Search

To build a behaviour decision method that can converge to stability in a complex environment, this paper constructs an intelligent learning algorithm which can quickly find the shortest path by simulating the memory replay in the hippocampus. During the environment identification, a dynamic top-k competition is designed so the agent can consider the influence of multiple obstacles. Meanwhile, a representation layer of the advanced characteristics is added to the algorithm, so the agent can adapt to the complex structural environment. In addition, simulating the memory playback in the hippocampus, through the reverse playback of memory sequence, the reward discount is introduced into the weight updating process of the motor neurons, making the agent understand the relation among the memory fragments and guarantee the global optimal decision. Physical experiments of a mobile robot show that the agent can find the shortest path, demonstrating the agent can make optimal decisions in a complex structural environment.

How to achieve flexible behavioral decision making in a dynamic environment is an important prerequisite for a mobile robot to execute various tasks. Traditional methods suffer from discontinuous and unstable learning ability. This article brings the mobile robot continuous and stable learning ability in the dynamic environment, through improving the binding manner between the cerebellum and basal ganglia in the neuromodulatory system proposed in our previous work, thus to enhance its flexible behavioral decision-making performance. Moreover, a more biological significance index, i.e., the curiosity index, $Cur$ , is designed to mimic the activity switch of the locus coeruleus between the tonic and phasic mode, to modulate the exploration-exploitation tradeoff in the reinforcement learning (RL) of the basal ganglia. Influence of varying learning rate and varying discount rate on the performance of the RL is also investigated. The experiment results in static and dynamic environments, as well as the comparative experiment in the complex maze environment and the complex dynamic environment, demonstrate the potential of the proposed neuromodulatory system.

Path planning is one of the key technologies of mobile robots. This paper applies the Q-learning algorithm to path planning of mobile robots. To solve the blindness of the mobile robot in exploring the environment, this paper combines RRT algorithm and improves it to increase the goal-orientated performance of the robot in exploring the environment. To overcome the slow convergence rate of the classical Q-learning, the exploration rate, discount rate, learning rate and other parameters in Q-learning are modified. A modified ant colony optimisation (ACO) algorithm is designed to make the robot consider the information of other robots and improve the precision of action decision. The hierarchical idea in the grey wolves optimisation algorithm is introduced into the modified ACO algorithm to realise the dynamic regulation of pheromone. Finally, the modified Dyna-2 algorithm is designed to enhance the generalisation of the Q-learning algorithm. The potential of the proposed hybrid intelligent algorithm is proved by the path planning experiments in two static complex environments.

For the global path planning of mobile robots in static environment, the traditional ant colony optimization (ACO) algorithm has some disadvantages, such as slow convergence speed and easy to fall into local optimum. In order to deal with these problems, a novel improved ACO algorithm is proposed in this paper. First, in order to accelerate the convergence of the traditional ACO algorithm, several coefficients, such as pheromone volatility coefficient, path feasible coefficient, pheromone eliciting factor, expectation eliciting factor, memory parameter, social phobia parameter, personality factor and emotion fluctuation coefficient, were designed in the improved ACO algorithm. Meanwhile, in order to decrease the probability of the traditional ACO algorithm falling into the local optimum, a specific parameter is designed to make the ants evolve continuously, so the ants can reach the target by the shortest path. Finally, three simulation experiments are carried out in static environment to verify the potential of improved ant colony algorithm in mobile robot path planning.

During the perceptual learning (PL) of a mobile robot, how to improve the efficiency of its behavior decision is a great challenge. The existing methods suffer from the less flexibility and low efficiency. This article proposes a novel methodology to address these problems. It mimics the principle that the human’s brain can still think, even when he/she neither receives any sensory stimulus nor performs any action, during the off-task process, a gated self-organization mechanism is used to trigger the “brain” of the mobile robot to work, analyze the scenarios that it encountered in the former tasks to decide the best moving direction, then store the sensed scenario information by means of the lateral excitation of the internal neurons in a developmental network (DN), and set up the weight connections from the internal area to high-level decision-making area of the DN. In the following PL, if the robot encounters a similar scenario, it can make a better and faster behavior decision. Therefore, this methodology can greatly enhance the efficiency of behavior decision with the limited training samples, without retraining in facing a new scenario. Most importantly, this methodology makes the robot continuously improve its intelligence through the autonomous learning, even in offline state. Extensive simulation and experiment results of the mobile robot PL demonstrate its potential. Since it is task nonspecific, the same learning principles are potentially suitable for other fields. To the best of our knowledge, it is the first trial to apply the transfer learning through the lateral excitation mechanism of internal neurons to the robot field with the emergent representation.

This paper proposes an event-triggered adaptive fuzzy asymptotic predefined-time controller for uncertain nonlinear systems (UNS) subject to external disturbances. By introducing a set of well-designed asymptotic performance adjustment (APA) functions, we establish a general framework for analyzing asymptotic predefined-time stability (APDTS). Under the control scheme based on the established framework, the system state is guaranteed to converge to a controllable area around zero within a user-defined time and eventually asymptotically converge to zero. Finally, numerical simulations of a robotic arm system verify the effectiveness and superiority of control approaches. • Asymptotic predefined time stability theorem • Improved event triggering strategies • New controllers for uncertain nonlinear systems [ABSTRACT FROM AUTHOR]

Online model-free reinforcement learning (RL) approaches play a crucial role in coping with the real-world applications, such as the behavioral decision making in robotics. How to balance the exploration and exploitation processes is a central problem in RL. A balanced ratio of exploration/exploitation has a great influence on the total learning time and the quality of the learned strategy. Therefore, various action selection policies have been presented to obtain a balance between the exploration and exploitation procedures. However, these approaches are rarely, automatically, and dynamically regulated to the environment variations. One of the most amazing self-adaptation mechanisms in animals is their capacity to dynamically switch between exploration and exploitation strategies. This article proposes a novel neurophysiologically motivated model which simulates the role of medial prefrontal cortex (MPFC) and lateral prefrontal cortex (LPFC) in behavior decision. The sensory input is transmitted to the MPFC, then the ventral tegmental area (VTA) receives a reward and calculates a dopaminergic reinforcement signal, and the feedback categorization neurons in anterior cingulate cortex (ACC) calculate the vigilance according to the dopaminergic reinforcement signal. Then the vigilance is transformed to LPFC to regulate the exploration rate, finally the exploration rate is transmitted to thalamus to calculate the corresponding action probability. This action selection mechanism is introduced to the actor–critic model of the basal ganglia, combining with the cerebellum model based on the developmental network to construct a new hybrid neuromodulatory model to select the action of the agent. Both the simulation comparison with other four traditional action selection policies and the physical experiment results demonstrate the potential of the proposed neuromodulatory model in action selection.

During the movement of the mobile robot, except the common scenario, sometimes, the robot has to face the unexpected uncertainty in the environment. Traditional methods to address this problem generally use the task-specific method from the engineering perspective, lacking flexibility in the changing environment and being difficult to respond to the environmental challenges. Because of the function of the brain’s neuromodulatory system, human beings have ability to respond to the ever-changing environment quickly. To simulate the working mechanism of the human brain in responding to the unexpected uncertainty in the environment, this article presents a motivated developmental network (MDN) to offer a control configuration for an artificial agent to face the unexpected uncertainty in the environment, through introducing the acetylcholine/norepinephrine (ACh/NE) systems to the MDN. Taking the regulation role of ACh/NE in serotonin and dopamine (DA) into account, a novel learning rate for the hidden layer neurons of the MDN is proposed. Moreover, a novel composite mechanism is presented to decide the moving direction of the agent. Under the modulation of DA, serotonin, ACh, and NE, the agent can perform specific functions effectively, e.g., chase a target and elude the obstacle, especially the sudden obstacle. Goal-directed pursuing behavior in three simulation cases illustrates the effect of the presented neural modulatory systems, for instance, dealing with the unexpected uncertainty, realizing the attentional effort, reinforcement learning, etc. To the best of our knowledge, this article is the first endeavor to address the unexpected uncertainty with the MDN.

ABSTRACTSpeech recognition performance of the machine has been greatly improved using artificial intelligence. However, compared with the superior recognition ability of human auditory system, the machine still has some problems to deal with. Based on the existing physiological principle of human auditory system, this paper proposes a novel emergent auditory model. This model simulates each key part of the human auditory pathway with a deep developmental network (DDN). Furthermore, this model simulates the function of the superior colliculus in the thalamus, i.e., context integration, as an additional layer in the DDN. Mel-frequency cepstral coefficients (MFCC) are used to extract the speech signal features to be inputs of the DDN. This work is different from other previous models as we emphasise the mechanism that makes a system to develop its emergent representations from its operational experience, i.e., the internal unsupervised neurons of the DDN are utilised to depict the short contexts, and competitions among them afford an interpretation of how such internal neurons denote the different speech contexts when they are not supervised by the external world. Experimental results show the advantage of the proposed DNN compared to the state-of-the-art methods for the recognition accuracies of English words and phrases.

Face recognition is one of the core and challenging issues in computer vision field. Compared to computer vision, human visual system can identify a target from complex backgrounds quickly and accurately. This paper proposes a new network model deriving from Where-What Networks (WWNs), which can approximately simulate the information processing pathways (i.e., dorsal pathway and ventral pathway) of human visual cortex and recognize different types of faces with different locations and sizes in complex background. To enhance the recognition performance, synapse maintenance mechanism and neuron regenesis mechanism are both introduced. Synapse maintenance is used to reduce the background interference while neuron regenesis mechanism is introduced to regulate the neuron resource dynamically to improve the network usage efficiency. Experiments have been conducted on human face images of 5 types, 11 sizes, and 225 locations in complex backgrounds. Experiment results demonstrate that the proposed WWN model can basically learn three concepts (type, location and size) simultaneously. The experiment results also show the advantages of the enhanced WWN-7 model for face recognition in comparison with several existing methods.

High-throughput screening and fast identification of single bacterial cells are crucial for clinical diagnosis, bioengineering, and fermentation engineering. Although single-cell technologies have been developed extensively in recent years, the single-cell technologies for bacteria still need further exploration. In this study, we demonstrate an identification and screening technology for single bacterial cells based on a large-scale nanobowl array, which is well-ordered and size-adjustable for use with different kinds of bacteria. When the culture medium with monodispersed bacteria was placed on the nanobowl array, it successfully enabled loading of single bacterium into a single nanobowl. Because of the limitative size and depth of the nanobowls, mixture of different bacteria species could be screened according to their sizes. In addition, with the help of a low electrical current, the bacteria can be further screened according to their intrinsic surface charges. If combined with micromanipulation technology, high-throughput single bacterial selection can be achieved in future.

ABSTRACTBrucellosis, caused by Brucella, is a severe zoonosis, and the current Brucellalive attenuated vaccine cannot be used in humans due to major safety risks. Although polysaccharide antigens can be used to prepare the Brucellavaccine, their lower immunogenicity limits them from producing efficient and broad protection. In this study, we produced a high-performance bioconjugate nanovaccine against different species of Brucellaby introducing a self-assembly nanoparticle platform and an O-linked glycosylation system into Yersinia enterocoliticaserotype O:9, which has an O-polysaccharide composed of the same unit as Brucella. After successfully preparing the vaccine and confirming its stability, we subsequently demonstrated the safety of the vaccine in mice by high-dose immunization. Then, by a series of mouse experiments, we found that the nanovaccine greatly promoted antibody responses. In particular, the increase of IgG2a was more obvious than that of IgG1. Most importantly, this nanovaccine could provide cross-protection against B. abortus, B. melitensis, and B. suisstrains by lethal dose challenged models, and could improve the clearance of B. melitensis, the most common pathogenic species in human brucellosis, by non-lethal dose infection. Overall, for the first time, we biocoupled polysaccharide antigens with nano carriers to prepare a Brucellavaccine, which showed pronounced and extensive protective effects in mice. Thus, we provided a potential candidate vaccine and a new direction for Brucellavaccine design.

Estimation of face orientation has been a topic of intense research in the recent past. Most of the prior face orientation methods use symbolic methods or handcrafted internal representations which are not sufficiently brain based. An emergent developmental network (DN) is presented to recognize the face orientation, from sensory and motor experience. This work is different in the sense that we focused on mechanisms that enable a system to develop its emergent representations from its operational experience. In this work, internal unsupervised neurons of the DN are used to represent the face orientation, and the competitions among the internal neurons enable them to represent different face orientations. To illustrate the recognition effect, we study and compare the recognition effects among the BP, LVQ, PNN, and DN. Experiment results demonstrate efficiently how such internal neurons represent the face orientation while they are not directly supervised by the external environment. The presented network is developmental which means that the internal representations are directly learned from the signals of the input and motor ports, not designed internally for particular task; hence, the same learning principles are potentially suitable for other sensory modalities. As far as we know, it is the first trial to use the DN to recognize the face orientation.

Many methods for reinforcement learning use symbolic representations—nonemergent—such as Q-learning. We use emergent representations here, without human handcrafted symbolic states (i.e., each state corresponds to a different location). This paper models reinforcement learning for hidden neurons in emergent networks for sequential tasks. In this paper, their influences on sequential tasks (e.g., robot navigation in different scenarios) are investigated where the learned value and results of a behavior rely on not only the current experience just like in a pattern recognition (episodic) but also the prediction of future experiences (e.g., delayed rewards) and environments (e.g., previously learned navigational trajectories). We show that this new model of motivated learning amounts to the computation of the maximum-likelihood estimate through “life” where punishment and reward have increased weights. This new formulation avoids the greediness of time-discount in Q-learning. Its complex nonlinear sequential optimization has been solved in a closed-form procedure under the condition of the limited computational resources and limited learning experience so far, because we convert it into a simpler problem of incremental and linear estimation. The experimental results showed that the serotonin and dopamine systems speed up learning for sequential tasks, because not all events are equally important. As far as we know, this is the first work that studies the influences of reinforcers (via serotonin and dopamine) on hidden neurons (Y neurons) for sequential tasks in dynamic scenarios using emergent representations. [ABSTRACT FROM AUTHOR]

In this issue, the robust synchronization for a class of uncertain Cohen–Grossberg neural networks is studied, in which neuron activations are modelled by discontinuous functions(or piecewise continuous functions). Pinning state-feedback and adaptive controllers are designed to achieve global robust exponential synchronization and global robust asymptotical synchronization of drive-response-based discontinuous Cohen–Grossberg neural networks. By applying the theory of non-smooth analysis theory and the method of generalized Lyapunov functional, some criteria are given to show that the coupled discontinuous Cohen–Grossberg neural networks with parameter uncertainties can realized global robust synchronization. Some examples and numerical simulations are also shown to verify the validity of the proposed results. [ABSTRACT FROM AUTHOR]

This paper is concerned with the dissipativity and synchronization problems of a class of delayed bidirectional associative memory (BAM) neural networks in which neuron activations are modeled by discontinuous bivariate functions. First, the concept of the Filippov solution is extended to functional differential equations with discontinuous right-hand sides and mixed delays via functional differential inclusions. The global dissipativity of the Filippov solution to the considered BAM neural networks is proven using generalized Halanay inequalities and matrix measure approaches. Second, to realize global exponential complete synchronization of BAM neural networks with multivariate discontinuous activations, discontinuous state feedback controllers are designed using functional differential inclusions theory and nonsmooth analysis theory with generalized Lyapunov functional method. Finally, several numerical examples are provided to demonstrate the applicability and effectiveness of our proposed results. [ABSTRACT FROM AUTHOR]

This paper investigates the issue of global exponential synchronization for a class of general neural networks that contains discontinuous activation functions and mixed time delays. Functional differential inclusions and nonsmooth analysis theories are used as bases to design discontinuous controllers, such that the discontinuous neural networks can be exponential complete synchronized. This novel approach and its applicability to neural networks with continuous activations are also easily verified. Several numerical examples demonstrate the practicality and effectiveness of the design method. [ABSTRACT FROM AUTHOR]

This article investigates the fixed-time stability (FXTS) of discontinuous impulsive systems. Based on the generalized Lyapunov functional (LF) method and some inequality techniques, we propose some novel FXTS criteria of the systems. It is worth mentioning that the derivative of LF can be negative definite or indefinite, and these results extend the previous results significantly. Furthermore, by using the monotonicity of the integral functions and the existence theorem of zero points, we obtain the FXTS results under the influence of impulses. In particular, when taking some certain functions, we can easily estimate the settling-time. Finally, the main results are confirmed by numerical simulations. • Fixed time stability is possible when the derivative of the Lyapunov function is indefinite. • Parameters of the inequality satisfied by Lyapunov function are relaxed to be time-varying functions. • Integrals of coefficient functions can be estimated with non-linear functions. • Settling time is irrelevant to the initial state values. [ABSTRACT FROM AUTHOR]

Natural language acquisition is a crucial research domain of artificial intelligence. To enable the computer acquire and understand the human language and generate the corresponding response, we must research the principle of language acquiring of human being. Most of the prior language acquisition methods use handcrafted internal representation which is not sufficiently brain-based. An emergent developmental network (DN) is presented to acquire the certain speech extracted by mel-frequency cepstrum coefficient (MFCC), from sensory and motor experience. This work is different in the sense that we focused on mechanisms that enable a system to develop its emergent representations from its operational experience. In this work, internal unsupervised neurons of the DN are used to represent the short contexts, and the competitions among the internal neurons enable them to represent different short contexts. To demonstrate the acquisition effect, we study and analyze the influences of different network structure (i.e., different neuron number and weight threshold) on the language acquisition rate. Four speech acquisition experiments demonstrate efficiently how such internal neurons represent the short context while they are not directly supervised by the external environment. The presented network is developmental which means that the internal representations are directly learned from the signals of the input and motor ports, not designed internally for particular task, hence the same learning principles are potentially suitable for other sensory modalities.

This paper investigates the finite-time stabilization problem for a general class of memristor-based neural networks (MNNs). Firstly, based on set-valued analysis and Kakutani’s fixed point theorem of set-valued maps, the existence of equilibrium point can be guaranteed for MNNs. Then, by designing novel discontinuous controller, some sufficient conditions are proposed to stabilize the states of such MNNs in finite time. Moreover, we give the upper bound of the settling time for stabilization which depends on the system parameters and control gains. The main tools to be used involve the framework of Filippov differential inclusions, non-smooth analysis, matrix theory and the famous finite-time stability theorem of nonlinear system. Finally, the theoretical results are verified by concrete examples with computer simulations.

Rare-earth (RE) element addition can remarkably improve the mechanical properties of magnesium alloys through precipitation hardening. The morphology, distribution and crystal structure of precipitates are regarded as major strengthening mechanisms in the Mg-RE alloys. In order to understand the formation of precipitates during aging at 225 °C in a Mg-10Gd-3Y-0.4Zr alloy (GW103K) with high strength and heat resistance, a high-resolution transmission electron microscopy (HRTEM) was employed to characterize the microstructural evolution. It was found that three types of precipitates were observed in the alloy at the early stage, named as: single layer D019 structure, one single layer D019 structure and one layer of Mg, two parallel single layers (containing RE) and Mg layer in between, which was regarded as ordered segregation of RE, precursors to form β″ and β′ phase, respectively. Both of β″ and β′ phase were transformed from the precursors. It was also found that large size of β′ phase and the small size of β″ phase were constantly existent in the whole aging process. β′ phase played a major role in the strengthening of the GW103K alloys and the decrease of the hardness was caused by the coarsening of β′ phase.

In this paper, we consider a delayed n + m-species competition predator-prey system on time scales with Holling III functional response and multiple exploited (or harvesting) terms. By using the continuation theorem based on Gaines and Mawhin’s coincidence degree theory, easily verifiable criteria are established for global existence of multiple positive periodic solutions for the above system.

Simultaneous environment exploration and map building of a mobile robot in an unknown environment are studied. Based on the real time data acquired from a laser sensor, a suitable environment exploration strategy with obstacle avoidance ability is proposed. To handle the problems existing in producing and evaluating candidates, feasible approaches are proposed. They can maximise the expected information gain and keep the environment information integrated, and ensure the environment exploration's continuum and complete traversal. The new evaluation method overcomes the drawbacks (attend to one criterion and lose another) of the traditional weighted average method. It can comprehensively evaluate the travelling cost, expected information gain and rotating angle to guarantee the quality of the optimal candidate. Furthermore, a topological map model is proposed which uses the nodes of the growing neural gas network as the topological network nodes. Through the growing characteristic of the GNG network, new topological nodes are added into the network to abstract and express the holistic knowledge of the surrounding environment and construct the environment map. Simulation results of two different indoor environments demonstrate its effectiveness and feasibility.

In this paper, we consider a delayed predator–prey system on time scales with nonmonotonic functional response and multiple exploited (or harvesting) terms. Using the continuation theorem based on Gaines and Mawhin’s coincidence degree theory, easily verifiable criteria are established for global existence of multiple periodic solutions to the above system on time scales. Moreover, our results generalize and improve the main results of Hu and Zhang (Comput Math Appl 56:3015–3022, 2008).

Highlights: [•] Delayed CGNN with discontinuous right-hand sides was investigated. [•] The global exponential stability of almost periodic solution to the system was obtained. [•] The assumptions on monotonicity of the discontinuous activation has been eliminated. [•] The restriction condition has also been eliminated. [ABSTRACT FROM AUTHOR]

Selection of optimal measurement configurations for the kinematic parameter calibration is investigated in this paper. To minimise the effects of measurement and modelling errors on parameter estimation, a series of optimal calibration configurations is obtained with our modified simulated annealing algorithm. To accelerate the convergence rate, a suitable cooling schedule is designed. Independent and interactive effects of the parameters in the cooling schedule on algorithm performance are discussed. Two observability indices are optimised with the modified simulated annealing algorithm to obtain the optimal measurement configurations of the 5-DOF polishing robot. Simulated calibration shows that these optimal configurations improve the calibration accuracy significantly. Experimental results, compared with those of random pose selection and a local search algorithm, are presented to demonstrate the feasibility of the proposed approach.

Shigella flexneri, which is closely related to Escherichia coli, is the most common cause of the endemic form of shigellosis. In this study, 53 homomultimeric protein complexes and nine heteromultimeric protein complexes from S. flexneri2a strain 2457T were separated and identified. Among these, three potential homomultimeric protein complexes had not been previously described. One complex, thought to be composed of 12 PhoN1 subunits, is a periplasmic protein with an unknown physiological role encoded on the virulence plasmid of S. flexneri. The abundance of the protein complexes was compared following growth at 37 or 30°C, and the abundance of three protein complexes (PyrB-PyrI, GlmS, and MglB) related to the synthesis of lipopolysaccharides (LPS) appeared to be temperature-dependent. Many studies have shown that LPS is essential to the virulence of S. flexneri. Here, we report the influence of temperature on the amount of LPS.

• In order to reestablish the random initialization input weights of ESN, a single layer ESN with a global reversible AE (GRAE) algorithm is proposed to ensure that the outputs of GRAE are remarkably correlated with the input data. • The reservoir layer with a reversible function is calculated by pulling the ESN output back and injecting it into the reservoir layer. Thus, the feature learning is enriched by additional information, which results in a better performance. • The GRAE which is based on ridge regression to solve a least squares problem is entirely different from existing back-propagation (BP) based AEs. So, the training speed of GRAE is many times faster than that of the BP based AEs. • In existing ESN autoencoder methods, all weights in the reservoir layer are generated randomly. The simple cycle connection is utilized in the reservoir layer of ESN to make sure the reservoir weights can be generated deterministically. An echo state network (z) can provide an efficient dynamic solution for predicting time series problems. However, in most cases, ESN models are applied for predictions rather than classifications. The applications of ESN in time series classification (TSC) problems have yet to be fully studied. Moreover, the conventional randomly generated ESN is unlikely to be optimal because of the randomly generated input and reservoir weights, which are not always guaranteed to be optimal. Randomly generating all layer weights is improper, because a purely random layer might destroy the useful features. To overcome this disadvantage, this study provides a new input weight establishment framework of ESN based on autoencoder (AE) theory for TSC tasks. A global reversible AE (GRAE) algorithm is proposed to reestablish the random initialization input weights of the ESN. In existing ESN-AEs, the output weights obtained in the encoding process are directly reused as the initial input weights. By contrast, in GRAE, the reservoir layer with a reversible activation function is calculated by pulling the decoding layer output back and injecting it into the reservoir layer. Thus, feature learning is enriched by additional information, which results in improved performance. The current weights of the encoding layer are iteratively replaced by the decoding layer to ensure that the outputs of the GRAE are remarkably correlated with the input data. Visualization analyses and experiments of the input weights on a massive set of UCR time series datasets indicate that the proposed GRAE method can considerably improve the original two-layer ESN-based classifiers and the proposed GRAE-ESN classifier yields better performance compared with traditional state-of-the-art TSC classifiers. Furthermore, the proposed method can provide comparable performance and considerably faster training speed compared with three deep learning classifiers. [ABSTRACT FROM AUTHOR]

ABSTRACT Conjugate vaccines are known to be one of the most effective and safest types of vaccines against bacterial pathogens. Previously, vaccine biosynthesis has been performed by using N-linked glycosylation systems. However, the structural specificity of these systems for sugar substrates has hindered their application. Here, we report a novel protein glycosylation system (O-linked glycosylation via Neisseria meningitidis) that can transfer virtually any glycan to produce a conjugate vaccine. We successfully established this system in Shigellaspp., avoiding the construction of an expression vector for polysaccharide synthesis. We further found that different protein substrates can be glycosylated using this system and that the O-linked glycosylation system can also effectively function in other Gram-negative bacteria, including some strains whose polysaccharide structure was not suitable for conjugation using the N-linked glycosylation system. The results from a series of animal experiments show that the conjugate vaccine produced by this O-linked glycosylation system offered a potentially protective antibody response. Furthermore, we elucidated and optimized the recognition motif, named MOOR, for the O-glycosyltransferase PglL. Finally, we demonstrated that the fusion of other peptides recognized by major histocompatibility complex class II around MOOR had no adverse effects on substrate glycosylation, suggesting that this optimized system will be useful for future vaccine development. Our results expand the glycoengineering toolbox and provide a simpler and more robust strategy for producing bioconjugate vaccines against a variety of pathogens.IMPORTANCERecently, the rapid development of synthetic biology has allowed bioconjugate vaccines with N-linked protein glycosylation to become a reality. However, the difficulty of reestablishing the exogenous polysaccharide synthetic pathway in Escherichia colihinders their application. Here, we show that an O-linked protein glycosylation system from Neisseria meningitidis, which has a lower structure specificity for sugar substrates, could be engineered directly in attenuated pathogens to produce effective conjugate vaccines. To facilitate the further design of next-generation bioconjugate vaccines, we optimized a novel short motif consisting of 8 amino acids that is sufficient for glycosylation. Our results expand the application potential of O-linked protein glycosylation and demonstrate a simpler and more robust strategy for producing bioconjugate vaccines against different pathogens. In the future, bacterial antigenic polysaccharides could be attached to major histocompatibility complex binding peptides to improve immunological memory or attached to protein subunit vaccine candidates to provide double immune stimulation.

In this paper, we consider the possibility to implement the periodic synchronization control of memristive neural networks with time-varying delay. On the one hand, we apply a new method based on Kakutani׳s fixed point theorem of set-valued maps to obtain the existence of periodic solutions for memristive neural networks. On the other hand, by designing novel discontinuous state-feedback controller and introducing some new analytic techniques, the drive–response structure of memristor-based neural networks with periodic coefficients can realize the p -type exponential complete synchronization. Moreover, the estimated rate of exponential synchronization depends on the time delay and system parameters. The analysis in this paper is performed via functional differential inclusions following the Filippov theory. Finally, the proposed design method and theoretical results are verified by numerical example with MATLAB programming. [ABSTRACT FROM AUTHOR]

This paper concerns a reaction diffusion epidemic model under logistic source. First, we explore uniform boundedness of the system. Moreover, by constructing two Lyapunov functions, we show that the disease-free equilibrium of system is globally asymptotically stable and the large time dynamics of constant endemic equilibrium is also considered. Especially, we further investigate the convergence rate of the nonnegative globally bounded classical solution. Finally, two numerical simulations are displayed to confirm the results. [ABSTRACT FROM AUTHOR]