Crosstalk among Oxidative Stress, Autophagy, and Apoptosis in the Protective Effects of Ginsenoside Rb1 on Brain Microvascular Endothelial Cells: A Mixed Computational and Experimental Study.

OBJECTIVE: Brain microvascular endothelial cells (BMECs) were found to shift from their usually inactive state to an active state in ischemic stroke (IS) and cause neuronal damage. Ginsenoside Rb1 (GRb1), a component derived from medicinal plants, is known for its pharmacological benefits in IS, but its protective effects on BMECs have yet to be explored. This study aimed to investigate the potential protective effects of GRb1 on BMECs. METHODS: An in vitro oxygen-glucose deprivation/reperfusion (OGD/R) model was established to mimic ischemia-reperfusion (I/R) injury. Bulk RNA-sequencing data were analyzed by using the Human Autophagy Database and various bioinformatic tools, including gene set enrichment analysis (GSEA), Gene Ontology (GO) classification and enrichment analysis, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, protein-protein interaction network analysis, and molecular docking. Experimental validation was also performed to ensure the reliability of our findings. RESULTS: Rb1 had a protective effect on BMECs subjected to OGD/R injury. Specifically, GRb1 was found to modulate the interplay between oxidative stress, apoptosis, and autophagy in BMECs. Key targets such as sequestosome 1 (SQSTM1/p62), autophagy related 5 (ATG5), and hypoxia-inducible factor 1-alpha (HIF-1α) were identified, highlighting their potential roles in mediating the protective effects of GRb1 against IS-induced damage. CONCLUSION: GRbl protects BMECs against OGD/R injury by influencing oxidative stress, apoptosis, and autophagy. The identification of SQSTM1/p62, ATG5, and HIF-1α as promising targets further supports the potential of GRb1 as a therapeutic agent for IS, providing a foundation for future research into its mechanisms and applications in IS treatment.

The Relationship Between Stressful Life Events and Depressive Symptoms in College Students: mediation by Parenting Style and Gender's Moderating Effect.

Aim: Stressful life events have a significant impact on the mental health of college students. Depression, as a prevalent psychological issue, has garnered attention in the field of college student mental health and is closely linked to it. Additionally, parenting style is identified as an important factor influencing the development of college students' mental health. Therefore, this study aims to explore the relationship between these three factors. Methods: A total of 8079 first-year college students from two medical universities in Shandong Province, China were surveyed. The Beck Depression Inventory was utilized to evaluate depressive symptoms among the college students, while the Adolescent Self-rating Life Events Checklist and the Egna Minnen Beträfande Uppfostran were employed to gather data. Subsequently, the SPSS macro program PROCESS was utilized to analyze both the mediating and moderating effects. All statistical analyses were conducted using SPSS 26.0. Results: The study found a detection rate of 6.3% for depressive symptoms among college students. The correlation analysis of this study showed that the stressful life events of college students were significantly positively correlated with depressive symptoms (r=0.261, p< 0.01). Each dimension of parenting style was associated with depressive symptoms in different degrees and directions. At the same time, parenting styles of all sizes play a partial mediating role between stressful life events and depressive symptoms in college students, gender plays a crucial regulatory role in this mediation. Conclusion: Stressful life events experienced by college students have a significant impact on their mental health. Early intervention through positive parenting styles from parents may prove to be beneficial in promoting the development of good mental health among college students.

Advancements in research on the effects of panax notoginseng saponin constituents in ameliorating learning and memory disorders.

Learning and memory disorder is a cluster of symptoms caused by neuronal aging and other diseases of the central nervous system (CNS). Panax notoginseng saponins (PNS) are a series of saponins derived from the natural active ingredients of traditional Chinese medicine (TCM) that have neuroprotective effects on the central nervous system. In this paper, we review the ameliorative effects and mechanisms of Panax notoginseng saponin-like components on learning and memory disorders to provide valuable references and insights for the development of new drugs for the treatment of learning and memory disorders. Our summary results suggest that Panax ginseng saponins have significant effects on improving learning and memory disorders, and these effects and potential mechanisms are mediated by their anti-inflammatory, anti-apoptotic, antioxidant, ß-amyloid lowering, mitochondrial homeostasis in vivo, neuronal structure and function improving, neurogenesis promoting, neurotransmitter release regulating, and probiotic homeostasis in vivo activities. These findings suggest the potential of Panax notoginseng saponin-like constituents as drug candidates for improving learning and memory disorders.

Qilong capsule prevents myocardial ischemia/reperfusion injury by inhibiting platelet activation via the platelet CD36 signaling pathway.

ETHNOPHARMACOLOGICAL RELEVANCE: Qilong capsule (QC) is developed from the traditional Chinese medicine formula Buyang Huanwu Decoction, which has been clinically used to invigorate Qi and promote blood circulation to eliminate blood stasis. Myocardial ischemia‒reperfusion injury (MIRI) can be attributed to Qi deficiency and blood stasis. However, the effects of QC on MIRI remain unclear. AIM OF THE STUDY: This study aimed to investigate the protective effect and possible mechanism of QC on platelet function in MIRI rats. MATERIALS AND METHODS: The left anterior descending artery of adult Sprague‒Dawley rats was ligated for 30 min and then reperfused for 120 min with or without QC treatment. Then, the whole blood viscosity, plasma viscosity, coagulation, platelet adhesion rate, platelet aggregation, and platelet release factors were evaluated. Platelet CD36 and its downstream signaling pathway-related proteins were detected by western blotting. Furthermore, the active components of QC and the molecular mechanism by which QC regulates platelet function were assessed via molecular docking, platelet aggregation tests in vitro and BLI analysis. RESULTS: We found that QC significantly reduced the whole blood viscosity, plasma viscosity, platelet adhesion rate, and platelet aggregation induced by ADP or AA in rats with MIRI. The inhibition of platelet activation by QC was associated with reduced levels of ß-TG, PF-4, P-selectin and PAF. Mechanistically, QC effectively attenuated the expression of platelet CD36 and thus inhibited the activation of Src, ERK5, and p38. The active components of QC apparently suppressed platelet aggregation in vitro and regulated the CD36 signaling pathway. CONCLUSIONS: QC improves MIRI-induced hemorheological disorders, which might be partly attributed to the inhibition of platelet activation via CD36-mediated platelet signaling pathways.

Influence of root distribution patterns on soil dynamic characteristics.

Slopes along the highway and railway routes are subjected to not only static loads but also dynamic loads generated by vehicles and trains. The induced excessive deformation potentially poses a threat to slope stability. In terms of the extensive application of ecological slope protection, plants play a critical role in slope stability, as the roots can enhance the shear strength of the soil. This study aims to investigate the influence of different root distribution patterns on the dynamic characteristics induced by cyclic loading. By conducting a group of dynamic triaxial tests, the results indicate that the root system can significantly enhance the liquefaction resistance of the soil when the soil is subjected to lower dynamic loads, and the cross arrangement has a better-reinforced effect than the mixed arrangement. The reinforced effect was not obvious when the soil was subjected to a dynamic load with a larger stress amplitude. In addition, based on the validation of the seed model, a new pore water pressure development model was proposed according to the test results. Overall, the research provides a new model and some innovative observations to better understand the dynamic behavior of root-reinforced soil.

Development of Whole Genome SNP-CAPS Markers and Preliminary QTL Mapping of Fruit Pedicel Traits in Watermelon.

Fruit pedicel (FP) is an important determinant of premium fruit quality that directly affects commercial market value. However, in-depth molecular and genetic basis of pedicel-related traits has not been identified in watermelon. Herein, a quantitative trait locus (QTL) mapping strategy was used to identify the potential genetic regions controlling FP traits based on newly derived whole-genome single nucleotide polymorphism based cleaved amplified polymorphism sequence (SNP-CAPS) markers. Next-generation sequencing based whole-genome re-sequencing of two watermelon parent lines revealed 98.30 and 98.40% of average coverage, 4,989,869 SNP variants, and 182,949 CAPS loci pairs across the reference genome, respectively. A total of 221 sets of codominant markers exhibited 46.42% polymorphism rate and were effectively genotyped within 100-F2:3 derived mapping population. The developed linkage map covered a total of 2,630.49 cM genetic length with averaged 11.90 cM, and depicted a valid marker-trait association. In total, 6 QTLs (qFPL4.1, qFPW4.1, qFPD2.1, qFPD2.2, qFPD8.1, qFPD10.1) were mapped with five major effects and one minor effect between the whole genome adjacent markers positioned over distinct chromosomes (02, 04, 08, 10), based on the ICIM-ADD mapping approach. These significant QTLs were similarly mapped in delimited flanking regions of 675.10, 751.38, 859.24, 948.39, and 947.51 kb, which collectively explained 8.64-13.60% PVE, respectively. A highly significant and positive correlation was found among the observed variables. To our knowledge, we first time reported the mapped QTLs/genes affecting FP traits of watermelon, and our illustrated outcomes will deliver the potential insights for fine genetic mapping as well as functional gene analysis through MAS-based breeding approaches.

Identification of Candidate Chromosome Region Related to Melon (Cucumis melo L.) Fruit Surface Groove Trait Through Biparental Genetic Mapping and Genome-Wide Association Study.

The melon fruit surface groove (fsg) not only affects peel structure and causes stress-induced fruit cracking but also fits consumers' requirements in different regions. In this study, genetic inheritance analysis of three F2 populations derived from six parental lines revealed that the fsg trait is controlled by a simple recessive inherited gene. Through bulked segregant analysis sequencing (BSA-seq), the Cmfsg locus was detected in an 8.96 Mb interval on chromosome 11 and then initially mapped to a region of approximately 1.15 Mb. Further fine mapping with a large F2 population including 1,200 plants narrowed this region to 207 kb containing 11 genes. A genome-wide association study (GWAS) with 187 melon accessions also produced the same chromosome region for the Cmfsg locus. Due to the rare molecular markers and lack of mutations in the coding and promoter regions of the 11 candidate genes in the fine-mapped interval, we conducted in silico BSA to explore the natural melon panel to predict candidate genes for the Cmfsg locus. A 1.07 kb segment upstream of MELO3C019694.2 (annotated as the AGAMOUS MADS-box transcription factor) exhibited a correlation with the grooved and non-grooved accessions among the F2 individuals, and a natural panel consisted of 17 melon accessions. The expression level of MELO3C019694.2 in the pericarp was higher in grooved lines than in non-grooved lines and was specifically expressed in fruit compared with other tissues (female flower, male flower, root, and leaf). This work provides fundamental information for further research on melon fsg trait formation and molecular markers for melon breeding.

Machine learning-driven multiscale modeling reveals lipid-dependent dynamics of RAS signaling proteins.

RAS is a signaling protein associated with the cell membrane that is mutated in up to 30% of human cancers. RAS signaling has been proposed to be regulated by dynamic heterogeneity of the cell membrane. Investigating such a mechanism requires near-atomistic detail at macroscopic temporal and spatial scales, which is not possible with conventional computational or experimental techniques. We demonstrate here a multiscale simulation infrastructure that uses machine learning to create a scale-bridging ensemble of over 100,000 simulations of active wild-type KRAS on a complex, asymmetric membrane. Initialized and validated with experimental data (including a new structure of active wild-type KRAS), these simulations represent a substantial advance in the ability to characterize RAS-membrane biology. We report distinctive patterns of local lipid composition that correlate with interfacially promiscuous RAS multimerization. These lipid fingerprints are coupled to RAS dynamics, predicted to influence effector binding, and therefore may be a mechanism for regulating cell signaling cascades.

Attribution-Driven Explanation of the Deep Neural Network Model via Conditional Microstructure Image Synthesis.

The materials science community has been increasingly interested in harnessing the power of deep learning to solve various domain challenges. However, despite their effectiveness in building highly predictive models, e.g., predicting material properties from microstructure imaging, due to their opaque nature fundamental challenges exist in extracting meaningful domain knowledge from the deep neural networks. In this work, we propose a technique for interpreting the behavior of deep learning models by injecting domain-specific attributes as tunable "knobs" in the material optimization analysis pipeline. By incorporating the material concepts in a generative modeling framework, we are able to explain what structure-to-property linkages these black-box models have learned, which provides scientists with a tool to leverage the full potential of deep learning for domain discoveries.

Adapted SIMPLE Algorithm for Incompressible SPH Fluids With a Broad Range Viscosity.

In simulating viscous incompressible SPH fluids, incompressibility and viscosity are typically solved in two separate stages. However, the interference between pressure and shear forces could cause the missing of behaviors that include preservation of sharp surface details and remarkable viscous behaviors such as buckling and rope coiling. To alleviate this problem, we introduce for the first time the semi-implicit method for pressure linked equations (SIMPLE) into SPH to solve incompressible fluids with a broad range viscosity. We propose to link incompressibility and viscosity solvers, and impose incompressibility and viscosity constraints iteratively to gradually remove the interference between pressure and shear forces. We will also discuss how to solve the particle deficiency problem for both incompressibility and viscosity solvers. Our method is stable at simulating incompressible fluids whose viscosity can range from zero to an extremely high value. Compared to state-of-the-art methods, our method not only produces realistic viscous behaviors, but is also better at preserving sharp surface details.

A Visual Comparison of Silent Error Propagation.

High-performance computing (HPC) systems play a critical role in facilitating scientific discoveries. Their scale and complexity (e.g., the number of computational units and software stack) continue to grow as new systems are expected to process increasingly more data and reduce computing time. However, with more processing elements, the probability that these systems will experience a random bit-flip error that corrupts a program's output also increases, which is often recognized as silent data corruption. Analyzing the resiliency of HPC applications in extreme-scale computing to silent data corruption is crucial but difficult. An HPC application often contains a large number of computation units that need to be tested, and error propagation caused by error corruption is complex and difficult to interpret. To accommodate this challenge, we propose an interactive visualization system that helps HPC researchers understand the resiliency of HPC applications and compare their error propagation. Our system models an application's error propagation to study a program's resiliency by constructing and visualizing its fault tolerance boundary. Coordinating with multiple interactive designs, our system enables domain experts to efficiently explore the complicated spatial and temporal correlation between error propagations. At the end, the system integrated a nonmonotonic error propagation analysis with an adjustable graph propagation visualization to help domain experts examine the details of error propagation and answer such questions as why an error is mitigated or amplified by program execution.

First Report of Maize Ear Rot Caused by Exserohilum rostratum in Hainan Province in Southern China.

Maize (Zea mays L.) is one of three major grain crops in China, with production reaching 261 million tons in 2019(NBS, 2020). Some fungi cause maize ear rot which lead to significant yield and quality losses. In 2016, about 5% of maize ears were dark brown and covered with a white mould in seed production fields in Lingshui, Hainan Province, China. These ears were brought back to the laboratory for analysis. Molded kernels were surface sterilized in 75% ethanol for 3 min and in 10% sodium hypochlorite for 3 min, subsequently rinsed three times in sterile-distilled water, placed onto potato dextrose agar (PDA), and incubated at 28°C in the dark for 3 days. mycelia tips grown from kernels were transferred into fresh PDA plates. Seven fungal isolates with similar morphology characteristics were obtained, and three of them were identified by morphology and molecular identification. The colonies grew rapidly. The aerial mycelia turned white to black with age. Conidia were straight to slightly curved, oval, pyriform or geniculate, brown to dark brown, and had 2 to 7 septa, with both basal and caudal septa thicker and darker than others, 39.47 to 78.66 ×13.96 to 22.78 µm, with a distinctly protruding hilum swelled from the basal cell. Conidiophores were dark brown, with geniculate tip and many septa. For molecular identification, genomic DNA of isolate was extracted from mycelia. The internal transcribed spacer (ITS), 1,3,8-trihydroxynaphthalene reductase (Brn) and glyceraldehyde-3-phosphate dehydrogenase-like (Gpd) genes were amplified with primers ITS1/ITS4 (White et al. 1990), Brn01/Brn02 (Shimizu et al. 1998) and gpd1/gpd 2 (Berbee et al. 1999) , respectively. BLASTn analysis showed that high identities with Exserohilum rostratum (ITS, LT837845.1, 100%; Brn, AY621165.1, 99.87%; Gpd, LT882543.1, 99.75%). Sequences of ITS, Brn and Gpd were deposited in GenBank with accession numbers MW362495, MW363953 and MW363954, respectively. Based on morphological characteristics and molecular analysis, the isolate was identified as E. rostratum (Hernández-Restrepo et al. 2018). Koch's postulates were completed by using ears of maize inbred line Huangzaosi and Chang7-2 growing in the experimental field of Baoding, Hebei Province. Three days post-silk emergence, each of the four maize ears was injected with 2 ml conidial suspension (1×106 conidia/ml) of isolate ZBSF005 through the silk channel. In the control groups, three ears were inoculated with an equal amount of sterile-distilled water. The inoculated ears grew under natural conditions for 30 days, the diseased kernels and ear tips were black brown and the surface covered with white or gray black mildew layer. The kernels with severe infection were wizened. But the bract could not be infected by the pathogen. Meanwhile, the control remained asymptomatic. The same fungus was successfully re-isolated from the inoculated kernels, and its identity was confirmed by morphological and molecular biology approaches, thus fulfilling Koch's postulates. E. rostratum has been reported to cause leaf spots in a wide range of hosts, such as Calathea picturata, Lagenaria siceraria, Saccharum officinarum, Ananas comosus, Hevea brasiliensis, Zea mays and so on (Chern et al. 2011; Ahmadpour et al. 2013; Choudhary et al. 2018), and it was also reported to cause root rot in Lactuca saliva (Saad et al. 2019). To our knowledge, this is the first report of E. rostratum causing maize ear rot in China. The pathogen was simultaneously inoculated to 8 maize inbred lines in Hebei province, but the disease only occurred in some varieties and the incidence area was large. Therefore, attention should be paid to the prevention and treatment of ear rot caused by this pathogen in the breeding process.

Talaromyces funiculosus, a Novel Causal Agent of Maize Ear Rot and Its Sensitivity to Fungicides.

Ear rot is one of the most prevalent and destructive diseases of maize. During field surveys, it was found that a Penicillium ear rot broke out in some areas of Shanxi, Shaanxi, Hebei, and Tianjin in China, with an incidence of 3 to 90%. A Penicillium sp. was isolated from diseased kernels covered with greyish green mold, and three isolates were identified by morphologic and molecular characteristics. The pathogenicity of isolate ZBS205 to maize ears was further determined by artificial inoculation in a field. Furthermore, the sensitivity of isolate ZBS205 against six commonly used fungicides was also evaluated. According to macro- and micromorphologic characteristics, isolate ZBS205 was generally identical to Talaromyces funiculosus (teleomorph of Penicillium funiculosum). The partial gene sequences of the nuclear ribosomal ITS1-5.8S-ITS2 (ITS) region, ß-tubulin, putative ribosome biogenesis protein (Tsr1), and the second largest subunit of the RNA polymerase II (RPB2) from isolates ZBS205, D49-1, and S73-1 showed the highest nucleotide identity to T. funiculosus strain X33, and the phylogenetic analysis conducted by the neighbor-joining method with the combined data of the four genes demonstrated that these three isolates clustered with T. funiculosus strain X33. These results suggested that the fungus isolated from diseased maize kernels was T. funiculosus. Pathogenicity testing showed that the T. funiculosus isolate ZBS205 was pathogenic to maize ears, which showed symptoms of rotted cob and deteriorated kernels. This is the first report of T. funiculosus as the definitive pathogen causing maize ear rot. The result of fungal sensitivity against fungicides showed that pyraclostrobin exhibited the highest toxicity to mycelial growth and could be used as a candidate agent for the prevention and control of T. funiculosus ear rot. The results of the present study provide a basis for understanding ear rot caused by T. funiculosus, and they should play an important role in disease management.

First Report of Root Rot Caused by Bipolaris zeicola on Maize in Hebei Province.

Maize (Zea mays L.) is one of the most important cereal crops in China, and the planting area reached 41.3 million hectares in 2019. Root rot is a widespread disease that occurs at the seedling stage of maize, resulting in leaf wilting, root rot and even plant death, and consequently yield and quality losses. During an investigation of spring maize in 2020, seedlings with wilted leaves and dark brown necrotic spots on root were observed in the fields in Kuancheng Manchu Autonomous County, Hebei Province, China. Symptomatic plants were collected for pathogen isolation and identification. The soil on roots was washed off with running water. Then, 2-3 mm necrotic root segments were sampled and surface sterilized with 75% ethanol for 2 min, rinsed three times with sterile distilled water, air-dried on sterile filter paper, and plated on potato dextrose agar (PDA). Plates were incubated at 28℃ in darkness for 3 days. A nonsporulating, dematiaceous fungus growing from root segments was transferred to fresh PDA plates. The colonies were round or irregular round, black, villiform with dense grayish white mycelia. Water agar amended with wheat straw was used for sporulation. Conidiophores were single, light brown, multiseptate, geniculate. Conidia were 38.68 x 10.69 to 71.98 x 20.57 µm, brown, oval, slightly curved, with 2 to 8 septa, and an obviously flattened hilum on the basal cell. Conidia germinated from both poles. The causal agent was identified as Bipolaris zeicola (G.L. Stout) Shoemaker (teleomorph = Cochliobolus carbonum R. R. Nelson) based on its morphological features. For molecular identification, genomic DNA was extracted from fresh mycelia cultured on PDA plates. Partial sequences of ITS-rDNA region and Brn1 reductase melanin biosynthesis gene were amplified using primers ITS1/ ITS4 (TCCGTAGGTGAACCTGCGG/ TCCTCCGCTTATTGATATGC) (White et al. 1990) and Brn01/ Brn02 (GCCAACATCGAGCAAACATGG/ GCAAGCAGCACCGTCAATACCAAT) (Shimizu et al. 1998), respectively. A DNA fragment of 532 bp was obtained from ITS-rDNA region and the sequence (GenBank Accession No. MW407046) was 100% identical to sequence of B. zeicola (GenBank Accession MH864760). The sequence of Brn1 gene was 816 bp (GenBank Accession No. MW415899) and was 99.75% identical to sequence of C. carbonum (GenBank Accession No. AB011658). The morphological and molecular evidence proved that the causal agent isolated from maize roots in Hebei province was B. zeicola. Pathogenicity assays were conducted with one week old (V1 stage) maize seedlings grown from the surface-sterilized seed of cv. Zhengdan 958. The mesocotyl and radicle of each plant (N=3) were inoculated with a 5 mm fungal disk of B. zeicola. Mock-inoculated plants (N=3) with sterile PDA disk served as the negative control. After 7 days, plants inoculated with B. zeicola were wilted with dark brown necrotic spots on mesocotyl and radicle. Meanwhile, the negative controls did not present any symptoms. Koch's postulate was proved with successful re-isolation of the same fungus from the inoculated maize plants. These results confirmed the pathogenicity of B. zeicola on maize root. B. zeicola mainly causes an important foliar disease in many regions of the world, known as Northern corn leaf spot, in addition, it can also cause ear rot and stalk rot of maize (Liu et al. 2015). To our knowledge, this is the first report of root rot caused by B. zeicola on maize in China, which extends the known agents of maize root rot. Therefore, it is necessary to explore effective seed-applied fungicides for disease control. Also, more attention should be paid to develop hybrids with resistance to this disease.

SpotSDC: Revealing the Silent Data Corruption Propagation in High-Performance Computing Systems.

The trend of rapid technology scaling is expected to make the hardware of high-performance computing (HPC) systems more susceptible to computational errors due to random bit flips. Some bit flips may cause a program to crash or have a minimal effect on the output, but others may lead to silent data corruption (SDC), i.e., undetected yet significant output errors. Classical fault injection analysis methods employ uniform sampling of random bit flips during program execution to derive a statistical resiliency profile. However, summarizing such fault injection result with sufficient detail is difficult, and understanding the behavior of the fault-corrupted program is still a challenge. In this article, we introduce SpotSDC, a visualization system to facilitate the analysis of a program's resilience to SDC. SpotSDC provides multiple perspectives at various levels of detail of the impact on the output relative to where in the source code the flipped bit occurs, which bit is flipped, and when during the execution it happens. SpotSDC also enables users to study the code protection and provide new insights to understand the behavior of a fault-injected program. Based on lessons learned, we demonstrate how what we found can improve the fault injection campaign method.

Nanomaterial Synthesis Insights from Machine Learning of Scientific Articles by Extracting, Structuring, and Visualizing Knowledge.

Nanomaterials of varying compositions and morphologies are of interest for many applications from catalysis to optics, but the synthesis of nanomaterials and their scale-up are most often time-consuming and Edisonian processes. Information gleaned from the scientific literature can help inform and accelerate nanomaterials development, but again, searching the literature and digesting the information are time-consuming manual processes for researchers. To help address these challenges, we developed scientific article-processing tools that extract and structure information from the text and figures of nanomaterials articles, thereby enabling the creation of a personalized knowledgebase for nanomaterials synthesis that can be mined to help inform further nanomaterials development. Starting with a corpus of ∼35k nanomaterials-related articles, we developed models to classify articles according to the nanomaterial composition and morphology, extract synthesis protocols from within the articles' text, and extract, normalize, and categorize chemical terms within synthesis protocols. We demonstrate the efficiency of the proposed pipeline on an expert-labeled set of nanomaterials synthesis articles, achieving 100% accuracy on composition prediction, 95% accuracy on morphology prediction, 0.99 AUC on protocol identification, and up to a 0.87 F1-score on chemical entity recognition. In addition to processing articles' text, microscopy images of nanomaterials within the articles are also automatically identified and analyzed to determine the nanomaterials' morphologies and size distributions. To enable users to easily explore the database, we developed a complementary browser-based visualization tool that provides flexibility in comparing across subsets of articles of interest. We use these tools and information to identify trends in nanomaterials synthesis, such as the correlation of certain reagents with various nanomaterial morphologies, which is useful in guiding hypotheses and reducing the potential parameter space during experimental design.

The Construction of a Hydrophilic Inorganic Layer Enables Mechanochemically Robust Super Antifouling UHMWPE Composite Membrane Surfaces.

In this study, a facile and effective method is adopted to prepare mechanochemically robust super antifouling membrane surfaces. During the process, vinyl trimethoxy silane (VTMS) was used as the reactive intermediate for coupling the hydrophilic inorganic SiO2 nanoparticle layer on to the organic ultra-high-molecular-weight polyethylene (UHMWPE) membrane surface, which created hierarchical nanostructures and lower surface energy simultaneously. The physical and chemical properties of the modified UHMWPE composite membrane surface were investigated. FTIR and XPS showed the successful chemical grafting of VTMS and SiO2 immobilization, and this modification could effectively enhance the membrane's surface hydrophilicity and filtration property with obviously decreased surface contact angle, the pure water flux and bovine serum albumin (BSA) rejection were 805 L·m-2·h-1 and 93%, respectively. The construction of the hydrophilic nano-SiO2 layer on the composite membrane surface for the improvement of membrane antifouling performance was universal, water flux recovery ratio values of BSA, humic acid (HA), and sodium alginate (SA) were all up to 90%. The aim of this paper is to provide an effective approach for the enhancement of membrane antifouling performance by the construction of a hydrophilic inorganic layer on an organic membrane surface.

Scalable Topological Data Analysis and Visualization for Evaluating Data-Driven Models in Scientific Applications.

With the rapid adoption of machine learning techniques for large-scale applications in science and engineering comes the convergence of two grand challenges in visualization. First, the utilization of black box models (e.g., deep neural networks) calls for advanced techniques in exploring and interpreting model behaviors. Second, the rapid growth in computing has produced enormous datasets that require techniques that can handle millions or more samples. Although some solutions to these interpretability challenges have been proposed, they typically do not scale beyond thousands of samples, nor do they provide the high-level intuition scientists are looking for. Here, we present the first scalable solution to explore and analyze high-dimensional functions often encountered in the scientific data analysis pipeline. By combining a new streaming neighborhood graph construction, the corresponding topology computation, and a novel data aggregation scheme, namely topology aware datacubes, we enable interactive exploration of both the topological and the geometric aspect of high-dimensional data. Following two use cases from high-energy-density (HED) physics and computational biology, we demonstrate how these capabilities have led to crucial new insights in both applications.

NLIZE: A Perturbation-Driven Visual Interrogation Tool for Analyzing and Interpreting Natural Language Inference Models.

With the recent advances in deep learning, neural network models have obtained state-of-the-art performances for many linguistic tasks in natural language processing. However, this rapid progress also brings enormous challenges. The opaque nature of a neural network model leads to hard-to-debug-systems and difficult-to-interpret mechanisms. Here, we introduce a visualization system that, through a tight yet flexible integration between visualization elements and the underlying model, allows a user to interrogate the model by perturbing the input, internal state, and prediction while observing changes in other parts of the pipeline. We use the natural language inference problem as an example to illustrate how a perturbation-driven paradigm can help domain experts assess the potential limitation of a model, probe its inner states, and interpret and form hypotheses about fundamental model mechanisms such as attention.

Visual Exploration of Semantic Relationships in Neural Word Embeddings.

Constructing distributed representations for words through neural language models and using the resulting vector spaces for analysis has become a crucial component of natural language processing (NLP). However, despite their widespread application, little is known about the structure and properties of these spaces. To gain insights into the relationship between words, the NLP community has begun to adapt high-dimensional visualization techniques. In particular, researchers commonly use t-distributed stochastic neighbor embeddings (t-SNE) and principal component analysis (PCA) to create two-dimensional embeddings for assessing the overall structure and exploring linear relationships (e.g., word analogies), respectively. Unfortunately, these techniques often produce mediocre or even misleading results and cannot address domain-specific visualization challenges that are crucial for understanding semantic relationships in word embeddings. Here, we introduce new embedding techniques for visualizing semantic and syntactic analogies, and the corresponding tests to determine whether the resulting views capture salient structures. Additionally, we introduce two novel views for a comprehensive study of analogy relationships. Finally, we augment t-SNE embeddings to convey uncertainty information in order to allow a reliable interpretation. Combined, the different views address a number of domain-specific tasks difficult to solve with existing tools.