Search

The function labeling of enzymes has a wide range of application value in the medical field, industrial biology and other fields. Scientists define enzyme categories by enzyme commission (EC) numbers. At present, although there are some tools for enzyme function prediction, their effects have not reached the application level. To improve the precision of enzyme function prediction, we propose a parallel convolutional contrastive learning (PCCL) method to predict enzyme functions. First, we use the advanced protein language model ESM-2 to preprocess the protein sequences. Second, PCCL combines convolutional neural networks (CNNs) and contrastive learning to improve the prediction precision of multifunctional enzymes. Contrastive learning can make the model better deal with the problem of class imbalance. Finally, the deep learning framework is mainly composed of three parallel CNNs for fully extracting sample features. we compare PCCL with state-of-art enzyme function prediction methods based on three evaluation metrics. The performance of our model improves on both two test sets. Especially on the smaller test set, PCCL improves the AUC by 2.57%.

Genome-wide association studies have shown that common genetic variants associated with complex diseases are mostly located in non-coding regions, which may not be causal. In addition, the limited number of validated non-coding functional variants makes it difficult to develop an effective supervised learning model. Therefore, improving the accuracy of predicting non-coding causal variants has become critical. This study aims to build a transfer learning-based machine learning method for predicting regulatory variants to overcome the problem of limited sample size. This paper presents a supervised learning method transfer support vector machine (TSVM) for massively parallel reporter assays (MPRA) validated regulatory variants prediction. First, uses a convolutional neural network to extract features with transfer learning. Second, the extracted features are selected by random forest method. Third, the selected features are used to train support vector machine for classification. We performed scale sensitivity experiments on the MPRA dataset and validated the effectiveness of transfer learning. The model achieves the Mcc of 0.326 and the AUC of 0.720, which are higher than the state-of-the-art method.

RNA-binding proteins are important for the process of cell life activities. High-throughput technique experimental method to discover RNA-protein binding sites is time-consuming and expensive. Deep learning is an effective theory for predicting RNA-protein binding sites. Using weighted voting method to integrate multiple basic classifier models can improve model performance. Thus, in our study, we propose a weighted voting deep learning model (WVDL), which uses weighted voting method to combine convolutional neural network (CNN), long short term memory network (LSTM) and residual network (ResNet). First, the final forecast result of WVDL outperforms the basic classifier models and other ensemble strategies. Second, WVDL can extract more effective features by using weighted voting to find the best weighted combination. And, the CNN model also can draw the predicted motif pictures. Third, WVDL gets a competitive experiment result on public RBP-24 datasets comparing with other state-of-the-art methods. The source code of our proposed WVDL can be found in https://github.com/biomg/WVDL.

Computational prediction of the RBP bound sites using features learned from existing annotation knowledge is an effective method because high-throughput experiments are complex, expensive and time-consuming. Many methods have been proposed to predict RNA-protein binding sites. However, the partial information of RNA sequence is not fully used. In this study, we propose multiple convolutional neural networks (MCNN) method, which predicts RNA-protein binding sites by integrating multiple convolutional neural networks constructed by RNA sequence information extracted from windows with different lengths. First, MCNN trains multiple CNNs base on RNA sequences extracted by different window lengths. Second, MCNN can extract more binding patterns of RBPs by combining these trained multiple CNNs previously. Third, MCNN only uses RNA base sequence information for RNA-protein binding sites prediction, which extracts sequence binding features and predicts the result with same architecture. This avoids the information loss of feature extraction step. Our proposed MCNN demonstrates a competitive performance comparing with other methods on a large-scale dataset derived from CLIP-seq, which is an effective method for RNA-protein binding sites prediction. The source code of our proposed MCNN method can be found in https://github.com/biomg/MCNN.

With the development of the automobile industry, the improvement of the reliability test level and the intelligent fault diagnosis level of the engine, as the heart component of the automobile and the assembly with the highest failure rate, becomes more and more important. In a context where technology is being upgraded, the development of intelligent automotive engine inspection system based on IoT thinking is set to become the major trend in the future. The objective of this article would be to have better knowledge about smart car engine system in the 5G IoT era. It is also through the study, a dynamic testing and characterization of the engine. The performance of the automobile engine, especially the dynamic performance, is further optimized. In this article, we focus on the PUMAOPEN test system of AVL and the system under test, the state management of the test process, and the intelligent transformation of key components. The results show that: studies and analyses the dynamic working conditions of intelligent vehicles on the Internet of Things, and compares the dynamic working conditions with the steady working conditions from qualitative to quantitative. By analyzing the influence of throttle opening change rate and speed change rate on dynamic performance and engine dynamic torque estimation, the fast sensing and response-ability in the engine testing process is improved. When the engine is accelerated by the experimental method presented in this article, and when the throttle opening is more than 30% and the engine speed varies at 100 and 200 rpm/s, the output value of the torque is higher than the output value of the steady-state torque or has little difference with the output value of the steady-state torque, and tends to be stable. In deceleration performance, the dynamic output torque of the engine is asymmetric. And it will deviate from the steady-state output torque, showing different dynamic response characteristics in different working sections. Throttle delay, overshoot characteristics, mechanical inertia, inflation coefficient, air–fuel ratio, and ignition advance angle when engine speed changes sharply will make engine dynamic characteristics deviate from steady-state characteristics.

Control of magnetism has attracted interest in achieving low-power and high-speed applications such as magnetic data storage and spintronic devices. Two-dimensional magnets allow for control of magnetic properties using the electric field, electrostatic doping and strain. In two-dimensional atomically thin magnets, a non-volatile all-optical method would offer the distinct advantage of switching magnetic states without application of an external field. Here, we demonstrate such all-optical magnetization switching in the atomically thin ferromagnetic semiconductor, CrI3, triggered by circularly polarized light pulses. The magnetization switching behaviour strongly depends on the exciting photon energy and polarization, in correspondence with excitonic transitions in CrI3, indicating that the switching process is related to spin angular momentum transfer from photoexcited carriers to local magnetic moments. Such an all-optical magnetization switching should allow for further exploration of magneto-optical interactions and open up applications in high-speed and low-power spintronic devices.

In the era of new media, it is of great significance to explore the strategies to enhance the international discourse power of the Chinese Wushu and to help build China into a strong sports country. Through literature study, text analysis, induction and deduction, this paper explored into the opportunities, challenges and countermeasures of promoting the international discourse power of the Chinese Wushu in the new media era. It was found that the new media could enrich the content and expression of the Chinese Wushu discourse, broaden the subject, form and channel of the Chinese Wushu discourse communication, and enhance the effectiveness of the Chinese Wushu discourse communication, which provided opportunities for the promotion of international discourse power of the Chinese Wushu faced with excessive media dependence, digital division, fragmented communication and discourse carnival. Based on this, this paper put forward the strategies to strengthen the Chinese Wushu, to strengthen the mass media communication, to optimize the Wushu communication talents, to enhance the dialogue ability of the discourse subjects, to strengthen the international discourse power. It was necessary to reconstruct the communication discourse system of Wushu, eliminate the barrier of cross-cultural communication, build the brand of international Wushu events, promote the gem of the Chinese traditional culture, subdivide the communication field of new media users, and create a harmonious network discourse space. [ABSTRACT FROM AUTHOR]

Polymer/graphene composites have attracted much attention for their conductivity and other functions. To control structure of the composites more effectively, we adopted suction filtration combined with the layer-by-layer (LbL) method based on our proposed particle construction method by using polystyrene/graphene (PS/rGO) composite particles as building blocks to prepare PS/rGO composites having a graphene orientation-3D network binary structure. The resulting PS/rGO composites had an excellent conductivity of 1344.3 S/m at a rGO content of 2.10 vol %. Besides, there were always an upgrade in conductivity of the composites along with an improvement of the preparation strategy, from particle constructing, suction filtration, and LbL methods to appropriate adjustment of rGO content in different layers. The results show that there was a clear relationship between the composite structure and the electrical properties and a low efficiency–consumption ratio for both electrical and mechanical properties.

In the era of big data, computer science has been applied to every aspect of biomedical field. At the same time, transforming biomedical data into valuable knowledge is one of the most important challenges of bioinformatics. Consequently, the birth of deep learning algorithms has greatly promoted the development of bioinformatics as a result of its ability to incorporate existing knowledge from vast datasets. In this review, we focus on the related deep learning techniques based on omics in the direction of disease prediction and present brief descriptions and comments of these studies based on DNA, RNA, protein and multi-omics datasets. Finally, we give the statistical results of the types and numbers of papers published in the field of disease prediction and discuss the potential development of deep learning in this field.

Catechol, nitrite, and dissolved metals are ubiquitous in source drinking water. Catechol and nitrite have been identified as precursors for halonitromethanes (HNMs), but the effect of metal ions on HNM formation during chlorination remains unclear. The main objective of this study was to investigate the effect of metal ions (Fe3+, Ti4+, Al3+) on the formation of trichloronitromethane (TCNM) (the most representative HNM species in disinfected water) on chlorinating catechol and nitrite. Trichloronitromethane was extracted by methyl tert‐butyl ether and detected by gas chromatography. The results show that metal ions promoted the formation of TCNM and that the enhancement efficiency followed the order of Fe3+> Ti4+> Al3+. Trichloronitromethane formation increased greatly within 2 h, and a basic condition (pH 8–9) favored TCNM formation more than acidic or neutral conditions. The conjoint effect of the metal–ion mixtures was shown to be similar to that of the single metal ion having the highest promoting effect on TCNM formation. Our results strongly suggest that metal ions play a significant role in enhancing TCNM formation. Metal ions increased TCNM formation during chlorination of catechol and nitrite.Type and concentration of metal ions affected TCNM yield.Effect of metal ions on the formation of TCNM varied with pH.Conjoint effect of metal ions on TCNM formation was not obvious.

Using literature review, the present paper analyzes the PDCA cycle of result optimization -- oriented sports curriculum. Meanwhile, it assesses the curriculum evaluation method by combining empirical and humanity methods. The study holds that the PDCA cycle includes four steps, namely, design of evaluation plan, data collection, operation of original curriculum program and effect verification of revised curriculum program. The four steps form a continuous cycle, thereby ensuring the constant improvement of physical education curriculum program. Our findings include; the PDCA cycle can ensure the achievement of learning objectives in a comprehensive and consistent way through vertical decomposition of the curriculum objectives. Besides, in order to save time, teachers can combine formal and informal evaluation, make sampling and complete evaluation in the process of evaluation. Teachers can also employ appropriate tools to assess in a comprehensive way students action, attention, understanding and efforts so that they can give students feedback information, find out students advantages and disadvantages, figure out the reasons way some students are at a disadvantage, and finally put forward hypotheses regarding curriculum improvement and revise the curriculum program. [ABSTRACT FROM AUTHOR]

With the development of the times, there is a huge amount of data in every industry. Big data technology is to collect, analyze, process and information from these huge data and apply it to all aspects of our life to improve people's production life. The proposal and development of smart agriculture will play a significant role in the further implementation of the country's rural revitalization strategy. However, the research on risk monitoring in smart agriculture is not yet systematic enough. The purpose of this article is to strengthen the development of smart agriculture under big data, focusing on risk monitoring. To this end, this article studies big data monitoring through data analysis methods and Internet of Things technologies, and discusses the principles of big data and key technology principles of the Internet of Things. Explained, and proposed a modern agricultural technology platform based on the Internet of Things and big data. This platform is established on the basis of precision agriculture and wireless sensor network work, and analyzes the accuracy of various types of light wavelengths for determining wheat rust. The analysis results show that the accuracy of light with multiple wavelengths is not necessarily better than the light with a single wavelength. The accuracy of 600 nm wavelength can reach 100 %, and the accuracy of monitoring with 4 wavelengths together is rather low. However, less consideration is given to the factors in this article, and wheat does not necessarily cause only one disease. Comprehensive analysis is required. With the help of spectral research and aerial photography of drones, the severity of the disease and the outbreak area can be speculated.

Patients with congenital absence of the portal vein (CAPV) often suffer from additional medical complications such as hepatic tumors and cardiac malformations. Congenital absence of the portal vein (CAPV) is a rare malformation. We present a case of a 16-year-old Chinese girl with CAPV with multiple pathology-proven hepatic focal nodular hyperplasias (FNHs) and ventricular septal defect (VSD). The CT and MRI features of this case are described, and previously reported cases are reviewed. CAPV is a rare congenital anomaly and in such patients, clarifying the site of portosystemic shunts, liver disease, and other anomalies is critical for appropriate treatment selection and accurate prognosis determination. Close follow-up, including laboratory testing and radiologic imaging, is recommended for all CAPV patients.

The surfaces of nanotube arrays were coated with poly(methyl methacrylate) (PMMA) using an imprinting method with an anodized alumina membrane as the template. The prepared nanotube array surfaces then either remained untreated or were coated with NH2(CH2)3Si(OCH3)3(PDNS) or CF3(CF2)7CH2CH2Si(OC2H5)3(PFO). Thus, nanotube arrays with three different surfaces, PDNS, PMMA (without coating), and PFO, were obtained. All three surfaces (PDNS, PMMA, and PFO) exhibited superhydrophobic properties with contact angles (CA) of 155, 166, and 168°, respectively, and their intrinsic water contact angles were 30, 79, and 118°, respectively. The superhydrophobic stabilities of these three surfaces were examined under dynamic impact and static pressures in terms of the transition from the Cassie–Baxter mode to the Wenzel mode. This transition was determined by the maximum pressure (pmax), which is dependent on the intrinsic contact angle and the nanotube density of the surface. A pmaxgreater than 10 kPa, which is sufficiently large to maintain stable superhydrophobicity under extreme weather conditions, such as in heavy rain, was expected from the PFO surface. Interestingly, the PDNS surface, with an intrinsic CA of only 30°, also displayed superhydrophobicity, with a CA of 155°. This property was partially maintained under the dynamic impact and static pressure tests. However, under an extremely high pressure (0.5 MPa), all three surfaces transitioned from the Cassie–Baxter mode to the Wenzel mode. Furthermore, the lost superhydrophobicity could not be recovered by simply relieving the pressure. This result indicates that the best way to maintain superhydrophobicity is to increase the pmaxof the surface to a value higher than the applied external pressure by using low surface energy materials and having high-density binary nano-/microstructures on the surface.

A simple and low-cost technique for the preparation of silicon-oil-infused polydimethylsiloxane (PDMS) coatings with different silicon oil contents have been developed and studied. This material is designed for ice-phobic applications, and thus a high hydrophobic property of PDMS is maintained by avoiding any polar groups such as CO and OH in the structure. Therefore, the polymer main chain was attached with vinyl and Si–H groups to obtain a cross-linking capability, meanwhile to ensure a nonpolar chemical structure. Its ice-phobic property has been investigated in terms of ice adhesion strength (tensile and shear), water contact angle, icing dynamics using high-speed photography and morphology using TEM, SEM and AFM. The prepared coating surface shows a low surface energy and very low ice adhesion strength of 50 kPa, only about 3% of the value on a bare aluminum (Al) surface. In the silicon oil infused PDMS coatings, the low surface energy of the silicon oil and PDMS, and the high mobility of silicon oil played an important role on the ice-phobic property. Both of these factors offer the surface a large water contact angle and hence a small contact area, leading to the formation of a loose ice layer. In addition, the oil infused polymer structure significantly reduces the contact area of the ice with solid substrate since the ice mostly contacts with the mobile oil. This leads to a very weak interaction between the substrate and ice, consequently significantly reduces the ice adhesion strength on the surface. Therefore, such material could be a good candidate for ice-phobic coatings on which the accumulated ice may be easily removed by a nature force, such as wind, gravity, and vibration.

Four aluminum surfaces with wettability varied from superhydrophilic to superhydrophobic were prepared by combining an etching and a coating process. The surface wettability was checked in terms of water contact angle (CA) and sliding angle (SA) under different humidity at −10 °C. High-speed photography was applied to study water droplet impact dynamics on these surfaces. It was found that single and successive water droplets could rebound on the superhydrophobic surface and roll off at a tilt angle larger than 30° under an extremely condensing weather condition (−10 °C and relative humidity of 85–90%). In addition, the superhydrophobic surface showed a strong icephobic property, the ice adhesion on this surface was only 13% of that on the superhydrophilic surface, though they had a similar nano/microtopological structure. Moreover, this superhydrophobic surface displayed an excellent durability of the icephobic property. The ice adhesion only increased to 20% and 16% of that on the superhydrophobic surface after the surface was undergone 20 icing/ice-breaking cycles and 40 icing/ice-melting cycles, respectively. Surface profile and XPS studies on these surfaces indicated a minor damage of the surface nano/microstructure and the coating layer upon these multiple ice-breaking and ice-melting processes. Therefore, this superhydrophobic surface could be a good candidate for icephobic applications.

In this paper, we have comparatively investigated the stability of superhydrophobic behaviors of fresh and biomimetic lotus leaf surfaces under controlled water condensation conditions. The binary micro/nano structures of the superhydrophobic surfaces are observed with electron micrographs. Contact and sliding angles are evaluated by syringing water droplets on the substrates with surface temperatures and humidity precisely controlled between −10 and 30 °C, and RH = 10, 30, 60, and 90%. According to the calculations on the solid−liquid contact area fraction in different environmental conditions based on a micro/nano binary structure model, the effects of condensed water on superhydrophobic surfaces are assessed quantitatively. Both the calculated and experimental results indicate that the temperature difference between surface temperature and the dew point during measurement is essential to the occurrence of water condensation while the effect of condensation on the surface wettability also depends on the topology of hierarchical structured surfaces. The loss of water repellency that usually appears on the artificial superhydrophobic surface under low temperature and high humidity conditions is proved to be reversible, showing a bidirectional transition of the equilibrium state between Wenzel and Cassie−Baxter.

A comb copolymer with a fluorinated FPAE main chain and narrowly dispersed PMS graft chains has been prepared. The micellization behavior of this copolymer in a methanolacetone mixture was studied. DLS studies showed hydrodynamic diameters of the micelles between 50 and 80 nm, while the diameters of dry micelles were 40 nm in TEM images. Micelle films with an interesting porous composite structure of nanofibers bearing micelle particles on the surface have been prepared by double coating the micelle solution. Superhydrophobicity with a WCA of 165° and a sliding angle of 3° was observed. As comb copolymers can adjust their molecular weight and block length independently, materials with a high physical strength and controllable morphological structure can be obtained.

Multicontrast, high-resolution carotid magnetic resonance imaging (MRI) has been validated with histology to quantify atherosclerotic plaque morphology and composition. For evaluating the lipid-rich necrotic core (LRNC) and fibrous cap, both of which are key elements in determining plaque stability, the combined pre- and postcontrast T1-weighted (T1W) sequences have been recently shown to have a higher reproducibility than other contrast weightings. In this study, we sought to determine whether contrast weightings beyond T1W (pre- and postcontrast) are necessary for comprehensive, quantitative, carotid plaque interpretation.

A series of hydroxy-terminated polybutadiene (HTPB) polyureas modified with different liquid polysulfide content was synthesized and their structure and oil-resistant properties were studied by attenuated total reflectance–FTIR spectroscopy, dynamic mechanical analysis, isothermal aging and differential scanning calorimetry, stress–strain analysis, oil absorption, and oil-resistance test. The results showed that polysulfide–polybutadiene polyureas retained low temperature flexibility and had lower oil absorption and better oil resistance than that of HTPB-based polyureas. © 2003 Wiley Periodicals, Inc. J Appl Polym Sci 89: 2672–2675, 2003

The delta-1-pyrroline-5-carboxylate synthetase (P5CS) gene exercises a protective function in stressed plants. However, the relationship between proline accumulation caused by P5CSand abiotic stress tolerance in plants is not always clear, as P5CSoverexpression has been reported to repress plant growth under normal conditions in several reports. We re-evaluated the role of P5CSin drought-tolerant rice breeding by expressing the AtP5CS1and feedback-inhibition-removed AtP5CS1(AtP5CS1F128A) genes under the regulation of an ABA-inducible promoter to avoid the potential side effects of P5CSoverexpression under normal conditions. ABA-inducible AtP5CS1and AtP5CS1F128Aincreased seedling growth in a nutrient solution (under osmotic stress) and grain yield in pot plants. However, the evidently deleterious effects of AtP5CS1on grain quality, tiller number, and grain yield in the field indicated the unsuitability of P5CSfor drought-tolerance breeding.

We systematically evaluated water contact and sliding angles on the lotus leaf from −10 to 80 °C under either extremely low or high relative humidity. SEM images revealed changes of the hierarchical structures after thermal treatment at different temperatures. The wettability recovery test showed consistent results that a restorable loss of superhydrophobic behavior appeared under low temperature and high humidity due to water condensation, whereas structural destruction at high temperatures caused the lotus leaf to lose extraordinary water repellency permanently.