Pattern recognition by primary and secondary response of an Artificial Immune System

Summary In this paper we show how an Artificial Immune System can be used to study pattern recognition processes and learning. In particular we show the ability of the model to discover and maintain coverage of the diverse patterns through mechanism of evolution and mutation.     

Network Intrusion Detection Based on a General Regression Neural Network Optimized by an Improved Artificial Immune Algorithm

To effectively and accurately detect and classify network intrusion data, this paper introduces a general regression neural network (GRNN) based on the artificial immune algorithm with elitist strategies (AIAE). The elitist archive and elitist crossover were combined with the artificial immune algorithm (AIA) to produce the AIAE-GRNN algorithm, with the aim of improving its adaptivity and accuracy. In this paper, the mean square errors (MSEs) were considered the affinity function. The AIAE was used to optimize the smooth factors of the GRNN; then, the optimal smooth factor was solved and substituted into the trained GRNN. Thus, the intrusive data were classified. The paper selected a GRNN that was separately optimized using a genetic algorithm (GA), particle swarm optimization (PSO), and fuzzy C-mean clustering (FCM) to enable a comparison of these approaches. As shown in the results, the AIAE-GRNN achieves a higher classification accuracy than PSO-GRNN, but the running time of AIAE-GRNN is long, which was proved first. FCM and GA-GRNN were eliminated because of their deficiencies in terms of accuracy and convergence. To improve the running speed, the paper adopted principal component analysis (PCA) to reduce the dimensions of the intrusive data. With the reduction in dimensionality, the PCA-AIAE-GRNN decreases in accuracy less and has better convergence than the PCA-PSO-GRNN, and the running speed of the PCA-AIAE-GRNN was relatively improved. The experimental results show that the AIAE-GRNN has a higher robustness and accuracy than the other algorithms considered and can thus be used to classify the intrusive data.     

Application of Artificial Bee Colony Algorithm to Maximum Likelihood DOA Estimation

Maximum Likelihood (ML) method has an excellent performance for Direction-Of-Arrival (DOA) estimation, but a multidimensional nonlinear solution search is required which complicates the computation and prevents the method from practical use. To reduce the high computational burden of ML method and make it more suitable to engineering applications, we apply the Artificial Bee Colony (ABC) algorithm to maximize the likelihood function for DOA estimation. As a recently proposed bio-inspired computing algorithm, ABC algorithm is originally used to optimize multivariable functions by imitating the behavior of bee colony finding excellent nectar sources in the nature environment. It offers an excellent alternative to the conventional methods in ML-DOA estimation. The performance of ABC-based ML and other popular meta-heuristic-based ML methods for DOA estimation are compared for various scenarios of convergence, Signal-to-Noise Ratio (SNR), and number of iterations. The computation loads of ABC-based ML and the conventional ML methods for DOA estimation are also investigated. Simulation results demonstrate that the proposed ABC based method is more efficient in computation and statistical performance than other ML-based DOA estimation methods.     

The superior fault tolerance of artificial neural network training with a fault/noise injectionbased genetic algorithm

Artificial neural networks (ANNs) are powerful computational tools that are designed to replicate the human brain and adopted to solve a variety of problems in many different fields. Fault tolerance (FT), an important property of ANNs, ensures their reliability when significant portions of a network are lost. In this paper, a fault/noise injection-based (FIB) genetic algorithm (GA) is proposed to construct fault-tolerant ANNs. The FT performance of an FIB-GA was compared with that of a common genetic algorithm, the back-propagation algorithm, and the modification of weights algorithm. The FIB-GA showed a slower fitting speed when solving the exclusive OR (XOR) problem and the overlapping classification problem, but it significantly reduced the errors in cases of single or multiple faults in ANN weights or nodes. Further analysis revealed that the fit weights showed no correlation with the fitting errors in the ANNs constructed with the FIB-GA, suggesting a relatively even distribution of the various fitting parameters. In contrast, the output weights in the training of ANNs implemented with the use the other three algorithms demonstrated a positive correlation with the errors. Our findings therefore indicate that a combination of the fault/noise injection-based method and a GA is capable of introducing FT to ANNs and imply that the distributed ANNs demonstrate superior FT performance.     

Adaptive Neural Network Structure Optimization Algorithm Based on Dynamic Nodes

Large-scale artificial neural networks have many redundant structures, making the network fall into the issue of local optimization and extended training time. Moreover, existing neural network topology optimization algorithms have the disadvantage of many calculations and complex network structure modeling. We propose a Dynamic Node-based neural network Structure optimization algorithm (DNS) to handle these issues. DNS consists of two steps: the generation step and the pruning step. In the generation step, the network generates hidden layers layer by layer until accuracy reaches the threshold. Then, the network uses a pruning algorithm based on Hebb’s rule or Pearson’s correlation for adaptation in the pruning step. In addition, we combine genetic algorithm to optimize DNS (GA-DNS). Experimental results show that compared with traditional neural network topology optimization algorithms, GA-DNS can generate neural networks with higher construction efficiency, lower structure complexity, and higher classification accuracy.     

Immune checkpoint blockade opens a new way to cancer immunotherapy

Among the main promising systems to triggering therapeutic antitumor immunity is the blockade of immune checkpoints. Immune checkpoint pathways regulate the control and eradication of infections, malignancies, and resistance against a host of autoantigens. Initiation point of the immune response is T cells, which have a critical role in this pathway. As several immune checkpoints are initiated by ligand–receptor interactions, they can be freely blocked by antibodies or modulated by recombinant forms of ligands or receptors. Antibodies against cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) were the first immunotherapeutics that achieved the US Food and Drug Administration approval. Preliminary clinical results with the blockers of additional immune checkpoint proteins, such as programmed cell death protein 1 (PD-1) indicate extensive and different chances to boost antitumor immunity with the objective of conferring permanent clinical effects. This study provides an overview of the immune checkpoint pathways, including CTLA-4, PD-1, lymphocyte activation gene 3, T-cell immunoglobulin and mucin domain 3, B7-H3, and diacylglycerol kinase α and implications of their inhibition in the cancer therapy.     

Microsatellite markers associated with quantitative trait loci controlling antibody response to Escherichia coli and Salmonella enteritidis in young broilers

A unique resource population was produced to facilitate detection of microsatellite markers associated with quantitative trait loci controlling antibody (Ab) response in broiler chickens. Three F1 males were produced by mating two lines divergently selected on Ab response to Escherichia coli vaccination. Each F1 male was mated with females from four genetic backgrounds: F1, high-Ab line (HH), low-Ab line and commercial line, producing three resource families, each with four progeny types. About 1700 chicks were immunized with E. coli and Salmonella enteritidis vaccines. Selective genotyping was conducted on the individuals with highest or lowest average Ab to E. coli and S. enteritidis within each progeny type in each sire family. Twelve markers were significantly associated with Ab to E. coli and six of them were also associated with Ab to S. enteritidis, mostly exhibiting a similar low effect (approximately 0.35 phenotypic SD) in all progeny types. Four markers exhibited a highly significant and much larger effect (approximately 1.7 SD), but only in progeny of females from the HH, suggesting that a backcross to the high parental line should be preferred over the commonly used F2 population. Results from two markers suggested a quantitative trait locus on chromosome 2 around 400 cM. The marker MCW0083, significant in two sire families, is closely linked to the bone morphogenetic protein 2 (BMP2) gene, known to be associated with the control of T-cell transformation in humans.     

A Linear Domain System Identification for Small Unmanned Aerial Rotorcraft Based on Adaptive Genetic Algorithm

<正> This paper proposes a new adaptive linear domain system identification method for small unmanned aerial rotorcraft.Byusing the flash memory integrated into the micro guide navigation control module, system records the data sequences of flighttests as inputs (control signals for servos) and outputs (aircraft's attitude and velocity information).After data preprocessing, thesystem constructs the horizontal and vertical dynamic model for the small unmanned aerial rotorcraft using adaptive geneticalgorithm.The identified model is verified by a series of simulations and tests.Comparison between flight data and the one-stepprediction data obtained from the identification model shows that the dynamic model has a good estimation for real unmannedaerial rotorcraft system.Based on the proposed dynamic model, the small unmanned aerial rotorcraft can perform hovering,turning, and straight flight tasks in real flight tests.     

一种构建高质量随机肽库的有效方法

为构建完全随机化的基因工程肽库 ,克服现有简并方案中终止密码子和序列组成偏歧的不足 ,提出了一种新的简并DNA文库合成方式。通过这种分组式合成方式构建的肽库可以避免终止密码子的出现和氨基酸组成偏歧的发生 ,还可以控制随机化过程中不同氨基酸的参入比例。以一个 13肽库的合成过程为例对分组式合成法进行了实验 ,测序结果和对 19种氨基酸出现频率的统计表明没有终止密码子和半胱氨酸密码子出现 ,各氨基酸的出现频率接近均值 ,表明这种分组 混合 分组 ,辅以简并合成的方法是行之有效的 ,能满足各类高容量基因工程随机肽库要求。     

An Improved Method with High Anti-interference Ability for R Peak Detection in Wearable Devices

The rapid development of the wearable electrocardiogram monitoring equipment increases the requirements for R peak detection in wearable devices. An improved method called ISC algorithm is proposed with high anti-interference ability for R peak detection in wearable devices based on a simple basic algorithm called SC algorithm. The proposed method is characterized by using the updated amplitude selection threshold, updated slope comparison threshold and RR interval judgement to reduce false positives and false negatives. For data from MIT-BIH Arrhythmia Database, the positive predictivity P+ of ISC algorithm can reach 99.12%, and the sensitivity Se of ISC algorithm is more than 95%. For MIT-BIH Noise Stress Test Database, the accuracy of ISC algorithm for both sensitivity Se and positive predictivity P+ can exceed 94% under three common noise, baseline wander, muscle artifact, and electrode motion artifact, where the positive predictivity P+ of ISC algorithm is 44.46% higher than that of SC algorithm on average. For wearable devices in exercise, even under the exercise intensity of 7 km per hour, the average positive predictivity P+ of ISC algorithm is 99.32%, which is 60.93% higher than that of SC algorithm. The high anti-interference ability shows that ISC algorithm is suitable for R peak detection in wearable devices.     

An Improved Rapid Method for Selecting Monoclonal Antibodies with High Catalytic Activities

Monoclonal antibodies were raised against a -naphthyl phosphonate hapten (1) to elicit antibodies capable of catalyzing the hydrolysis of -naphthyl acetate (3). After cell fusion, potential catalytic antibody-producing hybridomas were selected, by use of a competitive inhibition assay on the basis of the binding activity for a short transition-state analogue (inhibitor 5), followed by use of high-performance liquid chromatography analysis for the hybridoma supernatants to screen the antibodies processing catalytic activities. It was shown that supernatants of 12 wells had high binding activity with inhibitor and of them, 7 had catalytic activities. After cloning by limiting dilution, we got two hybridoma clones producing monoclonal antibodies which catalyzed the hydrolysis of -naphthyl acetate. This combination of competitive inhibition assay with high-performance liquid chromatography analysis represents an improved rapid approach for the screening of potential catalytic antibodies and significantly increases the possibility of obtaining efficient catalytic monoclonal antibodies. Further study of the catalytic antibodies revealed significant rate enhancement (K _cat/K _uncat 10⁶) and specificity.     

Implementing Dynamic Clamp with Synaptic and Artificial Conductances in Mouse Retinal Ganglion Cells

Ganglion cells are the output neurons of the retina and their activity reflects the integration of multiple synaptic inputs arising from specific neural circuits. Patch clamp techniques, in voltage clamp and current clamp configurations, are commonly used to study the physiological properties of neurons and to characterize their synaptic inputs. Although the application of these techniques is highly informative, they pose various limitations. For example, it is difficult to quantify how the precise interactions of excitatory and inhibitory inputs determine response output. To address this issue, we used a modified current clamp technique, dynamic clamp, also called conductance clamp ^{1, 2, 3} and examined the impact of excitatory and inhibitory synaptic inputs on neuronal excitability. This technique requires the injection of current into the cell and is dependent on the real-time feedback of its membrane potential at that time. The injected current is calculated from predetermined excitatory and inhibitory synaptic conductances, their reversal potentials and the cell''s instantaneous membrane potential. Details on the experimental procedures, patch clamping cells to achieve a whole-cell configuration and employment of the dynamic clamp technique are illustrated in this video article. Here, we show the responses of mouse retinal ganglion cells to various conductance waveforms obtained from physiological experiments in control conditions or in the presence of drugs. Furthermore, we show the use of artificial excitatory and inhibitory conductances generated using alpha functions to investigate the responses of the cells.     

On Indifference Zone Selection with a Preference Threshold

This paper discusses a selection criterion that generalizes the well-known concept of indifference zone selection through a preference threshold. A population is preferred to another population if the difference in the sums of observed values exceeds a given nonnegative threshold value. We present an argument for this selection rule by modelling preference by imprecise previsions. We aim at guidelines to design a selection experiment, which is characterized by two numbers: the number of necessary observations per population, and the preference threshold. Next to the probability of correct selection we also need a second specification. In this paper we consider a probability of false selection that is strongly related to the minimum probability of correct selection. Based on this model the outcome of an experiment may be ‘no selection’, at least not based on strong preference of a single population. The ideas are presented through a simple selection problem for normal populations with common known variance. Although the theory has a frequentist nature, the derivation and justification of the selection rule through imprecise previsions relies on Bayesian foundations, and via this route we gain more insight into the selection criterion.     

Optimization of a fermentation medium using neural networks and genetic algorithms

Artificial neural networks and genetic algorithms are used to model and optimize a fermentation medium for the production of the enzyme hydantoinase by Agrobacterium radiobacter. Experimental data reported in the literature were used to build two neural network models. The concentrations of four medium components served as inputs to the neural network models, and hydantoinase or cell concentration served as a single output of each model. Genetic algorithms were used to optimize the input space of the neural network models to find the optimum settings for maximum enzyme and cell production. Using this procedure, two artificial intelligence techniques have been effectively integrated to create a powerful tool for process modeling and optimization.     

Human Artificial Chromosome with a Conditional Centromere for Gene Delivery and Gene Expression

Human artificial chromosomes (HACs), which carry a fully functional centromere and are maintained as a single-copy episome, are not associated with random mutagenesis and offer greater control over expression of ectopic genes on the HAC. Recently, we generated a HAC with a conditional centromere, which includes the tetracycline operator (tet-O) sequence embedded in the alphoid DNA array. This conditional centromere can be inactivated, loss of the alphoid^tet-O (tet-O HAC) by expression of tet-repressor fusion proteins. In this report, we describe adaptation of the tet-O HAC vector for gene delivery and gene expression in human cells. A loxP cassette was inserted into the tet-O HAC by homologous recombination in chicken DT40 cells following a microcell-mediated chromosome transfer (MMCT). The tet-O HAC with the loxP cassette was then transferred into Chinese hamster ovary cells, and EGFP transgene was efficiently and accurately incorporated into the tet-O HAC vector. The EGFP transgene was stably expressed in human cells after transfer via MMCT. Because the transgenes inserted on the tet-O HAC can be eliminated from cells by HAC loss due to centromere inactivation, this HAC vector system provides important novel features and has potential applications for gene expression studies and gene therapy.     

用神经网络和遗传算法优化怀槐悬浮细胞合成异黄酮

为了获得怀槐悬浮细胞合成异黄酮的最适培养条件 ,采用ANNs(人工神经网络 )结合RAGA(实数编码加速遗传算法 )对培养基组成进行全局寻优。培养基中影响异黄酮染料木素产率的主要组成是KNO3、(NH₄)₂SO₄ 、2 ,4 D和 6 BA。在它们的有效作用浓度范围内 ,用随机10组培养基组合及细胞染料木素产率为输入和输出由ANNs对数据建模 ,由RAGA优化模型参数。建立的优化模型准确性高 ,依赖模型由RAGA全局寻优获得的最佳培养基组合是149.68mg L (NH₄)₂SO₄ 、2.936 1.0mg LKNO3、0.01mg L 2 ,4 D和 0.19mg L6-BA ,染料木素产率达14.13mg L ,与模型预测值的误差为 7.38%。结果表明 ,运用神经网络结合遗传算法优化怀槐细胞合成异黄酮的培养条件是可行的 ,优化后的培养基使怀槐细胞异黄酮合成能力比优化前有很大的提高.     

Genetic Algorithm Based Scheduling of Meta-Tasks with Stochastic Execution Times in Heterogeneous Computing Systemst1

In this study, we address the meta-task scheduling problem in heterogeneous computing (HC) systems, which is to find a task assignment that minimizes the schedule length of a meta-task composed of several independent tasks with no data dependencies. The fact that the meta-task scheduling problem in HC systems is NP-hard has motivated the development of many heuristic scheduling algorithms. These heuristic algorithms, however, neglect the stochastic nature of task execution times in an attempt to minimize a deterministic objective function, which is the maximum of the expected values of machine loads. Contrary to existing heuristics, we account for this stochastic nature by modeling task execution times as random variables. We, then, formulate a stochastic scheduling problem where the objective is to minimize the expected value of the maximum of machine loads. We prove that this new objective is underestimated by the deterministic objective function and that an optimal task assignment obtained with respect to the deterministic objective function could be inefficient in a real computing platform. In order to solve the stochastic scheduling problem posed, we develop a genetic algorithm based scheduling heuristic. Our extensive simulation studies show that the proposed genetic algorithm can produce better task assignments as compared to existing heuristics. Specifically, we observe a performance improvement on the relative cost heuristic (M.-Y. Wu and W. Shu, A high-performance mapping algorithm for heterogeneous computing systems, in: Int. Parallel and Distributed Processing Symposium, San Francisco, CA, April 2001) by up to 61%.