PPII-AEAT: Prediction of protein-protein interaction inhibitors based on autoencoders with adversarial training.

Protein-protein interactions (PPIs) have shown increasing potential as novel drug targets. The design and development of small molecule inhibitors targeting specific PPIs are crucial for the prevention and treatment of related diseases. Accordingly, effective computational methods are highly desired to meet the emerging need for the large-scale accurate prediction of PPI inhibitors. However, existing machine learning models rely heavily on the manual screening of features and lack generalizability. Here, we propose a new PPI inhibitor prediction method based on autoencoders with adversarial training (named PPII-AEAT) that can adaptively learn molecule representation to cope with different PPI targets. First, Extended-connectivity fingerprints and Mordred descriptors are employed to extract the primary features of small molecular compounds. Then, an autoencoder architecture is trained in three phases to learn high-level representations and predict inhibitory scores. We evaluate PPII-AEAT on nine PPI targets and two different tasks, including the PPI inhibitor identification task and inhibitory potency prediction task. The experimental results show that our proposed PPII-AEAT outperforms state-of-the-art methods.

Salicylic acid inducing the expression of maize anti-insect gene SPI: a potential control strategy for Ostrinia furnacalis.

BACKGROUND: Due to being rooted in the ground, maize (Zea mays L.) is unable to actively escape the attacks of herbivorous insects such as the Asian corn borer (Ostrinia furnacalis). In contrast to the passive damage, plants have evolved defense mechanisms to protect themselves from herbivores. Salicylic acid, a widely present endogenous hormone in plants, has been found to play an important role in inducing plant resistance to insects. In this study, we screened and identified the insect resistance gene SPI, which is simultaneously induced by SA and O. furnacalis feeding, through preliminary transcriptome data analysis. The functional validation of SPI was carried out using bioinformatics, RT-qPCR, and heterologous expression protein feeding assays. RESULTS: Both SA and O. furnacalis treatment increased the expression abundance of SA-synthesis pathway genes and SPI in three maize strains, and the upregulation of SPI was observed strongly at 6 hours post-treatment. The expression of SPI showed a temporal relationship with SA pathway genes, indicating that SPI is a downstream defense gene regulated by SA. Protein feeding assays using two different expression vectors demonstrated that the variation in SPI protein activity among different strains is mainly due to protein modifications. CONCLUSIONS: Our research results indicate that SPI, as a downstream defense gene regulated by SA, is induced by SA and participates in maize's insect resistance. The differential expression levels of SPI gene and protein modifications among different maize strains are one of the reasons for the variation in insect resistance. This study provides new insights into ecological pest control in maize and valuable insights into plant responses to SA-induced insect resistance.

Magnetic Adsorbent Fe3O4/ZnO/LC for the Removal of Tetracycline and Congo Red from Aqueous Solution.

Zeolitic imidazolate frameworks (ZIFs) can be used as an adsorbent to efficiently adsorb organic pollutants. However, ZIF nanoparticles are easy to form aggregates, hampering the effective and practical application in practical adsorption. In this study, the ZIF-8 was successfully loaded onto lignocellulose (LC) to further produce ZnO/LC by in situ growth method and hydrothermal treatment, and then Fe3O4 nanoparticles (Fe3O4 NPs) were loaded onto ZnO/LC to prepare magnetic Fe3O4/ZnO/LC adsorbent for removing tetracycline (TC) and congo red (CR) pollutants from aqueous solution. The adsorption properties of the adsorbent were systematically analyzed for different conditions, such as adsorbent dosage, solution pH, contact time, temperature and initial concentration. The experimental data were fitted using adsorption kinetic and isotherm models. The results showed that the pseudo-second-order model and Sips model were well fitted to the adsorption kinetic and adsorption isotherm, respectively. The adsorption capacities of TC and CR reached the maximum value of 383.4 mg/g and 409.1 mg/g in experimental conditions. The mechanism of the removal mainly includes electrostatic interaction, hydrogen bonding and π-π stacking. This novel adsorbent could be rapidly separated from the aqueous solution, suggesting its high potential to remove pollutants in wastewater.

Simple ultrasonic integration of shapeable, rebuildable, and multifunctional MIL-53(Fe)@cellulose composite for remediation of aqueous contaminants.

Metal-organic frames (MOFs) have been recognized as one of the best candidates in the remediation of aqueous contaminants, while the fragile powder shape restricts the practical implementation. In this work, a shapeable, rebuildable, and multifunctional MOF composite (MIL-53@CF) was prepared from MIL-53 (Fe) and cellulose fiber (CF) using a simple ultrasonic method for adsorption and photocatalytic degradation of organic pollutants in wastewater. The results showed MIL-53(Fe) crystals were uniformly growth on CF surfaces and bonded with surface nanofibrils of CF through physical crosslinking and hydrogen bonding. Because of the high bonding strength, the MIL-53@CF composite exhibited an excellent compressive strength (3.53 MPa). More importantly, the MIL-53@CF composite was rebuildable through mechanical destruction followed by re-ultrasonication, suggesting the excellent reusability of MIL-53@CF for water remediation. The MIL-53@CF composite also had high adsorption capacities for methyl orange (884.6 mg·g-1), methylene blue (198.3 mg·g-1), and tetracycline (106.4 mg·g-1). MIL-53@CF composite could degrade TC through photocatalysis. The photocatalytic degradation mechanism was attributed to the Fe(II)/Fe(III) transform cycle reaction of MIL-53 crystal located on MIL-53@CF. Furthermore, the mechanical property and remoldability of MIL-53@CF composite increased its practicability. Comprehensively, MIL-53@CF composite provided a possible strategy to practically apply MOF in the remediation of aqueous contaminants.

Nanocellulose-based polyurethane foam adsorbent for pressure-driven dye-contaminated water purification.

Dye-contaminated water has caused a worldwide pollution, which is threatening aquatic organisms and human health. In this work, a pressure-driven foam adsorbent (PFA) was bioinspired from the tapestry turban for purifying the dye-contaminated water. The PFA was prepared using an one-step method from nanocellulose (NC), amino-functionalized ZIF-8 (ZIF-8-NH2), and high resilience polyurethane foam (PUF). It was applied to efficiently remove methyl orange (MO) and crystal violet (CV) dyes from dye-contaminated waste solutions. The maximum adsorption capacity of PFA for MO and CV was 225.9 mg/g (25 °C, pH = 2) and 41.6 mg/g (25 °C, pH = 10), respectively, which were acceptable as compared with the reported works. The dyes could be efficiently removed from various river water samples. After 5 cycles, the removal efficiencies of MO and CV decreased from 92.0% and 85.7% to 84.7% and 76.1%, respectively. Moreover, the PFA relied on pressure-driven force to release the purified water under a low pressure.

Prediction of drug protein interactions based on variable scale characteristic pyramid convolution network.

MOTIVATION: In the process of drug screening, it is significant to improve the accuracy of drug-target binding affinity prediction. A multilayer convolutional neural network is one of the most popular existing methods for predicting affinity based on deep learning. It uses multiple convolution layers to extract features from the simplified molecular input system (SMILES) strings of the compounds and amino acid sequences of proteins and then performs affinity prediction analysis. However, the semantic information contained in low-level features can gradually be lost due to the increasing network depth, which affects the prediction performance. RESULT: We propose a novel method called the Pyramid Network Convolution Drug-Target Binding Affinity (PCNN-DTA) method for drug-target binding affinity prediction. The PCNN-DTA method, which is based on a feature pyramid network (FPN), fuses the features extracted from each layer of a multilayer convolution network to retain more low-level feature information, thus improving the prediction accuracy. PCNN-DTA is compared with other typical algorithms on three benchmark datasets, namely, the KIBA, Davis, and Binding DB datasets. Experimental results show that the PCNN-DTA method is superior to existing regression prediction methods using convolutional neural networks, which further demonstrates its effectiveness.

Modeling DTA by Combining Multiple-Instance Learning with a Private-Public Mechanism.

The prediction of the strengths of drug-target interactions, also called drug-target binding affinities (DTA), plays a fundamental role in facilitating drug discovery, where the goal is to find prospective drug candidates. With the increase in the number of drug-protein interactions, machine learning techniques, especially deep learning methods, have become applicable for drug-target interaction discovery because they significantly reduce the required experimental workload. In this paper, we present a spontaneous formulation of the DTA prediction problem as an instance of multi-instance learning. We address the problem in three stages, first organizing given drug and target sequences into instances via a private-public mechanism, then identifying the predicted scores of all instances in the same bag, and finally combining all the predicted scores as the output prediction. A comprehensive evaluation demonstrates that the proposed method outperforms other state-of-the-art methods on three benchmark datasets.

Effect of Scanning Speed on Properties of Laser Surface Remelted 304 Stainless Steel.

In order to study the microstructure and properties of stainless steel after laser surface remelting, based on the theory of laser surface remelting, a simulation model of nanosecond-pulsed laser surface remelted stainless steel was established to study the evolution law of the Marangoni force of the molten pool during laser surface remelting. A single-lane laser remelting experiment was performed to study the variation of the scanning speed on the remelting width, roughness, and layer microtopography. The "S" scanning path was used to remelt the stainless steel surface to investigate the bonding force between the remelted layer and the substrate, the hardness, microscopic morphology, and corrosion resistance. The results show that the viscosity of the liquid metal in the molten pool increases with the increase of the scanning speed. Larger liquid viscosity and smaller surface tension temperature gradients promote a weaker flow of liquid metal, which reduces the velocity of the liquid metal flow in the molten pool. With the increase of scanning speed, the remelting width gradually decreases, but the roughness gradually increases. When the element content of Cr increases, the element content of Fe and O decreases. The surface is covered with an oxide film, the main components of which are oxides of Cr and Fe, the remelted layer is greater than that of the substrate, and the corrosion resistance is improved. Laser surface remelting technology can improve the structure and properties of 304 stainless steel.

Peridynamic Simulation to Fracture Mechanism of CBN Grain in the Honing Wheel Dressing Process.

Regularly dressing of CBN honing wheel is an effective way to keep its sharpness and correct geometry during honing process. This study aims to understand the fracture mechanism of single CBN grain in the dressing process of honing wheel. The honing wheel dressing process was simplified into the dressing process of grinding wheel, and the bond-based Peridynamic method considering bond rotation effect was developed to investigate the progressive fracture evolution, stress characteristics, and fracture modes of CBN grains in this process. It was found that fracture evolution of CBN grains mainly underwent four stages: elastic deformation, damage initiation, crack formation, and macro fracture. In addition, the fracture initiation and propagation were mainly determined by the tensile and shear stress, where the former led to mode I fractures and the latter led to mode II fractures. The propagation of mode I fractures was stable while the propagation of mode II fracture was unstable. The results show that the Peridynamic approach has great potential to predict the fracture mechanism of CBN grain in the dressing process of honing and grinding wheels.

Inverse identification of pollution source release information for surface river chemical spills using a hybrid optimization model.

To evaluate the liability of the spilled contaminant and to design comprehensive emergency response schemes, it is essential to estimate the contaminant source characteristic and identify where, when and how much the spilled contaminant is injected into a surface river. In this study, an effective pollution source inverse method is developed to reconstruct the release history of the injection location, time, and quantity, and provide appropriate emergency response schemes for dealing with surface river environmental pollution. The pollution source inverse method IGSAA is developed by an integration of genetic algorithm (IGA) and simulated annealing algorithm (SAA) in order to guarantee both the global searching ability and convergence speed. The pollution source inverse method IGSAA is then applied to a hypothetical study, comparing with the traditional GA-based and SAA-based methods, to verify the accuracy and efficiency of the contaminant source inverse, and to a trace study of Truckee River in west America to identify the contaminant source release history and characteristic under different scenarios. The pollution source inversion results can help decision-makers (DMs) to identify the contaminant source characteristics of a chemical spill, and carry out emergency disposal scheme for an emergency rescue in a quick response, and enhance the supervision and management ability for a real surface river system.

pth moment exponential stability of stochastic memristor-based bidirectional associative memory (BAM) neural networks with time delays.

Stochastic memristor-based bidirectional associative memory (BAM) neural networks with time delays play an increasingly important role in the design and implementation of neural network systems. Under the framework of Filippov solutions, the issues of the pth moment exponential stability of stochastic memristor-based BAM neural networks are investigated. By using the stochastic stability theory, Itô's differential formula and Young inequality, the criteria are derived. Meanwhile, with Lyapunov approach and Cauchy-Schwarz inequality, we derive some sufficient conditions for the mean square exponential stability of the above systems. The obtained results improve and extend previous works on memristor-based or usual neural networks dynamical systems. Four numerical examples are provided to illustrate the effectiveness of the proposed results.

Complex dynamics of a delayed discrete neural network of two nonidentical neurons.

In this paper, we discover that a delayed discrete Hopfield neural network of two nonidentical neurons with self-connections and no self-connections can demonstrate chaotic behaviors. To this end, we first transform the model, by a novel way, into an equivalent system which has some interesting properties. Then, we identify the chaotic invariant set for this system and show that the dynamics of this system within this set is topologically conjugate to the dynamics of the full shift map with two symbols. This confirms chaos in the sense of Devaney. Our main results generalize the relevant results of Huang and Zou [J. Nonlinear Sci. 15, 291-303 (2005)], Kaslik and Balint [J. Nonlinear Sci. 18, 415-432 (2008)] and Chen et al. [Sci. China Math. 56(9), 1869-1878 (2013)]. We also give some numeric simulations to verify our theoretical results.