Detection and recognition of metal surface corrosion based on CBG-YOLOv5s.

The automatic detection of the degree of surface corrosion on metal structures is of significant importance for assessing structural damage and safety. To effectively identify the corrosion status on the surface of coastal metal facilities, this study proposed a CBG-YOLOv5s model for metal surface corrosion detection, based on the YOLOv5s model. Firstly, we integrated the Convolutional Block Attention Module (CBAM) into the C3 module and developed the C3CBAM module. This module effectively enhanced the channel and spatial attention capabilities of the feature map, thereby improving the feature representation. Second, we introduced a multi-scale feature fusion concept in the feature fusion part of the model and added a small target detection layer to improve small target detection. Finally, we designed a lighter C3Ghost module, which reduced the number of parameters and the computational load of the model, thereby improving the running speed of the model. In addition, to verify the effectiveness of our method, we constructed a dataset containing 6000 typical images of metal surface corrosion and conducted extensive experiments on this dataset. The results showed that compared to the YOLOv5s model and several other commonly used object detection models, our method achieved superior performance in terms of detection accuracy and speed.

HDGFI: Hierarchical Dual-Level Graph Feature Interaction Model for Personalized Recommendation.

Under the background of information overload, the recommendation system has attracted wide attention as one of the most important means for this problem. Feature interaction considers not only the impact of each feature but also the combination of two or more features, which has become an important research field in recommendation systems. There are two essential problems in current feature interaction research. One is that not all feature interactions can generate positive gains, and some may lead to an increase in noise. The other is that the process of feature interactions is implicit and uninterpretable. In this paper, a Hierarchical Dual-level Graph Feature Interaction (HDGFI) model is proposed to solve these problems in the recommendation system. The model regards features as nodes and edges as interactions between features in the graph structure. Interaction noise is filtered by beneficial interaction selection based on a hierarchical edge selection module. At the same time, the importance of interaction between nodes is modeled in two perspectives in order to learn the representation of feature nodes at a finer granularity. Experimental results show that the proposed HDGFI model has higher accuracy than the existing models.

A Dual Adaptive Interaction Click-Through Rate Prediction Based on Attention Logarithmic Interaction Network.

Click-through rate (CTR) prediction is crucial for computing advertisement and recommender systems. The key challenge of CTR prediction is to accurately capture user interests and deliver suitable advertisements to the right people. However, there are an immense number of features in CTR prediction datasets, which hardly fit when only using an individual feature. To solve this problem, feature interaction that combines several features via an operation is introduced to enhance prediction performance. Many factorizations machine-based models and deep learning methods have been proposed to capture feature interaction for CTR prediction. They follow an enumeration-filter pattern that could not determine the appropriate order of feature interaction and useful feature interaction. The attention logarithmic network (ALN) is presented in this paper, which uses logarithmic neural networks (LNN) to model feature interactions, and the squeeze excitation (SE) mechanism to adaptively model the importance of higher-order feature interactions. At first, the embedding vector of the input was absolutized and a very small positive number was added to the zeros of the embedding vector, which made the LNN input positive. Then, the adaptive-order feature interactions were learned by logarithmic transformation and exponential transformation in the LNN. Finally, SE was applied to model the importance of high-order feature interactions adaptively for enhancing CTR performance. Based on this, the attention logarithmic interaction network (ALIN) was proposed for the effectiveness and accuracy of CTR, which integrated Newton's identity into ALN. ALIN supplements the loss of information, which is caused by the operation becoming positive and by adding a small positive value to the embedding vector. Experiments are conducted on two datasets, and the results prove that ALIN is efficient and effective.

Maricaulis alexandrii sp. nov., a novel active bioflocculants-bearing and dimorphic prosthecate bacterium isolated from marine phycosphere.

An aerobic, Gram-stain-negative, straight or curved rods and dimorphic prosthecate bacterium designated as strain LZ-16-1T was isolated from phycosphere microbiota of routinely laboratory-cultured and highly-toxic marine dinoflagellate Alexandrium catenella LZT09. Strain LZ-16-1T produces active bioflocculanting exopolysaccharides (EPS). Cells were dimorphic with non-motile prostheca, or non-stalked and motile by a single polar flagellum. Growth occurred at 10-40 °C, pH 5-9 and 1-8% (w/v) NaCl, with optimum growth at 25 °C, pH 7-8 in the presence of 2-4% (w/v) NaCl. Phylogenetic analysis based on 16S rRNA gene sequences indicated that strain LZ-16-1T was affiliated to the genus Maricaulis, and closely related to M. parjimensis MCS 25T (99.5%) and M. virginensis VC-5T (99.0%). However, based on genome sequencing and phylogenomic calculations, the average nucleotide identity (ANI) and digtal DNA-DNA genome hybridization (dDDH) values between strains LZ-16-1T and its closest relative, M. parjimensis MCS 25T were only 85.0 and 20.9%, respectively. The dominant fatty acids of strain LZ-16-1T were summed feature 8, C16:0, C17:0, C18:0, C18:1 ω9c and summed feature 9. Major polar lipids were sulfoquinovosyl diacylglycerol, six glycolipids, one unidentified phospholipid and one unidentified polar lipid. The predominant isoprenoid quinone was Q-10. The DNA G + C content calculated from the genome was 63.6 mol%. Physiological and chemotaxonomic characterizations further confirmed the distinctiveness of strain LZ-16-1T from other Maricaulis members. Thus, strain LZ-16-1T represents a novel species of the genus Maricaulis, for which the name Maricaulis alexandrii sp. nov. (type strain LZ-16-1T = KCTC 72194T = CCTCC AB 2019006T) is proposed.

Sphingopyxis microcysteis sp. nov., a novel bioactive exopolysaccharides-bearing Sphingomonadaceae isolated from the Microcystis phycosphere.

During the study into the microbial biodiversity and bioactivity of the Microcystis phycosphere, a new yellow-pigmented, non-motile, rod-shaped bacterium containing polyhydroxybutyrate granules designated as strain Z10-6T was isolated from highly-toxic Microcystis aeruginosa Kützing M.TN-2. The new isolate produces active bioflocculating exopolysaccharides. Phylogenetic analysis based on 16S rRNA gene sequences indicated strain Z10-6T belongs to the genus Sphingopyxis with highest similarity to Sphingopyxis solisilvae R366T (98.86%), and the similarity to other Sphingopyxis members was less than 98.65%. However, both low values obtained by phylogenomic calculation of average nucleotide identity (ANI, 85.5%) and digital DNA-DNA hybridization (dDDH, 29.8%) separated the new species from its closest relative. The main polar lipids were sphingoglycolipid, phosphatidylethanolamine, phosphatidylglycerol, diphosphatidylglycerol, one unidentified glycolipid and one unidentified aminophospholipid. The predominant fatty acids were summed feature 8, C17:1ω6c, summed feature 3, C16:0, C18:1ω7c 11-methyl and C14:0 2-OH. The respiratory quinone was ubiqunone-10, with spermidine as the major polyamine. The genomic DNA G + C content was 64.8 mol%. Several biosynthesis pathways encoding for potential new bacterial bioactive metabolites were found in the genome of strain Z10-6T. The polyphasic analyses clearly distinguished strain Z10-6T from its closest phylogenetic neighbors. Thus, it represents a novel species of the genus Sphingopyxis, for which the name Sphingopyxis microcysteis sp. nov. is proposed. The type strain is Z10-6T (= CCTCC AB2017276T = KCTC 62492T).

Conditional GWAS revealing genetic impacts of lifestyle behaviors on low-density lipoprotein (LDL).

BACKGROUND: Accumulation of LDL cholesterol (LDL-c) within artery walls is strongly associated with the initiation and progression of atherosclerosis development. This complex trait is affected by multifactor involving polygenes, environments, and their interactions. Uncovering genetic architecture of LDL may help to increase the understanding of the genetic mechanism of cardiovascular diseases. METHODS: We used a genetic model to analyze genetic effects including additive, dominance, epistasis, and ethnic interactions for data from the Multi-Ethnic Study of Atherosclerosis (MESA). Three lifestyle behaviors (reading, intentional exercising, smoking) were used as cofactor in conditional models. RESULTS: We identified 156 genetic effects of 10 quantitative trait SNPs (QTSs) in base model and three conditional models. The total estimated heritability of these genetic effects was approximately 72.88% in the base model. Five genes (CELSR2, MARK2, ADAMTS12, PFDN4, and MAGI2) have biological functions related to LDL. CONCLUSIONS: Compared with the based model LDL, the results in three conditional models revealed that intentional exercising and smoking could have impacts for causing and suppressing some of genetic effects and influence the levels of LDL. Furthermore, these two lifestyles could have different genetic effects for each ethnic group on a specific QTS. As most of the heritability in based model LDL and conditional model LDL|Smk was contributed from epistasis effects, our result indicated that epistasis effects played important roles in determining LDL levels. Our study provided useful insight into the biological mechanisms underlying regulation of LDL and might help in the discovery of novel therapeutic targets for cardiovascular disease.

Identifying cancer-related microRNAs based on subpathways.

MicroRNAs (miRNAs) are a class of small endogenous non-coding genes that play important roles in post-transcriptional regulation as well as other important biological processes. Accumulating evidence indicated that miRNAs were extensively involved in the pathology of cancer. However, determining which miRNAs are related to a specific cancer is problematic because one miRNA may target multiple genes and one gene may be targeted by multiple miRNAs. The authors proposed a new approach, named miR_SubPath, to identify cancer-associated miRNAs by three steps. The targeted genes were determined based on differentially expressed genes in significant dysfunctional subpathways. Then the candidate miRNAs were determined according to miRNA-genes associations. Finally, these candidate miRNAs were ranked based on their relations with some seed miRNAs in a functional similarity network. Results on real-world datasets showed that the proposed miR_SubPath method was more robust and could identify more cancer-related miRNAs than a prior approach, miR_Path, miR_Clust and Zhang's method.

In Situ Scanning Electron Microscope (SEM) Observations of Damage and Crack Growth of Shale.

To better understand the formation and evolution of hierarchical crack networks in shales, observations of microscopic damage, and crack growth were conducted using an in situ tensile apparatus inside a scanning electron microscope. An arched specimen with an artificial notch incised into the curved edge was shown to afford effective observation of the damage and crack growth process that occurs during the brittle fracturing of shale. Because this arched specimen design can induce a squeezing effect, reducing the tensile stress concentration at the crack tip, and preventing the brittle shale from unstable fracturing to some extent. Both induced and natural pores and cracks were observed at different scales around the main crack path or on fractured surfaces. Observations indicate that the crack initiation zone develops around the crack tip where tensile stresses are concentrated and micro/nanoscale cracks nucleate. Crack advancement generally occurs by the continuous generation and coalescence of damage zones having intermittent en echelon microscopic cracks located ahead of the crack tips. Mineral anisotropy and pressure build-up around crack tips causes crack kinking, deflection, and branching. Crack growth is often accompanied by the cessation or closure of former branch cracks due to elastic recovery and induced compressive stress. The branching and interactions of cracks form a three-dimensional hierarchical network that includes induced branch cracks having similar paths, as well as natural structures such as nanopores, bedding planes, and microscopic cracks.

Br-Doping CH3NH3PbI3-xBrx Thin Films for Efficient TiO2 Nanorod Array Perovskite Solar Cells.

CH3NH3PbI3-xBrx thin films with different Br contents were successfully prepared on TiO2 nanorod array using 1.7 M PbI2 · DMSO complex solution and 0.465 M CH3NH3x(x = Br, I) precursors with different CH3NH3Br contents (molar ratios) by sequential deposition method. The influence of CH3NH3Br contents on the chemical composition, crystallinity, optical absorption and surface morphology of the CH3NH3PbI3-xBrx thin films was systematically investigated, and the photovoltaic performance of the corresponding TiO2 nanorod array perovskite solar cells was evaluated. The results revealed that the CH3NH3PbI3-xBrx solar cells using CH3NH3x precursors with 5% CH3NH3Br exhibited the best photoelectric conversion efficiency (PCE) of 16.47%, along with an open-circuit voltage (Voc) of 1.02 V, short-circuit photocurrent density (Jsc) of 20.99 mA · cm-2 and fill factor (FF) of 0.77, and the average PCE of 16.06 ± 0.52%, along with Voc of 1.02 ± 0.01 V, Jsc of 20.41 ± 0.58 mA · cm-2 and FF of 0.77 ± 0.01 under illumination of simulated AM 1.5 sunlight (100 mA cm-2).