全β类蛋白编码序列的LZ复杂度对蛋白质折叠速率的影响

蛋白质折叠速率预测问题是计算生物学和生物信息学中的核心问题之一.科研工作者相继提出了许多参数和方法来探索折叠速率的决定因素.但蛋白质编码序列复杂度信息对蛋白质折叠速率的影响未被提及.提取编码序列LZ复杂度信息,融合多特征信息,建立线性回归模型进行折叠速率预测.该方法能在不需要结构信息的情况下,直接从蛋白质的编码序列出发对全β类蛋白质进行折叠速率进行预测.在卡方检验方法的验证下,发现折叠速率的预测值与实验值有很好的相关性,相关系数能达到0.9712.这一精度明显优于其他基于序列的方法,充分说明序列LZ复杂度是一个有效特征信息,蛋白质编码序列LZ复杂度信息确实影响蛋白质折叠速率及其结构.     

基于Hamilton路模型的蛋白质结构预测的研究

提出一种基于Hamilton路模型的新方法研究蛋白质结构预测问题,为使结构匹配序列,把已知蛋白质的3D结构信息转化为一个加权的完全图Kn,则求这个特定空间结构所匹配的氨基酸残基序列问题转化为求Kn图的最小H路问题.用此方法研究了72个单链蛋白质结构,结果表明Kn图的最小H路对应此蛋白质的序列,图的顶点数n与最小H路总长度成正比.     

蛋白质结构的主成分分析及氧链糖基化位点的人工神经网络预测

糖基化是蛋白质翻译后修饰的重要形式之一,氧链糖基化是糖基化的一种主要类型,对蛋白质氧链糖基化位点进行预测具有重要的意义.以窗口长度为41的蛋白质序列为研究对象,采用稀疏编码,利用主成分分析法研究了氧链糖基化蛋白质序列的结构特点;在提取主成分的基础上,设计了一个含单隐层的BP神经网络(256—8—4),对蛋白质氧链糖基化位点进行预测,把蛋白质序列分为4类;并同直接用BP神经网络分类的结果相比较,实验结果证明提出的方法省时,准确,预测的准确率达80~90%.     

基于k字位置序列的蛋白质序列分析方法及其应用

基于蛋白质序列的κ-字位置序列,利用标准化的κ-字区间平均距离和改进的标准化的κ-字区间平均距离的方法作为蛋白质序列的数字特征,并给出了比较蛋白质序列相似性的方法.最后,运用这两种方法分析了9个物种的ND5蛋白质序列和8个物种的ND6蛋白质序列的相似性,并利用交叉验证得出基于改进的标准化的κ-字区间平均距离的方法的准确度比基于标准化的κ-字区间平均距离的方法的准确度高.     

多种权重与信息扩散相结合的灾害风险评估模型的比较研究

用基于遗传-投影寻踪的分类权重、基于模糊理论的排序权重,和优化处理后得到的组合权重,分别和相应的按比例缩小的指标耦合,构造灾情指数序列和致灾源指数序列;通过二维正态扩散构造原始信息短阵、模糊关系矩阵,利用因素空间理论进行模糊近似推理,分别得到由致灾源因子近似估计灾情的风险估计值.结果表明,利用排序权重模型、组合权重模型计算输出的风险估计值与灾情指数序列基本一致;通过对这三种权重模型的风险估计值误差计算、灾情的风险估计值与灾情指数序列相关性计算、一致性检验以及K-S测试的比较分析发现,组合权重模型比分类权重和排序权重模型更能反映华南TCs灾害风险状况.     

基于方向多尺度变换和统计建模的纹理分类方法

纹理是图像分析和识别中经常使用的关键特征, 而小波变换则是图像纹理表示和分类中的常用工具. 然而, 基于小波变换的纹理分类方法常常忽略了小波低频子带信息, 并且无法提取图像纹理的块状奇异信息. 本文提出小波子带系数的局部能量直方图建模方法、轮廓波特征的Poisson 混合模型建模方法和基于轮廓波子带系数聚类的特征提取方法, 并将其应用于图像纹理分类上. 基于局部能量直方图的纹理分类方法解决了小波低频子带的建模难题, 基于Poisson 混合模型的纹理分类方法则首次将Poisson 混合模型用于轮廓子带特征的建模, 而基于轮廓波域聚类的纹理分类方法是一种快速的分类方法. 实验结果显示, 本文所提出的三类方法都超过了当前典型的纹理分类方法.     

一种能够提高建模精度的含参变量函数变换及其应用

根据传统预测模型适宜低增长、近指数序列这一基本事实出发,提出用原始序列本身的级比偏差来比较低增长的程度,用原始序列所对应的级比序列的级比偏差比较原始序列接近指数序列的程度.并通过此种方法找到了一种含参变量函数变换缩小级比偏差的范围,从而提高了建模精度,又因为此种变换能够保证还原相对误差保持不变,以至于最终达到了提高建模精度的目的.并通过实例验证了方法的可行性及其结果的正确性、优越性.     

利用提升小波的蛋白质相互作用特征提取

利用提升小波从蛋白质序列中提取出它们相互作用的频谱特征,经支持向量机训练学习后,用于预测蛋白质间的相互作用.模拟计算结果表明,在阳性数据和阴性数据平衡的前提下,利用提升小波获取的低维蛋白质相互作用特征向量可以得到较高预测精度.进一步阐述了不同物种的蛋白质相互作用网络有着不同特征,为了得到更准确的预测结果,需要利用不同的方法提取蛋白质相互作用的特征.     

数据预处理对替代数据检验方法的影响

替代数据检验法是检验时间序列中是否存在确定性非线性成分的重要统计方法.通过研究差分和数据平滑运算对替代数据检验方法的影响,指出常用的线性滤波等数据预处理步骤破坏了序列的静态性质,从而会导致对零假设的错误拒绝.因此,建议应该直接利用原始时间序列而非应用了差分等非静态滤波运算后的时间序列生成替代数据,再进行假设检验,以免造成对零假设的错误拒绝.     

平衡设计多向分类多元重复测量模型的Wilks检验

对平衡设计多向分类多元重复测量模型,利用极大似然比方法,推导了对各单个固定效应分别进行检验的Wilks型检验规则.并推导了对多个固定效应进行同时检验的检验规则.推导了非中心分布的参数与原始参数和样本容量的关系.     

Index-doubling in sequences by Aitken extrapolation

Aitken extrapolation, applied to certain sequences, yields the even-numbered subsequence of the original. We prove that this is true for sequences generated by iterating a linear fractional transformation, and for some sequences of convergents of the regular continued fractions of certain quadratic irrational numbers.     

Some test length bounds for nonrepeating functions in the {&;, ∨} basis

Testing relative to a nonrepeating alternative in a conjunction-disjunction basis is considered. A lower bound on the test length is established for all nonrepeating functions in this basis. A subsequence of easily testable functions is constructed and the corresponding tests are described. Individual lower test length bounds are proved for functions of a special form; minimality of the tests is established for the functions of the constructed subsequence.     

Classification of families of pr- and epr-sequences

This paper establishes new restrictions for attainable enhanced principal rank characteristic sequences (epr-sequences). These results are then used to classify two related families of sequences that are attainable by a real symmetric matrix: the family of principal rank characteristic sequences (pr-sequences) not containing three consecutive 1s and the family of epr-sequences which contains an N in every subsequence of length 3.     

Convex unconditionality and summability of weakly null sequences

It is proved that every normalized weakly null sequence has a subsequence which is convexly unconditional. Further, a hierarchy of summability methods is introduced and with this we give a complete classification of the complexity of weakly null sequences.     

Addition to the Aitken method for the extrapolation of the limit of slowly convergent sequences

The class of sequences and series in which the Aitken process accelerates the convergence is considerably extended. It is proved that a proper subsequence of a slowly convergent sequence satisfies the sufficient condition for accelerating the convergence of the Aitken transformation. Two numerical examples illustrate the highly accurate limit extrapolation.     

Approximation to the mean curve in the LCS problem

The problem of sequence comparison via optimal alignments occurs naturally in many areas of applications. The simplest such technique is based on evaluating a score given by the length of a longest common subsequence divided by the average length of the original sequences. In this paper we investigate the expected value of this score when the input sequences are random and their length tends to infinity. The corresponding limit exists but is not known precisely. We derive a theoretical large deviation, convex analysis and Monte Carlo based method to compute a consistent sequence of upper bounds on the unknown limit. An empirical practical version of our method produces promising numerical results.     

On Suboptimal LCS-alignments for Independent Bernoulli Sequences with Asymmetric Distributions

Let X = X ₁ ... X _n and Y = Y ₁ ... Y _n be two binary sequences with length n. A common subsequence of X and Y is any subsequence of X that at the same time is a subsequence of Y; The common subsequence with maximal length is called the longest common subsequence (LCS) of X and Y. LCS is a common tool for measuring the closeness of X and Y. In this note, we consider the case when X and Y are both i.i.d. Bernoulli sequences with the parameters ϵ and 1 − ϵ, respectively. Hence, typically the sequences consist of large and short blocks of different colors. This gives an idea to the so-called block-by-block alignment, where the short blocks in one sequence are matched to the long blocks of the same color in another sequence. Such and alignment is not necessarily a LCS, but it is computationally easy to obtain and, therefore, of practical interest. We investigate the asymptotical properties of several block-by-block type of alignments. The paper ends with the simulation study, where the of block-by-block type of alignments are compared with the LCS.     

A note on the longest matching consecutive subsequence

The longest matching consecutive subsequence plays an important role in information theory and molecular biology. We consider the Hausdorff dimension of the set of points whose rate of growth of the longest matching consecutive subsequence is almost equal to a class of monotonically increasing functions.     

Solutions to enumeration problems for single-transition serial sequences with an adjacent series height increment bounded above

Sets of n-valued finite serial sequences are investigated. Such a sequence consists of two serial subsequences, beginning with an increasing subsequence and ending in a decreasing one (and vice versa). The structure of these sequences is determined by constraints imposed on the number of series, on series lengths, and on series heights. For sets of sequences the difference between adjacent series heights in which does not exceed a certain given value 1 ≤ |h _j+1 ? h _j | ≤ δ, two algorithms are constructed of which one assigns smaller numbers to lexicographically lower sequences and the other assigns smaller numbers to lexicographically higher sequences.     

Upper bounds for the expected length of a longest common subsequence of two binary sequences

Let f(n) be the expected length of a longest common subsequence of two random binary sequences of length n. It is known that the limit γ = lim_{n→ ∞} n^?1 f(n) exists. Improved upper bounds for γ are given using a new method.