Explore the vast range of titles available.

Heterologous expression is the main approach for recombinant protein production ingenetic synthesis, for which codon optimization is necessary. The existing optimization methods are based on biological indexes. In this paper, we propose a novel codon optimization method based on deep learning. First, we introduce the concept of codon boxes, via which DNA sequences can be recoded into codon box sequences while ignoring the order of bases. Then, the problem of codon optimization can be converted to sequence annotation of corresponding amino acids with codon boxes. The codon optimization models for Escherichia Coli were trained by the Bidirectional Long-Short-Term Memory Conditional Random Field. Theoretically, deep learning is a good method to obtain the distribution characteristics of DNA. In addition to the comparison of the codon adaptation index, protein expression experiments for plasmodium falciparum  candidate vaccine and polymerase acidic protein were implemented for comparison with the original sequences and the optimized sequences from Genewiz and ThermoFisher. The results show that our method for enhancing protein expression is efficient and competitive.

Capsule networks (CapsNets), a new class of deep neural network architectures proposed recently by Hinton et al., have shown a great performance in many fields, particularly in image recognition and natural language processing. However, CapsNets have not yet been applied to drug discovery-related studies. As the first attempt, we in this investigation adopted CapsNets to develop classification models of hERG blockers/nonblockers; drugs with hERG blockade activity are thought to have a potential risk of cardiotoxicity. Two capsule network architectures were established: convolution-capsule network (Conv-CapsNet) and restricted Boltzmann machine-capsule networks (RBM-CapsNet), in which convolution and a restricted Boltzmann machine (RBM) were used as feature extractors, respectively. Two prediction models of hERG blockers/nonblockers were then developed by Conv-CapsNet and RBM-CapsNet with the Doddareddy's training set composed of 2,389 compounds. The established models showed excellent performance in an independent test set comprising 255 compounds, with prediction accuracies of 91.8 and 92.2% for Conv-CapsNet and RBM-CapsNet models, respectively. Various comparisons were also made between our models and those developed by other machine learning methods including deep belief network (DBN), convolutional neural network (CNN), multilayer perceptron (MLP), support vector machine (SVM), k-nearest neighbors (kNN), logistic regression (LR), and LightGBM, and with different training sets. All the results showed that the models by Conv-CapsNet and RBM-CapsNet are among the best classification models. Overall, the excellent performance of capsule networks achieved in this investigation highlights their potential in drug discovery-related studies.

Abstract Autism is a common neurodevelopmental disorder that severely affects patients’ quality of life. We aimed to investigate whether acupuncture at Zusanli (ST36) could alleviate the behavior disorder of autistic rats by inhibiting thioredoxin-interacting protein (TXNIP)-mediated activation of NLRP3. An autism model was induced by intraperitoneal injection of pregnant rats with valproic acid (VPA). The pups’ behaviors were analyzed using hot plate, open field, Morris water maze, and 3-chamber social interaction tests. Nissl staining was used to visualize neurons in prefrontal cortex. Levels of TXNIP, NLRP3, interleukin (IL)-1β, and caspase were determined by Western blot or quantitative real-time PCR. After ST36 acupuncture, pain sensitivity, autonomous activity, sociability index, sociability preference index, and learning and memory were improved in the autism model rats. Levels of TXNIP, NLRP3, IL-1β, and caspase 1 were decreased after acupuncture. Interference with TXNIP alleviated the behavior disorders and inhibited NLRP3, caspase 1, and IL-1β levels. In summary, ST36 acupuncture reduced TXNIP expression, inhibited the activation of the NLRP3 inflammasome, and alleviated the behavior disorder related to the prefrontal cortex of the autistic rats. These results point to a potential mechanism for acupuncture-induced improvement of autistic behavioral disorders.

Sentiment analysis (SA) is an important research area in cognitive computation—thus, in-depth studies of patterns of sentiment analysis are necessary. At present, rich-resource data-based SA has been well-developed, while the more challenging and practical multi-source unsupervised SA (i.e., a target-domain SA by transferring from multiple source domains) is seldom studied. The challenges behind this problem mainly locate in the lack of supervision information, the semantic gaps among domains (i.e., domain shifts), and the loss of knowledge. However, existing methods either lack the distinguishable capacity of the semantic gaps among domains or lose private knowledge. To alleviate these problems, we propose a two-stage domain adaptation framework. In the first stage, a multi-task methodology-based shared-private architecture is employed to explicitly model the domain-common features and the domain-specific features for the labeled source domains. In the second stage, two elaborate mechanisms are embedded in the shared-private architecture to transfer knowledge from multiple source domains. The first mechanism is a selective domain adaptation ( SDA ) method, which transfers knowledge from the closest source domain. And the second mechanism is a target-oriented ensemble ( TOE ) method, in which knowledge is transferred through a well-designed ensemble method. Extensive experiment evaluations verify that the performance of the proposed framework outperforms unsupervised state-of-the-art competitors. What can be concluded from the experiments is that transferring from very different distributed source domains may degrade the target-domain performance, and it is crucial to choose proper source domains to transfer from.

The aim of the study reported here was to evaluate whether the mitochondrial ATP-sensitive potassium (mitoK ATP ) channel could participate in the effect of dexmedetomidine on cerebral ischemia-reperfusion (I/R) rats. Forty rats were randomly assigned into 5 groups: sham operation (S) group; cerebral I/R group; dexmedetomidine (D) group; 5-hydroxydecanoate (5-HD) group; 5-HD + D group. The cerebral I/R were produced by 2 h right middle cerebral artery occlusion followed by 24 h reperfusion. Dexmedetomidine (50μg/kg) was injected intraperitoneally before ischemia and after the onset of reperfusion. 5-HD (30 mg/kg) was injected intraperitoneally at 1 h before ischemia. The neurological deficit score (NDS) and the levels of super oxide dismutase (SOD), malondialdehyde (MDA), myeloperoxidase (MPO), Interleukin 6 (IL-6) and tumor necrosis factor-α (TNF-α) were evaluated. Compared to group S, NDS and the levels of MDA, MPO, IL-6 and TNF-α were significantly higher, and SOD levels were significantly lower in the other groups ( P  < 0.05). Compared to group I/R,NDS and the levels of MDA, MPO, IL-6 and TNF-α were significantly lower, and SOD level was significantly higher in group D ( P  < 0.05). Compared to group D, NDS and the levels of MDA, MPO, IL-6 and TNF-α were significantly higher, and SOD level was significantly lower in group5-HD + D ( P  < 0.05). The activation of the mitoK ATP channel could contribute to the protective effect of dexmedetomidine on rats induced by focal cerebral ischemia-reperfusion injury.

The aim of the present study was to evaluate the effect of a combination of dexmedetomidine and fentanyl on peripheral oxygen saturation (SpO2) and hemodynamic stability in patients undergoing flexible bronchoscopy. One hundred patients undergoing elective flexible bronchoscopy were randomized into either a propofol-fentanyl group (PF group; n=50) or a dexmedetomidine-fentanyl group (DF group; n=50). SpO2 values, heart rate (HR), systolic and diastolic blood pressure (SBP and DBP), patients' cough scores and discomfort scores as determined by patients and bronchoscopists, levels of sedation, number of times that additional lidocaine was required, elapsed time until recovery, and adverse events were recorded. The mean SpO2 values in the DF group were significantly higher than those in the PF group (P<0.01), and HR, SBP and DBP were significantly lower in the DF group than in the PF group (P<0.05). There were no statistically significant differences between the two groups in terms of cough scores or discomfort scores, sedation levels, or number of times that additional lidocaine was required (P>0.05). Elapsed time until recovery in the DF group was significantly longer than in the PF group (P=0.002). The incidence of hypoxemia was significantly lower in the DF group than in the PF group (P=0.027), but the incidence of bradycardia was significantly higher in the DF group than in the PF group (P=0.037). Dexmedetomidine-fentanyl was superior to propofol-fentanyl in providing satisfactory SpO2. Furthermore, dexmedetomidine-fentanyl attenuated hemodynamic responses during bronchoscopy and maintained hemodynamic stability in the early stage of the procedure.

In the context of online education, designing an automatic solver for geometric problems has been considered a crucial step towards general math Artificial Intelligence (AI), empowered by natural language understanding and traditional logical inference. In most instances, problems are addressed by adding auxiliary components such as lines or points. However, adding auxiliary components automatically is challenging due to the complexity in selecting suitable auxiliary components especially when pivotal decisions have to be made. The state-of-the-art performance has been achieved by exhausting all possible strategies from the category library to identify the one with the maximum likelihood. However, an extensive strategy search have to be applied to trade accuracy for ef-ficiency. To add auxiliary components automatically and efficiently, we present deep reinforcement learning framework based on the language model, such as BERT. We firstly apply the graph attention mechanism to reduce the strategy searching space, called AttnStrategy, which only focus on the conclusion-related components. Meanwhile, a novel algorithm, named Automatically Adding Auxiliary Components using Reinforcement Learning framework (A3C-RL), is proposed by forcing an agent to select top strategies, which incorporates the AttnStrategy and BERT as the memory components. Results from extensive experiments show that the proposed A3C-RL algorithm can substantially enhance the average precision by 32.7% compared to the traditional MCTS. In addition, the A3C-RL algorithm outperforms humans on the geometric questions from the annual University Entrance Mathematical Examination of China.

Sentiment analysis (SA) is an important research area in cognitive computation-thus in-depth studies of patterns of sentiment analysis are necessary. At present, rich resource data-based SA has been well developed, while the more challenging and practical multi-source unsupervised SA (i.e. a target domain SA by transferring from multiple source domains) is seldom studied. The challenges behind this problem mainly locate in the lack of supervision information, the semantic gaps among domains (i.e., domain shifts), and the loss of knowledge. However, existing methods either lack the distinguishable capacity of the semantic gaps among domains or lose private knowledge. To alleviate these problems, we propose a two-stage domain adaptation framework. In the first stage, a multi-task methodology-based shared-private architecture is employed to explicitly model the domain common features and the domain-specific features for the labeled source domains. In the second stage, two elaborate mechanisms are embedded in the shared private architecture to transfer knowledge from multiple source domains. The first mechanism is a selective domain adaptation (SDA) method, which transfers knowledge from the closest source domain. And the second mechanism is a target-oriented ensemble (TOE) method, in which knowledge is transferred through a well-designed ensemble method. Extensive experiment evaluations verify that the performance of the proposed framework outperforms unsupervised state-of-the-art competitors. What can be concluded from the experiments is that transferring from very different distributed source domains may degrade the target-domain performance, and it is crucial to choose the proper source domains to transfer from.

At present, the great achievements of convolutional neural network(CNN) in feature and metric learning have attracted many researchers. However, the vast majority of deep network architectures have been used to represent based on real values. The research of complex-valued networks is seldom concerned due to the absence of effective models and suitable distance of complex-valued vector. Motived by recent works, complex vectors have been shown to have a richer representational capacity and efficient complex blocks have been reported, we propose a new approach for learning image descriptors with complex numbers to compare image patches. We also propose a new architecture to learn image similarity function directly based on complex-valued network. We show that our models can perform competitive results on benchmark datasets. We make the source code of our models publicly available.