Search

Abstract Machine learning with optical neural networks has featured unique advantages of the information processing including high speed, ultrawide bandwidths and low energy consumption because the optical dimensions (time, space, wavelength, and polarization) could be utilized to increase the degree of freedom. However, due to the lack of the capability to extract the information features in the orbital angular momentum (OAM) domain, the theoretically unlimited OAM states have never been exploited to represent the signal of the input/output nodes in the neural network model. Here, we demonstrate OAM-mediated machine learning with an all-optical convolutional neural network (CNN) based on Laguerre-Gaussian (LG) beam modes with diverse diffraction losses. The proposed CNN architecture is composed of a trainable OAM mode-dispersion impulse as a convolutional kernel for feature extraction, and deep-learning diffractive layers as a classifier. The resultant OAM mode-dispersion selectivity can be applied in information mode-feature encoding, leading to an accuracy as high as 97.2% for MNIST database through detecting the energy weighting coefficients of the encoded OAM modes, as well as a resistance to eavesdropping in point-to-point free-space transmission. Moreover, through extending the target encoded modes into multiplexed OAM states, we realize all-optical dimension reduction for anomaly detection with an accuracy of 85%. Our work provides a deep insight to the mechanism of machine learning with spatial modes basis, which can be further utilized to improve the performances of various machine-vision tasks by constructing the unsupervised learning-based auto-encoder.

Adeno-associated viral transduction allows the introduction of nucleic fragments into cells and is widely used to modulate gene expressions in vitro and in vivo. It enables the study of genetic functions and disease mechanisms and, more recently, serves as a tool for gene repair. To achieve optimal transduction performance for a given cell type, selecting an appropriate serotype and the number of virus particles per cell, also known as the multiplicity of infection, is critical. Fluorescent proteins are one of the common reporter genes to visualize successfully transduced cells and assess transduction efficiencies. Traditional methods of measuring fluorescence-positive cells are endpoint analysis by flow cytometry or manual counting with a fluorescence microscope. However, the flow cytometry analysis does not allow further measurement in a test run, and manual counting by microscopy is time-consuming. Here, we present a method that repeatedly evaluates transduction efficiencies by adding the DNA-stain Hoechst 33342 during the transduction process combined with a microscope or live-cell imager and microplate image analysis software. The method achieves fast, high-throughput, reproducible, and real-time post-transduction analysis and allows for optimizing transduction parameters and screening for a proper approach.

BackgroundDepression is widespread global problem that not only severely impacts individuals’ physical and mental health but also imposes a heavy disease burden on nations and societies. The role of inflammation in the pathogenesis and pathophysiology of depression has received much attention, but the precise relationship between the two remains unclear. This study aims to investigate the correlation between depression and inflammation using a network medicine approach.MethodsWe utilized a degree-preserving approach to identify the large connected component (LCC) of all depression-related proteins in the human interactome. The LCC was deemed as the disease module for depression. To measure the association between depression and other diseases, we calculated the overlap between these disease protein modules using the Sab algorithm. A smaller Sab value indicates a stronger association between diseases. Building on the results of this analysis, we further explored the correlation between inflammation and depression by conducting enrichment and pathway analyses of critical targets. Finally, we used a network proximity approach to calculate drug-disease proximity to predict the efficacy of drugs for the treatment of depression. We calculated and ranked the distances between depression disease modules and 6,100 drugs. The top-ranked drugs were selected to explore their potential for treating depression based on the hypothesis that their antidepressant effects are related to reducing inflammation.ResultsIn the human interactome, all depression-related proteins are clustered into a large connected component (LCC) consisting of 202 proteins and multiple small subgraphs. This indicates that depression-related proteins tend to form clusters within the same network. We used the 202 LCC proteins as the key disease module for depression. Next, we investigated the potential relationships between depression and 299 other diseases. Our analysis identified over 18 diseases that exhibited significant overlap with the depression module. Where SAB = −0.075 for the vascular disease and depressive disorders module, SAB = −0.070 for the gastrointestinal disease and depressive disorders module, and SAB = −0.062 for the endocrine system disease and depressive disorders module. The distance between them SAB

With the advent of the era of big data, data mining methods show their powerful information mining ability in various fields, seeking the association information hidden in the data, which is convenient for people to make scientific decisions. This paper analyses the butterfly effect in the agricultural product industry chain from the perspective of producer and consumer by using multidimensional time and space theory and proposes a new price forecasting method. We consider that the price change of agricultural products is not only affected by the balance of market supply and demand but also by the factors of time and space. Taking the pig industry chain of Sichuan Province as an example, this paper explores and excavates the data from 2010 to 2020 in the time dimension. Interestingly, we found that the price changes in pork in the market are generally highly correlated with the prices of slaughtered pigs, piglets a few weeks ago and the prices of multiple feed a few months ago. Based on the precise time-space factors, we improved the price forecasting model, greatly improved the accuracy of price prediction, and proved the effectiveness of multidimensional spatiotemporal association mining. The research in this paper is helpful to establish a brand-new agricultural product price prediction theory, which is of great significance to the development of the agricultural economy and global poverty alleviation.

PurposeTransforming growth factor-beta (TGFB)-mediated epithelial-mesenchymal transition (EMT) plays a crucial role in the pathogenesis of retinal fibrosis, which is one of the leading causes of impaired vision. Current approaches to treating retinal fibrosis focus, among other things, on inhibiting the TGFB signaling pathway. Transient expression of microRNAs (miRNAs) is one way to inhibit the TGFB pathway post-transcriptionally. Our previous study identified the miRNA miR-302d as a regulator of multiple TGFB-related genes in ARPE-19 cells. To further explore its effect on primary cells, the effect of miR-302d on TGFB-induced EMT in primary human retinal pigment epithelium (hRPE) was investigated in vitro.MethodshRPE cells were extracted from patients receiving enucleation. Transfection of hRPE cells with miR-302d was performed before or after TGFB1 stimulation. Live-cell imaging, immunocytochemistry staining, Western blot, and ELISA assays were utilized to identify the alterations of cellular morphology and EMT-related factors expressions in hRPE cells.ResultshRPE cells underwent EMT by TGFB1 exposure. The transfection of miR-302d inhibited the transition with decreased production of mesenchymal markers and increased epithelial factors. Meanwhile, the phosphorylation of SMAD2 activated by TGFB1 was suppressed. Moreover, miR-302d expression promoted TGFB1-induced fibroblast-like hRPE cells to revert towards an epithelial stage. As confirmed by ELISA, miR-302d reduced TGFB receptor 2 (TGFBR2) and vascular endothelial growth factor A (VEGFA) levels 48 hours after transfection.ConclusionsThe protective effect of miR-302d might be a promising approach for ameliorating retinal fibrosis and neovascularization. MiR-302d suppresses TGFB-induced EMT in hRPE cells via downregulation of TGFBR2, even reversing the process. Furthermore, miR-302d reduces the constitutive secretion of VEGFA from hRPE cells.

Abstract Background Several different technologies are used for prenatal screening procedures and genetic diagnostic technologies. We aimed to investigate the rates of chromosomal abnormalities in cases with different abnormal prenatal indications and to determine the relationships between fetal chromosomal abnormalities and indicators of prenatal abnormalities in Northeast China. Methods We evaluated 4953 16- to 23-week singleton gestation cases using amniocentesis and a total of 3583 participants received serological screening. Fetal chromosomal analyses were performed for all samples using fluorescence in situ hybridization and karyotyping. Results Among these samples, 204 (4.12%) had fetal chromosomal abnormalities. A total of 3583 participants received serological screening, among whom 102 (2.85%) exhibited positive results. A total of 309 participants had ultrasonography; 42 (13.6%) of these had abnormalities. Among 97 participants who had non-invasive prenatal testing (NIPT), 59 (61%) had positive results. Among 1265 participants with advanced maternal age, 78 (6.2%) had abnormal results. Conclusion The serological screening and NIPT that were included in the prenatal screening methods all had false positive and false negative rates. Although they are both prenatal screening techniques, maternal serum screening cannot be replaced by NIPT. The pregnancy women should accept NIPT in a qualified prenatal diagnostic center. We recommend that pregnant women at high or critical risk undergoing prenatal screening should confirm the fetal karyotype through amniocentesis. Moreover, if women receive a positive result via NIPT, they should not have a pregnancy termination without undergoing further prenatal diagnosis.

Down syndrome (DS) caused by the presence of part or all of a third copy of chromosome 21 is the most common form of aneuploidy. The prenatal screening for DS is a key component of antenatal care and is recommended to be universally offered to women irrespective of age or background. The objective of this paper is to introduce a noninvasive and accurate diagnosis procedure for DS and to minimize social and financial cost of prenatal diagnosis. Recently, machine learning has received considerable attention in predictive analytics for medical problems. However, there is few its applications on DS prediction reported due to the difficulty of dealing with highly imbalanced and feature-correlated screening data. In this paper, we propose a cascaded machine learning framework designed for DS prediction based on three complementary stages: 1) pre-judgment with isolation forest technique, 2) model ensemble by voting strategy, and 3) final judgment using logistic regression approach. The experimental results show that the performance of this framework on maternal serum screening data set, when evaluated with different evaluation parameters, is superior to those of some machine learning methods. The best suggested combination of input features for DS screening is the group of alpha-fetoprotein, human chorionic gonadotropin, unconjugated estriol, and maternal age. In addition, our method has the potential to generate further accurate prediction for imbalanced and feature-correlated data, thereby providing a novel and effective approach for certain diseases analysis.

Coupling graphene with narrow band-gap Ti2O3 nanoparticles can enable efficient mid-infrared photodetection. Here, the authors report a graphene-Ti2O3 based hybrid photodetector with high responsivity of ~300 A W-1 up to 10 μm by varying the number of graphene layers and size of Ti2O3 nanoparticles.

E-tailers can impact customers’ order incidence and cart size by offering an attractive shipping fee; they can also influence customers to purchase other products using a recommendation system. This article provides a combined model in which the shipping-fee discount and product recommendations occur jointly to maximize the sales volume of e-tailers. Selection of recommended products and discount of shipping fee are the dynamic issues involving the complexity of business environments. The proposed model first determines the optimal shipping-fee discount to attract customers and then establishes the most suitable product recommendation list to encourage customers to spend more money on other products. Through the use of a shipping-fee discount and simultaneous product recommendations, e-tailers maximize the sales volume. As for consumers, they can always get additional savings from the subsequent shopping process because the product’s shipping fee is lower than the original shipping costs. Experimental results indicate that the proposed approach is feasible and better than alternative methods for achieving sales volume maximization. Furthermore, sensitivity analysis validates the stability of the proposed model in the face of an uncertain environment. Therefore, using the shipping-fee discount and product recommendation system as a united marketing strategy can maximize e-tailers’ sales volume.

The identification of plant disease is the premise of the prevention of plant disease efficiently and precisely in the complex environment. With the rapid development of the smart farming, the identification of plant disease becomes digitalized and data-driven, enabling advanced decision support, smart analyses, and planning. This paper proposes a mathematical model of plant disease detection and recognition based on deep learning, which improves accuracy, generality, and training efficiency. Firstly, the region proposal network (RPN) is utilized to recognize and localize the leaves in complex surroundings. Then, images segmented based on the results of RPN algorithm contain the feature of symptoms through Chan–Vese (CV) algorithm. Finally, the segmented leaves are input into the transfer learning model and trained by the dataset of diseased leaves under simple background. Furthermore, the model is examined with black rot, bacterial plaque, and rust diseases. The results show that the accuracy of the method is 83.57%, which is better than the traditional method, thus reducing the influence of disease on agricultural production and being favorable to sustainable development of agriculture. Therefore, the deep learning algorithm proposed in the paper is of great significance in intelligent agriculture, ecological protection, and agricultural production.

Background/Aims: Monocyte chemotactic protein-induced protein 1 (MCPIP1) plays a crucial role in various cellular processes, including neurogenesis. However, the relationship between MCPIP1 and myocardial ischemia/reperfusion (I/R) injury remained illdefined. In this study, we explored whether the I/R-mediated up-regulation of MCPIP1 is critical in the modulation of both cell migration and apoptosis in human umbilical vein endothelial cells (HUVECs). Methods: Using Western blot analysis and quantitative real-time PCR, the protein expression and mRNA transcription, respectively, of MCPIP1 was detected in HUVECs. To investigate cell migration, an in vitro scratch assay and a nested matrix model were applied. Results: I/R increased the expression of MCPIP1 via the activation of the mitogen-activated protein kinase (MAPK) and PI3K/Akt pathways. I/R increased migration and apoptosis of HUVECs, which were significantly inhibited by MCPIP1 siRNA. Conclusion: These findings suggest that I/R-mediated up-regulation of MCPIP1 regulates migration and apoptosis in HUVECs. Understanding the regulation of MCPIP1 expression and function may aid in the development of an adjunct therapeutic strategy in the treatment of individuals with I/R injury.

Human skin sensing of mechanical stimuli originates from transduction of mechanoreceptors that converts external forces into electrical signals. Although imitating the spatial distribution of those mechanoreceptors can enable developments of electronic skins capable of decoupled sensing of normal/shear forces and strains, it remains elusive. We report a three-dimensionally (3D) architected electronic skin (denoted as 3DAE-Skin) with force and strain sensing components arranged in a 3D layout that mimics that of Merkel cells and Ruffini endings in human skin. This 3DAE-Skin shows excellent decoupled sensing performances of normal force, shear force, and strain and enables development of a tactile system for simultaneous modulus/curvature measurements of an object through touch. Demonstrations include rapid modulus measurements of fruits, bread, and cake with various shapes and degrees of freshness. [ABSTRACT FROM AUTHOR]

Based on the analysis of the importance of professional cluster construction by ecological theory, with the change of social demand for talents, this paper explores the practice of environmental chemical professional cluster construction in Pingdingshan University, including gradually perfecting teaching conditions and reforming teaching mode, breaking through the limitations of resources, integrating the boundaries of colleges and departments, integrating multiple resources, innovating systems and mechanisms, reconstructing professional clusters, deconstructing professional connotations, reorganizing curriculum systems, etc., in order to better build the ecological chain network of education in application-oriented colleges and universities, realize the deep integration of industry and education, train future-oriented interdisciplinary applied talents of new engineering, and realize the construction of characteristic professional cluster in application-oriented colleges. [ABSTRACT FROM AUTHOR]

In this article, we focus on prediction of a target model by transferring the information of source models. To be flexible, we use semiparametric additive frameworks for the target and source models. Inheriting the spirit of parameter-transfer learning, we assume that different models possibly share common knowledge across parametric components that is helpful for the target predictive task. Unlike existing parameter-transfer approaches, which need to construct auxiliary source models by parameter similarity with the target model and then adopt a regularization procedure, we propose a frequentist model averaging strategy with a J-fold cross-validation criterion so that auxiliary parameter information from different models can be adaptively transferred through data-driven weight assignments. The asymptotic optimality and weight convergence of our proposed method are built under some regularity conditions. Extensive numerical results demonstrate the superiority of the proposed method over competitive methods. [ABSTRACT FROM AUTHOR]

The quality and quantity of crop yields in agriculture primarily depend on the timing and precision of various implemented farming behaviors. Basins and hills dominate southwest China. Due to topographical constraints, the rate of agricultural mechanization in the region remains low, and agriculture remains primarily non-mechanized. The acquisition and recognition of information on farming behaviors play an important role in crop production. In this article, transfer learning was used in a current state-of-the-art 3DCNN-based behavior recognition model for farming behavior recognition and classification tasks. The focus was on fine-tuning and evaluating state-of-the-art 3D convolutional neural networks for farming behavior recognition. The evaluated architectures included Res3D, MC3, and R2+1D. The six common farming behaviors recognized include weeding, planting, harvesting, transplanting, fertilizing, and spraying. The accuracy of all models pretrained on Kinetics-400 after fine-tuning exceeded 90%, where MC3 had the best performance, with an accuracy of 0.9628, precision of 0.9647, sensitivity of 0.963, and specificity of 0.9925, which was slightly greater than the other two. MC3 was also the most lightweight of all models; its parameters were only 32.6% of Res3D and 36.7% of R2+1D. The experimental results demonstrated that the fine-tuned MC3 model offers high classification accuracy and effective recognition and classification of farming behaviors, which lays a good foundation for improved crop production. [ABSTRACT FROM AUTHOR]

Abstract Background Staphylococcus aureus and its single or mixed biofilm infections seriously threaten global public health. Phage therapy, which uses active phage particles or phage-derived endolysins, has emerged as a promising alternative strategy to antibiotic treatment. However, high-efficient phage therapeutic regimens have yet to be established. Results In this study, we used an enrichment procedure to isolate phages against methicillin-resistant S. aureus (MRSA) XN108. We characterized phage SYL, a new member of the Kayvirus genus, Herelleviridae family. The phage endolysin LysSYL was expressed. LysSYL demonstrated stability under various conditions and exhibited a broader range of efficacy against staphylococcal strains than its parent phage (100% vs. 41.7%). Moreover, dynamic live/dead bacterial observation demonstrated that LysSYL could completely lyse MRSA USA300 within 10 min. Scan and transmission electron microscopy revealed evident bacterial cell perforation and deformation. In addition, LysSYL displayed strong eradication activity against single- and mixed-species biofilms associated with S. aureus. It also had the ability to kill bacterial persisters, and proved highly effective in eliminating persistent S. aureus when combined with vancomycin. Furthermore, LysSYL protected BALB/c mice from lethal S. aureus infections. A single-dose treatment with 50 mg/kg of LysSYL resulted in a dramatic reduction in bacterial loads in the blood, liver, spleen, lungs, and kidneys of a peritonitis mouse model, which resulted in rescuing 100% of mice challenged with 108 colony forming units of S. aureus USA300. Conclusions Overall, the data provided in this study highlight the strong therapeutic potential of endolysin LysSYL in combating staphylococcal infections, including mono- and mixed-species biofilms related to S. aureus.