一种随机化检验算法及其Matlab实现

在数据满足特定假设的前提下,可用有关统计技术检验样品间的差异．然而,在自然情况下,这些假设往往不成立或未知成立．本研究建立了一种随机化检验算法,可对实数域上的两个样品进行差异显著性分析,给出了Matlab标准源程序．算法具有广适性．设定随机化模拟次数和差异显著性水平,则可计算出检验值P,及差异显著与否．该算法可使用各种差异函数,如欧氏距离等等．应用本算法,对数种实测样品数据,以不同的差异函数进行了差异性分析．     

基于盒维数的心音信号分形特征研究

在传统盒维数的基础上,从尺度变化的角度,提出一种计算心音信号时域波形分形维数的新的二进盒维数算法,并给出了算法思想和估算方法;然后用该方法对正常心音和几种典型的病态心音的分形维数进行计算,并对其分形特征进行了研究．研究结果表明：心音信号具有明显的分形特征,分形维数能够反映心音信号的复杂程度,并且能够明显地区分正常心音和病态心音．     

广义Kolmogorov模型的Lyapunov函数构造新算法及其应用

本文对广义Kolmogorov模型,给出构造Lyapunov函数的新算法,在文献1中只对其中某些特殊类型给出几种特殊的构造方法,而本文给出的是这类模型的一般新算法,应用较广泛。     

用网络方法识别生物序列motif

生物序列motif的识别是后基因组时代的一个核心问题。本文首先回顾了识别motif的几种主要算法,然后根据motif的重要性和随机性介绍了利用网络识别motif的两种具有代表性的方法：一种是建立一个随机网络混合模型,利用EM算法识别其中随机的网络motif;另一种用修正的参数流算法过滤出其中的最大密度予图,即为生物序列motif,并指出这两种方法的优劣,最后还对今后研究方向给出了讨论。     

基于信息论k—modes聚类法的基因表达数据分析

k-均值聚类算法是一种广泛应用于基因表达数据聚类分析中的迭代变换算法,它通常用距离法来表示基因间的关系,但不能有效的反应基因间的相互依赖的关系。为此,提出基于信息论的k-modes聚类算法,克服了以上缺点。另外,还引入了伪F统计量,一方面,可以对空间中有部分重叠的点进行有效的分类;另一方面,可以给出最佳聚类数目,从而弥补了k-modes聚类法的不足。使其成为一种非常有效的算法,从而达到较优的聚类效果。     

一种基于支持向量机的自适应肿瘤分类检测算法

根据肿瘤分类检测模型的特点,提出了一种新的算法,该算法结合使用了基因选择和数据抽取的有效方法,并在此基础上使用支持向量机对基因表达数据进行分类或者检测。其中乳腺癌的分类交叉验证结果由88．46％提高到100．0％,急性白血病的也由71．05％提高至100．0％。实验结果说明了这一方法的有效性,为在大量的基因表达数据中提高检测癌症的准确性提出了一种比较通用的方法。     

多通道神经元锋电位检测和分类的新方法

大脑神经元胞外单细胞动作电位(即锋电位)的检测和分类是提取神经元脉冲序列、研究神经系统信息处理机制的关键．为了提高锋电位的检出率和分类的正确性,设计了一种处理多通道锋电位记录信号的算法,用于分析微电极阵列记录的大鼠海马神经元锋电位信号,电极阵列上的测量点排列紧密,4个通道可以同时记录到来自相同神经元的信号．该算法首先利用一种多通道阈值检测法检出四通道记录信号中的锋电位,然后利用一种基于复合锋电位的主成分特征参数分类法将锋电位分类．仿真数据和实验记录信号的检验结果表明：与相应的单通道算法相比,该算法的锋电位检出率和分类的正确性显著提高,并且可以增加单次实验测得的神经元数目．因此,该算法为实现神经元锋电位的自动检测提供了一种简单有效的新 方法．     

实数遗传算法的改进研究

在现有文献研究的基础上,对实数遗传算法又作了进一步研究,提出了一种改进算法．该算法不仅可快速产生初始种群,而且实现了子代种群的产生在优化方向上进行,提高了算法的搜索能力,克服了子代个体位置限制的不足,有利于保持种群的多样性,提高了避免未成熟收敛于局部最优解的能力．     

非线性增长模型的回归分析

本文对Logistic模型和Gompertz模型给出了计算参数值的迭代公式,并提出了一种确定迭代初始值的新方法．     

基于后缀列的基因序列最大串联重复查找技术

重复序列分析在全基因组研究中起着重要作用,其首要任务就是在DNA序列中识别并定位所有的重复结构。本文提出了一种新的算法,此算法基于一种简单的数据结构——后缀数,用于查找给定的DNA序列中所有的最大串联重复。并且在该算法的基础上编写了一个有效实用的软件——RepLocate,同时给出了它应用到已知的DNA序列的实例。     

On testing biological data for the presence of a boundary

Under the boundary line model for a biological data set, where one variable is a biological response (e.g. crop yield) to an independent variable (e.g. available water content of the soil), we interpret the upper (or lower) boundary on a plot of the dependent variable (ordinate) against the independent variable (abscissa) as representing the maximum (or minimum) possible response for a given value of the independent variable. This concept has been widely used in soil science, agronomy and plant physiology; but it has been subject to criticism. In particular, no methods that are used to analyse the boundary line quantify the evidence that the envelope of the plot represents a boundary (in the sense of some limiting response to the independent variable) rather than simply being a fringe of extreme values of no intrinsic biological interest. In this article, we present a novel procedure that tests a data set for evidence of a boundary by considering its statistical properties in the region of the proposed boundary. The method is demonstrated using both simulated and real data sets.     

An Alternative Approach to Modelling the Assortative Mating Process

In this article we demonstrate how the difficulties encountered from consideration of two-sided assortation for a characteristic (or set of characteristics) in human populations can be overcome. This is achieved, essentially, by postulating a socio-environmental variable (based on an individual's lifetime experiences) over which individuals are mating assortatively. The value an individual has on this scale determines his mean value for the characteristic under consideration. In other words, there is a relationship between the characteristic and the socioenvironmental scale. As well, individuals may be mating assortatively for the characteristic under consideration (although not necessarily). Thus we achieve assortative mating for a single characteristic as a result of mixing, simultaneously, over more than just one variable. Expansions to more ‘mixing’ scales than a single socio-environmental variable are quite straightforward.     

Two-sample density-based empirical likelihood tests for incomplete data in application to a pneumonia study

In clinical trials examining the incidence of pneumonia it is a common practice to measure infection via both invasive and non-invasive procedures. In the context of a recently completed randomized trial comparing two treatments the invasive procedure was only utilized in certain scenarios due to the added risk involved, and given that the level of the non-invasive procedure surpassed a given threshold. Hence, what was observed was bivariate data with a pattern of missingness in the invasive variable dependent upon the value of the observed non-invasive observation within a given pair. In order to compare two treatments with bivariate observed data exhibiting this pattern of missingness we developed a semi-parametric methodology utilizing the density-based empirical likelihood approach in order to provide a non-parametric approximation to Neyman-Pearson-type test statistics. This novel empirical likelihood approach has both a parametric and non-parametric components. The non-parametric component utilizes the observations for the non-missing cases, while the parametric component is utilized to tackle the case where observations are missing with respect to the invasive variable. The method is illustrated through its application to the actual data obtained in the pneumonia study and is shown to be an efficient and practical method.     

Implementation of an adaptive controller for the startup and steady-state running of a biomethanation process operated in the CSTR mode

An adaptive control algorithm has been implemented on a biomethanation process to maintain propionate concentration, a stable variable, at a given low value, by steering the dilution rate. It was thereby expected to ensure the stability of the process during the startup and during steady-state running with an acceptable performance. The methane pilot reactor was operated in the completely mixed, once-through mode and computer-controlled during 161 days. The results yielded the real-life validation of the adaptive control algorithm, and documented the stability and acceptable performance expected.     

Exact solution of the multi-allelic diffusion model

We give an exact solution to the Kolmogorov equation describing genetic drift for an arbitrary number of alleles at a given locus. This is achieved by finding a change of variable which makes the equation separable, and therefore reduces the problem with an arbitrary number of alleles to the solution of a set of equations that are essentially no more complicated than that found in the two-allele case. The same change of variable also renders the Kolmogorov equation with the effect of mutations added separable, as long as the mutation matrix has equal entries in each row. Thus, this case can also be solved exactly for an arbitrary number of alleles. The general solution, which is in the form of a probability distribution, is in agreement with the previously known results. Results are also given for a wide range of other quantities of interest, such as the probabilities of extinction of various numbers of alleles, mean times to these extinctions, and the means and variances of the allele frequencies. To aid dissemination, these results are presented in two stages: first of all they are given without derivations and too much mathematical detail, and then subsequently derivations and a more technical discussion are provided.     

An efficient method for calculation of dynamic logarithmic gains in biochemical systems theory

Biochemical systems theory (BST) characterizes a given biochemical system based on the logarithmic gains, rate-constant sensitivities and kinetic-order sensitivities defined at a steady state. This paper describes an efficient method for calculation of the time courses of logarithmic gains, i.e. dynamic logarithmic gains L(Xi, Xj; t), which expresses the percentage change in the value of a dependent variable Xi at a time t in response to an infinitesimal percentage change in the value of an independent variable Xj at t=0. In this method, one first recasts the ordinary differential equations for the dependent variables into an exact canonical nonlinear representation (GMA system) through appropriate transformations of variables. Owing to the structured mathematical form of this representation, the recast system can be fully described by a set of numeric parameters, and the differential equations for the dynamic logarithmic gains can be set up automatically without resource to computer algebra. A simple general-purpose computer program can thus be written that requires only the relevant numeric parameters as input to calculate the time courses of the variables and of the dynamic logarithmic gains for both concentrations and fluxes. Unlike other methods, the proposed method does not require to derive any expression for the partial differentiation of flux expressions with respect to each independent variable. The proposed method has been applied to two kinds of reaction models to elucidate its usefulness.     

On the determination of optimal levels in phytosociological classification

A method is introduced to compare results of a clustering technique at different levels of abstraction, or of different clustering techniques. The method emphasizes within cluster homogeneity as well as discontinuities between clusters. It has been derived from Hogeweg's method with some important changes. First each cluster is handled separately to determine the ratio between homogeneity and similarity to the nearest neighbour cluster. For a given clustering a weighted average value is computed over all clusters. This average value is standardized using an expected average value for a cluster configuration with the same number of clusters having the same sizes. A low level of the ratio between expected and observed values is supposed to indicate an optimal clustering. A derivation of the criterion is given and results from three sets of data with different properties are evaluated.     

Note on the analysis of thymidine labeling in cells with chromosome-number dispersion

A method of analyzing thymidine labeling in a population of cells is formulated. The formulation establishes a unique relation between a specific set of labeling data and a specific set of cells in the population, viz. that set of cells having a particular chromosome number. The analysis employs a cell-state variable, i.e., a quantity which specifies the progress of a cell through its lifecycle. This variable is defined in terms of the nucleo-protein content and configuration of the chromosomes. The relation mentioned above leads immediately to an expression for the number of cells present at a particular time following labeling which have a given amount of label per cell and a given chromosome number.     

On the role of anatomy in learning by the cerebellar cortex

The properties due to the location of neurons, synapses, and possibly even synaptic channels, in neuron networks are still unknown. Our preliminary results suggest that not only the interconnections but also the relative positions of the different elements in the network are of importance in the learning process in the cerebellar cortex. We have used neural field equations to investigate the mechanisms of learning in the hierarchical neural network. The numerical resolution of these equations reveals two important properties: (i) The hierarchical structure of this network has the expected effect on learning because the flow of information at the neuronal level is controlled by the heterosynaptic effect through the synaptic density-connectivity function, i.e. the action potential field variable is controlled by the synaptic efficacy field variable at different points of the neuron. (ii) The geometry of the system involves different velocities of propagation along different fibers, i.e. different delays between cells, and thus has a stabilizing effect on the dynamics, allowing the Purkinje output to reach a given value. The field model proposed should be useful in the study of the spatial properties of hierarchical biological systems.