limiting factor for achieving a comprehensive understanding of long-term low-latitude terrestrial climatic changes. Here, we report paired mean annual air temperature (MAAT) and monsoon intensity proxy records over the past 88,000 years from Lake Tengchongqinghai in southwestern China. While summer monsoon intensity follows the ~23,000-year precession beat found also in previous studies, we identify previously unrecognized warm periods at 88,000–71,000 and 45,000–22,000 years ago, with 2–3 °C amplitudes that are close to our recorded full glacial-interglacial range. Using advanced transient climate simulations and comparing with forcing factors, we find that these warm periods in our MAAT record probably depends on local annual mean insolation, which is controlled by Earth’s ~41,000-year obliquity cycles and is anti-phased to annual mean insolation at high latitudes. The coincidence of our identified warm periods and intervals of high-frequent dated archaeological evidence highlights the importance of temperature on anatomically modern humans in Asia during the last glacial stage.
TL;DR: In this article, the hydropathy scale (HPS) coarse-grained (CG) model is used to simulate sequence-specific behavior of intrinsically disordered proteins (IDPs), including their liquid-liquid phase separation (LLPS).
Abstract: We present improvements to the hydropathy scale (HPS) coarse-grained (CG) model for simulating sequence-specific behavior of intrinsically disordered proteins (IDPs), including their liquid-liquid phase separation (LLPS). The previous model based on an atomistic hydropathy scale by Kapcha and Rossky (KR scale) is not able to capture some well-known LLPS trends such as reduced phase separation propensity upon mutations (R-to-K and Y-to-F). Here, we propose to use the Urry hydropathy scale instead, which was derived from the inverse temperature transitions in a model polypeptide with guest residues X. We introduce two free parameters to shift (Δ) and scale (µ) the overall interaction strengths for the new model (HPS-Urry) and use the experimental radius of gyration for a diverse group of IDPs to find their optimal values. Interestingly, many possible (Δ, µ) combinations can be used for typical IDPs, but the phase behavior of a low-complexity (LC) sequence FUS is only well described by one of these models, which highlights the need for a careful validation strategy based on multiple proteins. The CG HPS-Urry model should enable accurate simulations of protein LLPS and provide a microscopically detailed view of molecular interactions.