Efficient prediction of protein conformational pathways based on the hybrid elastic network model

Various computational models have gained immense attention by analyzing the dynamic characteristics of proteins. Several models have achieved recognition by fulfilling either theoretical or experimental predictions. Nonetheless, each method possesses limitations, mostly in computational outlay and physical reality. These limitations remind us that a new model or paradigm should advance theoretical principles to elucidate more precisely the biological functions of a protein and should increase computational efficiency. With these critical caveats, we have developed a new computational tool that satisfies both physical reality and computational efficiency. In the proposed hybrid elastic network model (HENM), a protein structure is represented as a mixture of rigid clusters and point masses that are connected with linear springs. Harmonic analyses based on the HENM have been performed to generate normal modes and conformational pathways. The results of the hybrid normal mode analyses give new physical insight to the 70S ribosome. The feasibility of the conformational pathways of hybrid elastic network interpolation (HENI) was quantitatively evaluated by comparing three different overlap values proposed in this paper. A remarkable observation is that the obtained mode shapes and conformational pathways are consistent with each other. Our timing results show that HENM has some advantage in computational efficiency over a coarse-grained model, especially for large proteins, even though it takes longer to construct the HENM. Consequently, the proposed HENM will be one of the best alternatives to the conventional coarse-grained ENMs and all-atom based methods (such as molecular dynamics) without loss of physical reality.     

Dynamic characteristics of a flagellar motor protein analyzed using an elastic network model

At the base of a flagellar motor, its rotational direction and speed are regulated by the interaction between rotor and stator proteins. A switching event occurs when the cytoplasmic rotor protein, called C-ring, changes its conformation in response to binding of the CheY signal protein. The C-ring structure consists of FliG, FliM, and FliN proteins and its conformational changes in FliM and FliG including Helix_MC play an important role in switching the motor direction. Therefore, clarifying their dynamic properties as well as conformational changes is a key to understanding the switching mechanism of the motor protein. In this study, to elucidate dynamic characteristics of the C-ring structure, both harmonic (intrinsic vibration) and anharmonic (transition pathway) analyses are conducted by using the symmetry-constrained elastic network model. As a result, the first three normal modes successfully capture the essence of transition pathway from wild type to CW-biased state. Their cumulative square overlap value reaches up to 0.842. Remarkably, it is also noted from the transition pathway that the cascade of interactions from the signal protein to FliM to FliG, highlighted by the major mode shapes from the first three normal modes, induces the reorientation (∼100° rotation of FliG_C5) of FliG C-terminal that directly interacts with the stator protein. Presumably, the rotational direction of the motor protein is switched by this substantial change in the stator-rotor interaction.     

基于深度卷积神经网络的心音分类算法

针对现有心音分类算法普适性差、依赖于对基本心音的精确分割、分类模型结构单一等问题,提出采用大量未经过精确分割的心音二维特征图训练深度卷积神经网络(CNN)的方法;首先采用滑动窗口方法和梅尔频率系数对心音信号进行预处理,得到大量未经过精确分割的心音特征图;然后利用深度CNN模型对心音特征图进行训练和测试;根据卷积层间连接方式的不同,设计了 3种深度CNN模型:基于单一连接的卷积神经网络、基于跳跃连接的卷积神经网络、基于密集连接的卷积神经网络;实验结果表明,基于密集连接的卷积神经网络比其他两种网络具备更大的潜力;与其他心音分类算法相比,该算法不依赖于对基本心音的精确分割,且在分类准确率、敏感性和特异性方面均有提升.     

基于小世界回声状态网的时间序列预测

为了提高时间序列的预测精度, 提出了利用改进的小世界网络优化泄露积分型回声状态网(Leaky-integrator echo state network, Leaky ESN)的时间序列预测方法. 首先提出一个改进型小世界网络, 其加边概率是节点间距离的负指数函数. 然后, 利用加边概率直接表示Leaky ESN储备池两个神经节点的连接权值, 取值范围为[0,1], 表征了节点间的连接程度. 利用这个新型小世界网络改进Leaky ESN的储备池神经节点的连接方式, 有目的地实现了稀疏连接, 减小了Leaky ESN储备池随机稀疏连接的盲目性, 提高了储备池的适应性.最后, 利用改进的Leaky ESN预测典型的非线性时间序列, 并利用Matlab仿真软件验证了本文提出方法的有效性. 与Leaky ESN相比, 本文提出的方法具有更高的预测精度和更短的训练时间.     

基于收益评估机制的连接数阈值动态调节算法

针对目前出口网关中并发连接数限制的定值方法的缺点,提出了一种基于收益评估机制的动态调节并发连接数阈值算法。根据用户链路某个时间段内的并发连接数变化特征,可利用收益评估的方法计算出一个最优化的连接数限制阈值,利用该值调节用户链路的并发连接数。通过调节有效连接与无效连接的效益值可适用于不同的网络偏好,对对等网（P2P）应用产生的大量无效连接进行有效抑制,保障了网络应用多元化的高效使用,同时给出了工程实现算法。     

基于连接自组织发育的稀疏跨越-侧抑制神经网络设计

针对跨越——侧抑制神经网络（Span-lateral inhibition neural network,S-LINN）的结构调整及参数学习问题,结合生物神经系统中神经元的稀疏连接特性,依据儿童及青少年智力发展水平与大脑皮层发育之间的相互关系,提出以小世界网络连接模式进行初始稀疏化的连接自组织发育稀疏跨越——侧抑制神经网络设计方法.定义网络连接稀疏度及神经元输出贡献率,设计网络连接增长——修剪规则,根据智力超常组皮层发育与智力水平的对应关系调整和控制网络连接权值,动态调整网络连接实现网络智力的自组织发育.通过非线性动力学系统辨识及函数逼近基准问题的求解,证明在同等连接复杂度的情况下,稀疏连接的跨越——侧抑制神经网络具有更好的泛化能力.     

A fuzzy connectionist expert system for visual pattern classification

In this paper, a connectionist model to integrate knowledge-based techniques into neural network approaches for visual pattern classification is presented. We propose a new structure of connectionist model which has rule-following capability as well as instance-based learning capability. Each node of the proposed network is doubly linked by two types of connections: positive connection and negative connection. Such connectionism provides a methodology to construct the classifier from the rule base and allows the expert knowledge to be utilized for the effective learning. For visual pattern classification, we present the techniques for knowledge representation and utilization using the concepts of fuzzy rules and fuzzy relations. We also discuss in this paper some advantageous characteristics of the model: result explanation capability and rule refinement capability. From the experimental results of the handwritten digit classification, the feasibility of the proposed model is evaluated.     

On Generalized Connection Caching

Cohen et al. [5] recently initiated the theoretical study of connection caching in the world-wide web. They extensively studied uniform connection caching, where the establishment cost is uniform for all connections [5], [6]. They showed that ordinary paging algorithms can be used to derive algorithms for uniform connection caching and analyzed various algorithms such as Belady's rule, LRU and Marking strategies. In particular, in [5] Cohen et al. showed that LRU yields a (2k-1) -competitive algorithm, where k is the size of the largest cache in the network. In [6] they investigated Marking algorithms with different types of communication among nodes and presented deterministic k -competitive algorithms. In this paper we study generalized connection caching , also introduced in [5], where connections can incur varying establishment costs. This model is reasonable because the cost of establishing a connection depends, for instance, on the distance of the nodes to be connected and on the congestion in the network. Algorithms for ordinary weighted caching can be used to derive algorithms for generalized connection caching. We present tight or nearly tight analyses on the performance achieved by the currently known weighted caching algorithms when applied in generalized connection caching. In particular we give online algorithms that achieve an optimal competitive ratio of k . Our deterministic algorithm uses extra communication while maintaining open connections. We develop a generalized algorithm that trades communication for performance and achieves a competitive ratio of (1+ε)k , for any 0<ε≤ 1 , using at most

基于无监督Hebb规则的入侵检测方法

通过分析现有的入侵检测方法,提出了一个基于无监督Hebb规则的入侵检测方法。此方法采用高效的抓包工具抓取计算机网络数据包;根据抓取到的网络数据包的信息定义行为变量;根据无监督的Hebb规则构建网络行为模型;采用Hamming距离进行检测。实验证明该方法能够正确地构造网络行为模型,并能准确地检测出异常行为。     

基于改进Elman神经网络的网络流量预测

针对网络系统非线性、多变量、时变性等特点,提出一种改进的Elman神经网络模型。在该模型的训练过程中引入了季节周期性学习方法,并对某高校主干网络出口流量进行实验检测。实验结果表明,该模型具有良好的预测效果,相对于传统线性模型、BP神经网络模型及标准Elman神经网络模型具有更高的预测精度和更好的自适应性。最后,通过自适应边界值方法进行检测,能够及时发现异常流量行为,说明该模型应用于网络流量预测是可行、有效的。     

蛋白质二级结构预测方法研究

为提高蛋白质二级结构预测精度,提出一种新的网络模型和编码方法。首先利用基因表达式编程（GEP）的全局搜索能力同时进化设计神经网络的结构和连接权;其次,对神经网络输入层编码进行了改进,添加了氨基酸残基所处的疏水环境。用PDBSelect25中的36条蛋白质共6 122个残基进行测试,结果表明提出的网络模型和编码方法能有效提高蛋白质二级结构预测的精度。     

一种蛋白质点突变计算机预测的并行模型

认识和预测蛋白质天然构象的波动对蛋白质-蛋白质对接和设计等应用是非常重要的.但是许多骨架柔性的方法会导致骨架较大幅度的波动.Backrub模型能够对骨架进行微小的扰动,符合高分辨率晶体结构中观察到的构象的微妙变化.本文提出了一种基于Backrub的并行扰动骨架和侧链的模型,可以对天然构象的等价状态进行模拟.这种并行扰动方式更加接近于真实情况下蛋白质构象的运动方式,更好地模拟了实验数据.通过预测10个点突变实例,相比串行随机扰动模型产生的构象,并行模型不仅从时间上提高了产生构象的速度,更提高了侧链的预测精度.     

可信网络连接中的可信度仿真评估*

可信网络连接(TNC)是对可信平台应用的扩展,也是可信计算机制与网络接入控制机制的结合。针对当前TNC研究中缺乏对网络连接进行定量评估的问题,提出了一种网络连接可信度仿真评估方法。该方法根据网络连接属性与评估值之间存在的非线性映射关系,利用BP网络模型及LM优化算法对网络连接可信度进行评估。仿真结果表明该方法具有较好的预测评估准确率。     

双端可共享网络的多模态行人重识别方法

针对多模态行人重识别中存在较大的类内差异和模态差异的问题,提出了一种使用双端共享网络的多模态行人重识别方法。通过裁剪和填充对不同模态的图片进行数据处理;将Resnet50的后4个卷积层中嵌入非局部注意力块,使用改进的Resnet50作为骨干网络分别对不同模态的图片进行特征提取,再将不同的特征输入共享网络;最后使用基于类内距离和模态差异的聚类损失对模型进行训练。实验结果表明,使用非局部注意力块和聚类损失的模型准确率有所提升,且模型更具有鲁棒性。     

面向WSI的乳腺病理亚型分类研究

为实现乳腺病理WSI图像的精准分类,提出了一种基于混合连接的门控卷积神经网络分类方法。搭建了局部残差连接和全局稠密连接的混合模块,将压缩激活门控单元嵌入混合模块,建立了混合模块与过渡层交替连接的骨干网络。结合基于四叉树分割的图像数据增强方法训练模型,基于BreastSet临床数据集的实验结果得出,该方法的图像级、患者级和病理级准确率分别达到92.24%、92.83%和92.18%,相较其他方法,其准确率提高,参数量和计算量降低,更具有临床应用价值。     

监控视频异常行为检测的概率记忆自编码网络

异常行为检测是智能监控系统中重要的功能之一, 在保障社会治安等方面发挥着积极的作用. 为提高监控视频中异常行为的检测率, 从学习正常行为分布的角度出发, 设计基于概率记忆模型的半监督异常行为检测网络, 解决正常行为数据与异常行为数据极度不均衡的问题. 该网络以自编码网络为主干网络, 利用预测的未来帧与真实帧之间的差距来衡量异常程度. 在主干网络提取时空特征时, 使用因果三维卷积和时间维度共享全连接层来避免未来信息的泄露, 保证信息的时序性. 在辅助模块方面, 从概率熵和正常行为数据模式多样性的角度, 设计概率模型和记忆模块提高主干网络视频帧重建质量. 概率模型利用自回归过程拟合输入数据分布, 促使模型收敛于正常分布的低熵状态; 记忆模块存储历史数据中的正常行为的原型特征, 实现多模式数据的共存, 同时避免主干网络的过度参与而造成对异常帧的重建. 最后, 利用公开数据集进行消融实验和与经典算法的对比实验, 以验证所提算法的有效性.     

Normal mode analysis of proteins: a comparison of rigid cluster modes with C(alpha) coarse graining

The ability to infer dynamic motions from an equilibrium (static) conformation of a protein can be essential in establishing structure-function relationships. In particular, the low-frequency motions are of functional interest because statistical mechanics predicts these motions will have the largest amplitudes. In this paper, we address the computational cost of normal mode analysis (NMA) applied to a C(alpha)-based elastic network model (C(alpha)-NMA) and present a new coarse-grained rigid-body-based analysis (cluster-NMA). This new method represents a protein as a collection of rigid bodies interconnected with harmonic potentials. This representation produces reduced degree-of-freedom (DOF) equations of motion (EOMs) which, even in the case of large structures (10(3+) residues), enables the computation of normal modes to be done on a desktop PC. We present the complete theory and analysis of cluster-NMA and also include its application to a variety of structures. The results of the new method are compared with C(alpha)-NMA and it is shown that cluster-NMA produces very good approximations to the lowest modes at a fraction of the computational cost.     

轻量级水下目标检测器LUDet

针对传统水下目标检测器受环境影响较大的问题,使用一种新的轻量级网络LUNet提取特征,结合两阶段检测算法提出轻量级检测器LUDet。首先,网络的第1个阶段使用高效卷积池化来获取不同特征表达。然后,在稠密连接结构的基础上增加两路稠密连接,以提高网络表征能力。网络由卷积池化层与两路稠密连接结构构成,网络中使用GhostModel代替1×1点卷积。使用CAFIR10和CAFIR100数据集进行分类实验验证了提出的骨干网的有效性。针对检测任务,LUDet通过通道注意力、多阶段融合后的特征图对目标进行检测。使用2个水下数据集对改进的检测器进行验证,水下生物数据集上检测的mAP达到了52.5%,水下垃圾数据集上检测的mAP达到了58.7%。     

基于互信息的生物神经网络功能性连接辨识

生物神经网络（ BNN）功能性连接的辨识方法被广泛地应用于使用BNN的多通道时间序列数据构建网络连接结构,帮助加深对BNN结构和功能间关系的认识和理解。首先,建立基于积分点火（ IF）机制的BNN模型,获得多通道神经元脉冲序列;然后,运用互信息（ MI）方法计算出各神经元间的MI值,超过一定阈值的MI表明两个神经元间存在相互连接关系。仿真结果表明：基于MI的网络辨识方法计算开销较小,对BNN功能性连接结构具有较高的辨识度。     

Modal synthesis method of lateral vibration analysis for rotor-bearing system

Rigid coupling and flexible connection made up of elastic coupling units are widely applied to rotor-bearing system with multi-branched shafting system. This paper proposes a modal synthesis method of lateral vibration analysis for such kind of rotor-bearing system. When the proposed approach is developed, the elastic coupling unit is defined as “flexible substructure” which is treated individually and the other parts are partitioned into some substructures which are analyzed by finite element method. The lower-frequency normal modes of the substructures are retained; whereas the higher-frequency normal modes are neglected by a frequency truncation criterion and the residual flexibility of those omitted modes is considered. The lower-frequency normal modes and residual flexibility are considered to be the assumed modes of Rayleigh–Ritz analysis of whole structure. The approach to the treatment of higher-frequency modes leads to a great reduction in the calculation time and a significant improvement in the efficiency of modal synthesis. Examples show that the proposed approach is very effective for the vibration analysis of rotor-bearing system with multi-branched shafting system and the results derived from this approach are in a good agreement with those from transfer matrix method.