Non-hypothetical projection pursuit regression for the prediction of hydration heat of Portland-cement-based cementitious system.

In this study, the non-hypothetical projection pursuit regression (NH-PPR) is proposed. The proposed NH-PPR model can predict the hydration heat based on the four cement phases, FA, SL, cement fineness and hydration time. The NH-PPR model is proposed by using the multiple layer iteration method and the non-hypothetical and non-parametric ridge functions to enhance accuracy and solve the problems caused by the parameter selection and the subjective hypothesis. The modeling data set is applied to train model, the testing data set is regressed and fitted into the model, and then the obtained results are compared with the BP model. To further validate the proposed model, another published data set is used to obtain a higher degree of confidence in the prediction. It is shown that the proposed model obtains the better accuracy, stability and versatility, and avoids the parameter selection and subjective hypothesis.

Taking Olefin Metathesis to the Limit: Stereocontrolled Synthesis of Trisubstituted Alkenes.

ConspectusIn this Account, we share the story of the development of catalytic olefin metathesis processes that efficiently deliver a wide range of acyclic and macrocyclic E- or Z-trisubstituted alkenes. The tale starts with us unveiling, in collaboration with Richard Schrock and his team, the blueprint in 2009 for the design of kinetically controlled Z-selective olefin metathesis reactions. This paved the way for the development of Mo-, W-, and Ru-based catalysts and strategies for synthesizing countless linear and macrocyclic Z-olefins. Six years later, in 2015, we found that abundant Z-alkene feedstocks, such as oleic acid, can be directly transformed to high-value and more difficult-to-access alkenes through a cross-metathesis reaction promoted by a Ru-catechothiolate complex that we had developed; the approach, later coined stereoretentive olefin metathesis, was extended to the synthesis of E-alkenes.It was all about disubstituted alkenes until when in 2017 we addressed the challenge of accessing stereodefined Z- and E-trisubstituted alkenes, key to medicine and materials research. These transformations can be most effectively catalyzed by Mo monoaryloxides pyrrolide (MAP) and chloride (MAC) complexes. A central aspect of the advance is the merging of olefin metathesis, which delivered trisubstituted alkenyl fluorides, chlorides, and bromides with cross-coupling. These catalytic and stereoretentive transformations can be used in various combinations, thereby enabling access to assorted Z- or E-trisubstituted alkene. Ensuing work led to the emergence of other transformations involving substrates that can be purchased with high stereoisomeric purity, notably E- and Z-trihalo alkenes. Trisubstituted olefins, Z or E, bearing a chemoselectively and stereoretentively alterable F,Cl-terminus or B(pin),Cl-terminus may, thus, be easily and reliably synthesized. Methods for stereoretentive preparation of other alkenyl bromide regioisomers and α,ß-unsaturated carboxylic and thiol esters, nitriles, and acid fluorides followed, along with stereoretentive ring-closing metathesis reactions that afford macrocyclic trisubstituted olefins. Z- and E-Macrocyclic trisubstituted olefins, including those that contain little or no entropic support for cyclization (minimally functionalized) and/or are disfavored under substrate-controlled conditions, can now be synthesized. The utility of this latest chapter in the history of olefin metathesis has been highlighted by applications to the synthesis of several biologically active compounds, as well as their analogues, such as those marked by one or more site-specifically incorporated fluorine atoms or more active but higher energy and otherwise unobtainable conformers.The investigations discussed here, which represent every stereoretentive method that has been reported thus far for preparing a trisubstituted olefin, underscore the inimitable power of Mo-based catalysts. This Account also showcases a variety of mechanistic attributes─some for the first time, and each instrumental in solving a problem. Extensive knowledge of mechanistic nuances will be needed if we are to address successfully the next challenging problem, namely, the development of catalysts and strategies that may be used to synthesize a wide range of tetrasubstituted alkenes, especially those that are readily modifiable, with high stereoisomeric purity.

Balancing Biases and Preserving Privacy on Balanced Faces in the Wild.

There are demographic biases present in current facial recognition (FR) models. To measure these biases across different ethnic and gender subgroups, we introduce our Balanced Faces in the Wild (BFW) dataset. This dataset allows for the characterization of FR performance per subgroup. We found that relying on a single score threshold to differentiate between genuine and imposters sample pairs leads to suboptimal results. Additionally, performance within subgroups often varies significantly from the global average. Therefore, specific error rates only hold for populations that match the validation data. To mitigate imbalanced performances, we propose a novel domain adaptation learning scheme that uses facial features extracted from state-of-the-art neural networks. This scheme boosts the average performance and preserves identity information while removing demographic knowledge. Removing demographic knowledge prevents potential biases from affecting decision-making and protects privacy by eliminating demographic information. We explore the proposed method and demonstrate that subgroup classifiers can no longer learn from features projected using our domain adaptation scheme. For access to the source code and data, please visit https://github.com/visionjo/facerec-bias-bfw.

An efficient approach to synthesize sterile allopolyploids through the combined reproduction mode of ameiotic oogenesis and sperm-egg fusion in the polyploid Carassius complex.

The association between polyploidy and reproduction transition, which is an intriguing issue in evolutionary genetics, can also be exploited as an approach for genetic improvement in agriculture. Recently, we generated novel amphitriploids (NA3n) by integrating the genomes of the gynogenetic Carassius gibelio and sexual C. auratus, and found gynogenesis was recovered in most NA3n females (NA3n♀I). Here, we discovered a unique reproduction mode, termed ameio-fusiongenesis, which combines the abilities of both ameiotic oogenesis and sperm-egg fusion, in a few NA3n females (NA3n♀II). These females inherited ameiotic oogenesis to produce unreduced eggs from gynogenetic C. gibelio and sperm-egg fusion from sexual C. auratus. Subsequently, we utilized this unique reproduction mode to generate a group of synthetic alloheptaploids by crossing NA3n♀II with Megalobrama amblycephala. They contained all chromosomes of maternal NA3n♀II and a chromosomal set of paternal M. amblycephala. Intergenomic chromosome translocations between NA3n♀II and M. amblycephala were also observed in a few somatic cells. Primary oocytes of the alloheptaploid underwent severe apoptosis owing to incomplete double-strand break repair at prophase I. Although spermatocytes displayed similar chromosome behavior at prophase I, they underwent apoptosis due to chromosome separation failure at metaphase I. Therefore, the alloheptaploid females and males were all sterile. Finally, we established a sustainable clone for the large-scale production of NA3n♀II and developed an efficient approach to synthesize diverse allopolyploids containing genomes of different cyprinid species. These findings not only broaden our understanding of reproduction transition but also offer a practical strategy for polyploidy breeding and heterosis fixing.

Global Aligned Structured Sparsity Learning for Efficient Image Super-Resolution.

Efficient image super-resolution (SR) has witnessed rapid progress thanks to novel lightweight architectures or model compression techniques (e.g., neural architecture search and knowledge distillation). Nevertheless, these methods consume considerable resources or/and neglect to squeeze out the network redundancy at a more fine-grained convolution filter level. Network pruning is a promising alternative to overcome these shortcomings. However, structured pruning is known to be tricky when applied to SR networks because the extensive residual blocks demand the pruned indices of different layers to be the same. Besides, the principled determination of proper layerwise sparsities remains challenging too. In this article, we present Global Aligned Structured Sparsity Learning (GASSL) to resolve these problems. GASSL has two major components: Hessian-Aided Regularization (HAIR) and Aligned Structured Sparsity Learning (ASSL). HAIR is a regularization-based sparsity auto-selection algorithm with Hessian considered implicitly. A proven proposition is introduced to justify its design. ASSL is for physically pruning SR networks. Particularly, a new penalty term Sparsity Structure Alignment (SSA) is proposed to align the pruned indices of different layers. With GASSL, we design two new efficient single image SR networks of different architecture genres, pushing the efficiency envelope of SR models one step forward. Extensive results demonstrate the merits of GASSL over other recent counterparts.

Catalytic Cross-Metathesis Reactions That Afford E- and Z-Trisubstituted Alkenyl Bromides: Scope, Applications, and Mechanistic Insights.

Stereochemically defined trisubstituted alkenes with a bromide and a methyl group at a terminus can be readily and stereoretentively derivatized through catalytic cross-coupling, affording unsaturated fragments found in many bioactive natural products. A direct method for generating such entities would be by stereocontrolled catalytic cross-metathesis (CM). Such methods are scarce however. Here, we present a stereoretentive strategy for CM between tri-, Z- or E-di, or monosubstituted olefins and Z- or E-2-bromo-2-butene, affording an assortment of E- or Z-trisubstituted alkenyl bromides. The majority of the transformations were catalyzed by two Mo monoaryloxide pyrrolide (MAP) complexes, one purchasable and the other accessible by well-established protocols. Substrates, such as feedstock trisubstituted olefins, can be purchased; the alkenyl bromide reagents are commercially available or can be prepared in two steps in a multigram scale. The catalytic process can be used to generate products that contain polar moieties, such as an amine or an alcohol, or sterically hindered alkenes that are α- or ß-branched. The utility of the approach is highlighted by a brief and stereocontrolled synthesis of an unsaturated fragment of phomactin A and a concise total synthesis of ambrein. An unexpected outcome of these investigations was the discovery of a new role for the presence of a small-molecule alkene in an olefin metathesis reaction. DFT studies indicate that this additive swiftly reacts with a short-lived Mo alkylidene and probably helps circumvent the formation of catalytically inactive square pyramidal metallacyclobutanes, enhancing the efficiency of a transformation.

Z-Trisubstituted α,ß-Unsaturated Esters and Acid Fluorides through Stereocontrolled Catalytic Cross-Metathesis.

Catalytic cross-metathesis (CM) reactions that can generate trisubstituted alkenes in high stereoisomeric purity are important but remain limited in scope. Here, CM reactions are introduced that generate Z-trisubstituted α-methyl, α,ß-unsaturated, alkyl and aryl esters, thiol esters, and acid fluorides. Transformations are promoted by a Mo bis-aryloxide, a monoaryloxide pyrrolide, or a monoaryloxide chloride complex; air-stable and commercially available paraffin tablets containing a Mo complex may also be used. Alkyl, aryl, and silyl carboxylic esters as well as thiol esters and acid fluoride reagents are either purchasable or can be prepared in one step. Products were obtained in 55-95% yield and in 88:12->98:2 Z/E ratio (typically >95:5). The applicability of the approach is highlighted by a two-step conversion of citronellol to an isomintlactone precursor (1.7 g, 73% yield, and 97:3 Z/E) and a single-step transformation of lanosterol acetate to 3-epi-anwuweizic acid (72% yield and 94:6 Z/E). Included are the outcomes of DFT studies, regarding several initially puzzling catalyst activity trends, providing the following information: (1) it is key that a disubstituted Mo alkylidene, generated by a competing homo-metathesis (HM) pathway, can re-enter the productive CM cycle. (2) Whereas in a CM cycle the formation of a molybdacyclobutane is likely turnover-limiting, the collapse of related metallacycles in a HM cycle is probably rate-determining. It is therefore the relative energy barrier required for these steps that determines whether CM or HM is dominant with a particular complex.

Semi-Supervised Domain Adaptive Structure Learning.

Semi-supervised domain adaptation (SSDA) is quite a challenging problem requiring methods to overcome both 1) overfitting towards poorly annotated data and 2) distribution shift across domains. Unfortunately, a simple combination of domain adaptation (DA) and semi-supervised learning (SSL) methods often fail to address such two objects because of training data bias towards labeled samples. In this paper, we introduce an adaptive structure learning method to regularize the cooperation of SSL and DA. Inspired by the multi-views learning, our proposed framework is composed of a shared feature encoder network and two classifier networks, trained for contradictory purposes. Among them, one of the classifiers is applied to group target features to improve intra-class density, enlarging the gap of categorical clusters for robust representation learning. Meanwhile, the other classifier, serviced as a regularizer, attempts to scatter the source features to enhance the smoothness of the decision boundary. The iterations of target clustering and source expansion make the target features being well-enclosed inside the dilated boundary of the corresponding source points. For the joint address of cross-domain features alignment and partially labeled data learning, we apply the maximum mean discrepancy (MMD) distance minimization and self-training (ST) to project the contradictory structures into a shared view to make the reliable final decision. The experimental results over the standard SSDA benchmarks, including DomainNet and Office-home, demonstrate both the accuracy and robustness of our method over the state-of-the-art approaches.

Self-directed online machine learning for topology optimization.

Topology optimization by optimally distributing materials in a given domain requires non-gradient optimizers to solve highly complicated problems. However, with hundreds of design variables or more involved, solving such problems would require millions of Finite Element Method (FEM) calculations whose computational cost is huge and impractical. Here we report Self-directed Online Learning Optimization (SOLO) which integrates Deep Neural Network (DNN) with FEM calculations. A DNN learns and substitutes the objective as a function of design variables. A small number of training data is generated dynamically based on the DNN's prediction of the optimum. The DNN adapts to the new training data and gives better prediction in the region of interest until convergence. The optimum predicted by the DNN is proved to converge to the true global optimum through iterations. Our algorithm was tested by four types of problems including compliance minimization, fluid-structure optimization, heat transfer enhancement and truss optimization. It reduced the computational time by 2 ~ 5 orders of magnitude compared with directly using heuristic methods, and outperformed all state-of-the-art algorithms tested in our experiments. This approach enables solving large multi-dimensional optimization problems.

Semi-Supervised Dual Relation Learning for Multi-Label Classification.

In a real-world scenario, an object could contain multiple tags instead of a single categorical label. To this end, multi-label learning (MLL) emerged. In MLL, the feature distributions are long-tailed and the complex semantic label relation and the long-tailed training samples are the main challenges. Semi-supervised learning is a potential solution. While, existing methods are mainly designed for single class scenario while ignoring the latent label relations. In addition, they cannot well handle the distribution shift commonly existing across source and target domains. To this end, a Semi-supervised Dual Relation Learning (SDRL) framework for multi-label classification is proposed. SDRL utilizes a few labeled samples as well as large scale unlabeled samples in the training stage. It jointly explores the inter-instance feature-level relation and the intra-instance label-level relation even from the unlabeled samples. In our model, a dual-classifier structure is deployed to obtain domain invariant representations. The prediction results from the classifiers are further compared and the most confident predictions are extracted as pseudo labels. A trainable label relation tensor is designed to explicitly explore the pairwise latent label relations and refine the predicted labels. SDRL is able to effectively and efficiently explore the feature-label relation as well as the label-label relation knowledge without any extra semantic knowledge. We evaluated SDRL in general and zero-shot multi-label classification tasks and we concluded that SDRL is superior to other SOTA baselines. Furthermore, extensive ablation studies have been done which reveal the effectiveness of each component in our framework.

Impact of Ethylene on Efficiency and Stereocontrol in Olefin Metathesis: When to Add It, When to Remove It, and When to Avoid It.

Ethylene is the byproduct of olefin metathesis reactions that involve one or more terminal alkenes. Its volatility is one reason why many cross-metathesis or ring-closing metathesis processes, which are reversible transformations, are efficient. However, because ethylene can be converted to a methylidene complex, which is a highly reactive but relatively unstable species, its concentration can impact olefin metathesis in other ways. In some cases, introducing excess ethylene can increase reaction rate owing to faster catalyst initiation. Ethylene and a derived methylidene complex can also advantageously inhibit substrate or product homocoupling, and/or divert a less selective pathway. In other instances, a methylidene's low stability and high activity may lead to erosion of efficiency and/or kinetic selectivity, making it preferable that ethylene is removed while being generated. If methylidene decomposition is so fast that there is little or no product formation, it is best that ethylene and methylidene complex formation is avoided altogether. This is accomplished by the use of di- or trisubstituted alkenes in stereoretentive processes, which includes adopting methylene capping strategy. Here, we analyze the different scenarios through which ethylene and the involvement of methylidene complexes can be manipulated and managed so that an olefin metathesis reaction may occur more efficiently and/or more stereoselectively.

[Value of ginsenoside Rb1 in alleviating coronary artery lesion in a mouse model of Kawasaki disease].

OBJECTIVE: To study the effect and related signaling pathways of ginsenoside Rb1 in the treatment of coronary artery lesion (CAL) in a mouse model of Kawasaki disease (KD). METHODS: BALB/c mice were randomly divided into a control group, a model group, an aspirin group, a low-dose ginsenoside Rb1 group (50 mg/kg), and a high-dose ginsenoside Rb1 group (100 mg/kg), with 12 mice in each group. All mice except those in the control group were given intermittent intraperitoneal injection of 10% bovine serum albumin to establish a mouse model of KD. The mice in the aspirin group, the low-dose ginsenoside Rb1 group, and the high-dose ginsenoside Rb1 group were given the corresponding drug by gavage for 20 days after modeling. Hematoxylin and eosin staining was used to observe the pathological changes of coronary artery tissue. ELISA was used to measure the levels of the inflammatory cytokines tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), and interleukin-1ß (IL-1ß) in serum and coronary artery tissue. Western blot was used to measure the relative expression levels of proteins involved in the regulation of the AMPK/mTOR autophagy signaling pathway and the PI3K/Akt oxidative stress signaling pathway in coronary artery tissue. RESULTS: The observation of pathological sections showed that compared with the model group, the high-dose ginsenoside Rb1 group had significant improvement in the symptoms of vascular wall thickening, intimal edema, fiber rupture, and inflammatory infiltration of endothelial cells. Compared with the control group, the model and low-dose ginsenoside Rb1 groups had significant increases in the levels of TNF-α, IL-6, and IL-1ß in serum and coronary artery tissue (P<0.05); the model group had significant increases in the expression levels of P-AMPK/AMPK, P-mTOR/mTOR, and P-P70S6/P70S6 in coronary artery tissue (P<0.05) and significant reductions in the expression levels of P-PI3K/PI3K, P-AKT/AKT, and P-GSK-3ß/GSK-3ß in coronary artery tissue (P<0.05). Compared with the model group, the aspirin group and the high-dose ginsenoside Rb1 group had significant reductions in the levels of TNF-α, IL-6, and IL-1ß (P<0.05); the low- and high-dose ginsenoside Rb1 groups had significant reductions in the expression levels of P-AMPK/AMPK, P-mTOR/mTOR, and P-P70S6/P70S6 (P<0.05) in a dose-dependent manner between the two groups (P<0.05); the low-dose ginsenoside Rb1 group had no significant change in the expression level of P-PI3K/PI3K (P>0.05) and had significant increases in the expression levels of P-AKT/AKT and P-GSK-3ß/GSK-3ß (P<0.05), while the high-dose ginsenoside Rb1 group had significant increases in the relative protein expression levels of the above three proteins (P<0.05). Compared with the low-dose ginsenoside Rb1 group, the aspirin group and the high-dose ginsenoside Rb1 group had significant reductions in the levels of TNF-α, IL-6, and IL-1ß (P<0.05); the high-dose ginsenoside Rb1 group had significant increases in the expression levels of P-PI3K/PI3K and P-AKT/AKT (P<0.05). CONCLUSIONS: Ginsenoside Rb1 can effectively alleviate CAL in a mouse model of KD in a dose-dependent manner, possibly by regulating the AMPK/mTOR/P70S6 autophagy signaling pathway to inhibit CAL inflammation and regulating the PI3K/AKT/GSK-3ß oxidative stress signaling pathway to exert a biological activity of protection against coronary artery endothelial cell injury.

Comparing sexual behaviours and knowledge between domestic students and Chinese international students in Australia: findings from two cross-sectional studies.

Few studies investigate sexual health among Chinese international students in Australia. We recruited domestic (n = 623) and Chinese international (n = 500) students for separate online surveys on sexual behaviours and knowledge. Samples were compared using Chi square, Fisher's exact and equality of medians tests. Domestic students were more likely than international students to have ever touched a partner's genitals (81% vs. 53%, p < 0.01), had oral sex (76% vs. 44%, p < 0.01), vaginal intercourse (67% vs. 41%, p < 0.01) and anal intercourse (31% vs. 6%, p < 0.01). Domestic students were younger when they first touched a partner's genitals (16 vs. 18 years, p < 0.01), had oral sex (17 vs. 18 years, p < 0.01) and vaginal intercourse (17 vs. 18 years, p < 0.01). Domestic students were less likely than Chinese international students to report only one lifetime partner for touching genitals (22% vs. 50%, p < 0.01), oral sex (25% vs. 55%, p < 0.01), vaginal intercourse (30% vs. 58%, p < 0.01) and anal intercourse (54% vs. 88%, p < 0.01). Domestic students were more likely than Chinese international students to use the oral contraceptive pill (48% vs. 16%, p < 0.01) and long-acting reversible contraceptives (19% vs. 1%, p < 0.01). Domestic students scored higher than international students on a contraception and chlamydia quiz (4/5 vs. 2/5, p < 0.01). Domestic and Chinese international students differed in sexual behaviours and knowledge highlighting the need for relevant sexual health promotion for both groups.

Sesquiterpenes with Phytopathogenic Fungi Inhibitory Activities from Fungus Trichoderma virens from Litchi chinensis Sonn.

A new monosesquiterpene diacetylgliocladic acid (1), a new dimeric sesquiterpene divirensol H (9), and two exceptionally novel trimeric sesquiterpene trivirensols A and B (11 and 12), together with another eight known congeners, were purified from an endophytic fungus Trichoderma virens FY06, derived from Litchi chinensis Sonn. whose fruit is a delicious and popular food. All of them were identified by comprehensive spectroscopic analysis, combined with biosynthetic considerations. Trivirensols A and B are unprecedented trimers of which three subunits are connected by two ester bonds of the sesquiterpene class. Relative to the positive control triadimefon, all the tested metabolites showed strong inhibitory activities against at least one phytopathogenic fungus among Penicillium italicum, Fusarium oxysporum, Fusarium graminearum, Colletotrichum musae, and Colletotrictum gloeosporioides. Notably, as metabolites of the endophytic fungus from L. chinensis, they all presented strong antifungal activities against C. gloeosporioides which causes anthracnose in L. chinensis.

Endobronchial ultrasound: Echoing in the field of pediatrics.

Endobronchial ultrasound (EBUS) is a useful technique for the diagnosis and staging of the lung and mediastinal lesions, which is crucial for selecting treatment protocol. Under EBUS guidance, transbronchial needle aspiration (TBNA) is widely applied for obtaining specimens for histological, cytological, and molecular evaluation. Recently, the EBUS scope designed for adults has been used in large pediatric candidates. The presence of lung masses and mediastinal lymphadenopathy in the pediatric population presents a diagnostic challenge; however, EBUS is a promising tool for pediatricians to address these challenging issues. In some centers, the adult EBUS echobronchoscope is applied in transesophageal procedures for pediatric patients. EBUS-guided TBNA can also be used to perform minimally invasive interventional therapy, such as fiducial marker placement for assisting precision radiation, brachytherapy, and radiofrequency ablation therapy. With the development of EBUS equipment designed specifically for children, pediatric EBUS will play an increasingly important role.

Permeable reactive barrier of coarse sand-supported zero valent iron for the removal of 2,4-dichlorophenol in groundwater.

In this study, coarse sand-supported zero valent iron (ZVI) composite was synthesized by adding sodium alginate to immobilize. Composite was detected by scanning electron microscope (SEM), X-ray diffraction (XRD), and X-ray fluorescence (XRF). SEM results showed that composite had core-shell structure and a wide porous distribution pattern. The synthesized composite was used for degradation of 2,4-dichlorophenol (2,4-DCP) contamination in groundwater. Experimental results demonstrated that degradation mechanism of 2,4-DCP using coarse sand-supported ZVI included adsorption, desorption, and dechlorination. 2,4-DCP adsorption was described as pseudo-second-order kinetic model. It was concluded that dechlorination was the key reaction pathway, ZVI and hydrogen are prime reductants in dechlorination of 2,4-DCP using ZVI.

Yidiyin, a Chinese herbal decoction, improves erectile dysfunction in diabetic patients and rats through the NO-cGMP pathway.

The nitric-oxide (NO)-cyclic-guanosine-monophosphate (cGMP) pathway plays a key role in penile erection. Erectile dysfunction (ED) is a complication in male diabetic patients that impacts their quality of 1ife. Recently, Yidiyin, a Chinese herbal decoction, is used to treat diabetic ED, but convincing evidence is lacking, and the potential mechanisms remain uncertain. In the study, diabetic ED patients had low scores on international index of erectile function-5 (IIEF-5), and administration of Yidiyin and hypoglycemic drugs for 16 weeks ameliorated patients' scores on IIEF-5 more than the hypoglycemic drug alone. Moreover, streptozotocin-induced diabetes severely impaired rats' erectile function and the activity of the NO-cGMP pathway in the corpora cavernosum, and treatment with Yidiyin for 4 weeks obviously increased the rats' erectile function, remarkably enhanced the activity of nitric oxide synthase (NOS), and elevated the contents of NO and cGMP. Our findings indicate that Yidiyin improves diabetic ED probably by enhancing the NO-cGMP pathway.