Phy-Taylor: Partially Physics-Knowledge-Enhanced Deep Neural Networks via NN Editing.

Purely data-driven deep neural networks (DNNs) applied to physical engineering systems can infer relations that violate physics laws, thus leading to unexpected consequences. To address this challenge, we propose a physics-knowledge-enhanced DNN framework called Phy-Taylor, accelerating learning-compliant representations with physics knowledge. The Phy-Taylor framework makes two key contributions; it introduces a new architectural physics-compatible neural network (PhN) and features a novel compliance mechanism, which we call physics-guided neural network (NN) editing. The PhN aims to directly capture nonlinear physical quantities, such as kinetic energy, electrical power, and aerodynamic drag force. To do so, the PhN augments NN layers with two key components: 1) monomials of the Taylor series for capturing physical quantities and 2) a suppressor for mitigating the influence of noise. The NN editing mechanism further modifies network links and activation functions consistently with physics knowledge. As an extension, we also propose a self-correcting Phy-Taylor framework for safety-critical control of autonomous systems, which introduces two additional capabilities: 1) safety relationship learning and 2) automatic output correction when safety violations occur. Through experiments, we show that Phy-Taylor features considerably fewer parameters and a remarkably accelerated training process while offering enhanced model robustness and accuracy.

Protein-ligand binding affinity prediction with edge awareness and supervised attention.

Accurate prediction of protein-ligand binding affinity is crucial in structure-based drug design but remains some challenges even with recent advances in deep learning: (1) Existing methods neglect the edge information in protein and ligand structure data; (2) current attention mechanisms struggle to capture true binding interactions in the small dataset. Herein, we proposed SEGSA_DTA, a SuperEdge Graph convolution-based and Supervised Attention-based Drug-Target Affinity prediction method, where the super edge graph convolution can comprehensively utilize node and edge information and the multi-supervised attention module can efficiently learn the attention distribution consistent with real protein-ligand interactions. Results on the multiple datasets show that SEGSA_DTA outperforms current state-of-the-art methods. We also applied SEGSA_DTA in repurposing FDA-approved drugs to identify potential coronavirus disease 2019 (COVID-19) treatments. Besides, by using SHapley Additive exPlanations (SHAP), we found that SEGSA_DTA is interpretable and further provides a new quantitative analytical solution for structure-based lead optimization.

How childhood maltreatment links to labor values? The mediating role of moral competence and prosocial normative tendency.

BACKGROUND: Labor values are important components of the individual value system and considered to be among the most important values of an individual, especially in China. In studies of values, childhood maltreatment is considered to have an important influence on the formation of individual values. However, there is no previous research about the relationship between childhood maltreatment and labor values. The mechanism of childhood maltreatment on labor values is not clear and requires further study. METHODS: This study intended to investigate the relationship between childhood maltreatment and labor values, and further verify whether moral competence or prosocial normative tendency mediated this correlation. Therefore, 2691 participants were recruited from primary and secondary schools, who completed Labor Values Scale, Childhood Trauma Questionnaire, Moral Competence subscale and Prosocial Norms subscale. RESULTS: Results revealed the negative correlation between childhood maltreatment and labor values. Importantly, childhood maltreatment also indirectly impacted labor values through moral competence and prosocial normative tendency. It indicated that both moral competence and prosocial normative tendency played a significant mediating role in this relationship. Our findings are valuable for understanding the underlying mechanism between early trauma and values. CONCLUSIONS: Childhood maltreatment has important implications for labor values. Moral competence and prosocial normative tendency mediate between childhood maltreatment and labor values. The results remind us to pay attention to the important influence of childhood maltreatment in the cultivation of labor values, and focus on the role of moral competence and prosocial normative tendency.

Effects of Labor Values on Subjective Well-Being: The Mediating Role of Altruistic Tendencies.

This study examined the mediating role of altruistic tendency in the association between labor values and subjective well-being (SWB). About 2,691 Chinese students (1,504 males and 1,187 females) completed the labor values scale (LVS), the Positive Affect and Negative Affect Scale, the Satisfaction With Life Scale, and the altruistic tendency scale. Results demonstrated that labor values were positively associated with life satisfaction and positive affect, while negatively with negative affect. The altruistic tendency was positively correlated with labor values, and positive affect, while negatively correlated with negative affect. Furthermore, altruistic tendency served as a mediator linking labor values and positive/negative affect. These results confirmed the relationship between labor values and SWB and revealed the mechanism of altruism tendency between the two.

Novel deep learning-based transcriptome data analysis for drug-drug interaction prediction with an application in diabetes.

BACKGROUND: Drug-drug interaction (DDI) is a serious public health issue. The L1000 database of the LINCS project has collected millions of genome-wide expressions induced by 20,000 small molecular compounds on 72 cell lines. Whether this unified and comprehensive transcriptome data resource can be used to build a better DDI prediction model is still unclear. Therefore, we developed and validated a novel deep learning model for predicting DDI using 89,970 known DDIs extracted from the DrugBank database (version 5.1.4). RESULTS: The proposed model consists of a graph convolutional autoencoder network (GCAN) for embedding drug-induced transcriptome data from the L1000 database of the LINCS project; and a long short-term memory (LSTM) for DDI prediction. Comparative evaluation of various machine learning methods demonstrated the superior performance of our proposed model for DDI prediction. Many of our predicted DDIs were revealed in the latest DrugBank database (version 5.1.7). In the case study, we predicted drugs interacting with sulfonylureas to cause hypoglycemia and drugs interacting with metformin to cause lactic acidosis, and showed both to induce effects on the proteins involved in the metabolic mechanism in vivo. CONCLUSIONS: The proposed deep learning model can accelerate the discovery of new DDIs. It can support future clinical research for safer and more effective drug co-prescription.

How Labor Values Affect Mental Health: An Analysis From the Perspective of Social Support.

To explore the positive and negative effects of labor values on mental health from the aspects of life satisfaction and psychological distress, and further verify the mediating role of social support. A total of 2,691 primary and secondary school students were surveyed by Labor Values Scale, the Multidimensional Scale of Social Support, General Health Questionnaire and Satisfaction with Life Scale, and the results of which showed that as: (1) labor values can positively predict life satisfaction, while they are negatively correlated with psychological distress; (2) social support can play a mediating role between labor values and life satisfaction; and (3) social support can also play a mediating role in the relationship between labor values and psychological distress. This study revealed that the specific path and mechanism of labor values on mental health. This provided a reference for families and schools to further implement the education of labor values on primary and secondary school students and helped to promote the social construction of an education system that aimed at cultivating individual all-round development.

Factors affecting sonolytic degradation of sulfamethazine in water.

In this study, the major factors affecting sonolytic degradation of sulfamethazine (SMT), a typical pharmaceutically active compound, in water were evaluated. The factors tested included two operational parameters (i.e. initial SMT concentration and ultrasonic power), three dissolved gases (i.e. Ar, O2 and N2), five most frequently found anions in water (NO3(-),Cl(-),SO4(2-),HCO3(-)andBr(-)), ferrous ion (Fe(2+)), and four alcohols (methanol, ethanol, isopropyl alcohol, tert-butyl alcohol). Typically, the degradation rate was increased with the increasing initial SMT concentration and power. The degradation rate was accelerated in the presence of argon or oxygen, but inhibited by nitrogen. Effects of anions on the ultrasonic treatment were species-dependent. The SMT degradation rate was slightly inhibited by NO3(-),Cl(-),and,SO4(2-) but significantly improved by HCO3(-)andBr(-). The negative effects of alcohols acted as hydroxyl radicals scavengers with the following order: tert-butyl alcohol>isopropyl alcohol>ethanol>methanol. The synergetic effect of ferrous ion was mainly due to production of additional hydroxyl radicals (·OH) through Fenton chemistry. LC/MS/MS analysis indicated that the degradation of SMT by ultrasonic irradiation is mainly ascribed to ·OH oxidation. Of interest, although the SMT could be rapidly degraded by ultrasonic irradiation, the degradation products were rarely mineralized. For example, ~100% of 180 µM SMT was decomposed, but only 8.31% TOC was reduced, within 2h at an irradiation frequency of 800 kHz and a power of 100 W. However, the products became much biodegradable (BOD5/COD was increased from 0.04 to 0.45). Therefore, an aerobic biological treatment may be an appropriate post-treatment to further decompose the SMT degradation products.

Photochemical degradation of ciprofloxacin in UV and UV/H2O2 process: kinetics, parameters, and products.

Photochemical degradation of fluoroquinolone ciprofloxacin (CIP) in water by UV and UV/H2O2 were investigated. The degradation rate of CIP was affected by pH, H2O2 dosage, as well as the presence of other inorganic components. The optimized pH value and H2O2 concentration were 7.0 and 5 mM. Carbonate and nitrate both impeded CIP degradation. According to liquid chromatography-tandem mass spectrometry analysis, four and 16 products were identified in UV and UV/H2O2 system, respectively. Proposed degradation pathways suggest that reactions including the piperazinyl substituent, quinolone moiety, and cyclopropyl group lead to the photochemical degradation of CIP. Toxicity of products assessed by Vibrio qinghaiensis demonstrated that UV/H2O2 process was more capable on controlling the toxicity of intermediates in CIP degradation than UV process.