复方配伍的摄动模糊聚类方法

利用摄动模糊聚类方法对中药复方桂枝汤的药群进行了分类,结果表明该方法优于传统模糊聚类方法,避免了利用传递闭包求模糊等价矩阵进行分类的失真问题,与中药传统组方原则相吻合。     

我国草兔的聚类研究

本文用新的聚类方法, 对中国草兔进行系统聚类。     

蛋白质序列的聚类结构分析

基于模糊邻近关系的粒度空间,对蛋白质序列进行聚类结构分析。利用MEGA软件计算选取的木聚糖酶序列间的比对距离,引入内积将其转化为模糊邻近关系(或矩阵),再应用算法求解其粒度空间,进行序列的聚类结构分析和最佳聚类确定研究。这些研究为蛋白质序列提供了定量分析的工具。     

基于改进Spectrum的肠道菌群聚类算法研究

肠道菌群与诸多人类重大疾病相关,研究在不同条件下的肠道菌群数据具有重要意义。由于菌群数据出现零膨胀现象,采用成对比率几何平均值（GMPR）方法对其进行归一化。本研究以2型糖尿病数据集为例,提出一种改进的Spectrum算法。首先,使用基于特征加权的相似度矩阵,避免忽视每个样本/特征所对应的不同特征值大小在该样本中所占据的权重;其次,将拉普拉斯矩阵替换为Hessian矩阵,避免传统谱聚类的灵敏度问题,将ISODATA聚类算法代替原本的K-means算法,有效地调整聚类中心数K。试验结果表明,GMPR+改进Spectrum在2型糖尿病中的标准化互信息（NMI）为0.423,戴维森堡丁指数（DBI）为4.751,Calinski-Harabasz指标（CH）为25.541,兰德指数（RI）为0.835,调整兰德指数（ARI）为0.019,较改进前的效果有所提升,并且该算法可以识别出不同类型患病人群在肠道菌群上的结构差异,挖掘出肠道微生物组的关键细菌。     

基于功率谱信息熵与GK模糊聚类的生物组织变性识别方法

本文提出了一种基于功率谱信息熵和GK模糊聚类相结合的方法,应用于生物组织变性识别判定中。利用高强度聚焦超声辐照离体新鲜猪肉来改变其特性,并用热电偶测量声焦区温度,同时采集了不同温度时的超声回波信号。通过对采集的信号进行截取,研究了分段数对功率谱信息熵辨识性能的影响。研究发现当分段数为26至32时,功率谱信息熵的准确度、灵敏度和特异度均有较高的值。本研究计算了分段数为30时的功率谱信息熵,此时的变性组织对应的功率谱信息熵比未变性组织对应的功率谱信息熵平均高出约0. 094,大约为7. 99%。采用功率谱信息熵作为特征参数时,GK模糊聚类效果优于模糊C均值聚类;采用GK模糊聚类方式,功率谱信息熵比小波熵具有更高的辨识率。     

用模糊聚类探讨福建三明格氏栲自然保护区植被类型的划分

初步统计三明格氏栲自然保护区有种子植物１４５科,４３４属,１０１０种（变种）。植物区系古老,地理成分以泛热和热带分成分较多,我们运用模糊聚类分析了保护区群落的划分,结果表明,划分的类型与用植物所在群落层中重要值来确定群落乔木层的优势种划分群系的方法基本相一致,计自然植被３个群系,格氏栲林,米储林,栲树林;人工植被５个群系：马尾松林,杉木林,建柏林,楠木林,毛竹林。     

直接模糊聚类的截取水平选择及其在植被数量分析中的应用

植被生态研究中常用的聚类法,是着眼于研究区域中植被和环境因子呈间断分布或变化梯度较大的一类情况,对原始数据中各个体按其属性进行归类。直接模糊聚类法则以各个体间的属性相近程度来定义一模糊关系矩阵,然后对矩阵取不同的水平截集,从而得出一等级分类。当模糊关系确定以后,截取水平的选择就成了聚类结果的决定性因素。至目前为止,直接模糊聚类中的截取水平通常由分析者主观给定,或者是以逐步试验,逐步修改的方法确定的。这样,聚类结果就不可避免地带有较大的主观和任意性。笔者认为截取水平应选在模糊关系变化较大之处,使聚类结果尽可能地反映原始数据的结构特征。这一原理已被实施于一通用软件中,实例分析表明,如此选择的截取水平确能比较客观地反映原始数据的特征,从而得出较为合理的聚类结果。     

基于聚类系数的DNA序列性质的复杂网络研究

利用一种新的基于图论理论的DNA序列(片段)分析的方法,即通过复杂网络研究生物体的拓扑结构,主要通过测量聚类系数(集团系数)构建网络的拓扑结构。依据DNA序列的前缀、后缀关联性质构造了所选取DNA序列(片段)的相关网络,发现该网络分布满足幂率特征,有较大的聚类系数。结果表明构建得到的网络同时满足小世界网络和无尺度网络的特征,证明DNA序列不全是随机的序列,而是有随机扰动的确定结构的序列。     

三维超声心脏图像的模糊聚类分割

采用三维超声心动图对小儿先天性心脏病进行诊断与治疗能达到比传统二维超声心动图更直观的效果。然而由于超声图像质量较差,三维超声心动图的可视化效果往往无法达到医生的要求。本文对三维超声心脏图像进行分割,以改进超声图像的可视化效果,并为参数提取等提供基础。首先采用快速的模糊c均值聚类得到初始分割结果;然后利用图像多分辨率技术进行修正;接着结合图像的对比度进行进一步的分割;最后,把处理后的图像用绘制的方法显示出来。本文的结果对超声图像的可视化效果有一定的改善.     

白桦核心种质构建的聚类方法研究

以白桦240个家系的胸径、树高、材积和纤维素含量数据为依据,采用马氏距离计算家系间距离、10%的取样比例和优先取样法,研究了最短距离法、最长距离法、中间距离法、重心法、类平均法、加权配对算术平均法、可变法和离差平方和聚类法建构的核心种质与原种质的遗传参数、性状相关性及分布格局.结果表明,最短距离法构建白桦初级核心种质均值差异百分率、极差符合率、方差差异百分率和极差符合率分别为0、100%、75%和143%,4个性状相关性显著、相关系数均超过0.5,保持了原种质资源的空间分布格局,是构建白桦核心种质最佳方法.     

A heterogeneity measure for cluster identification with application to disease mapping

Mapping of disease incidence has long been of importance to epidemiology and public health. In this paper, we consider identification of clusters of spatial units with elevated disease rates and develop a new approach that estimates the relative disease risk in association with potential risk factors and simultaneously identifies clusters corresponding to elevated risks. A heterogeneity measure is proposed to enable the comparison of a candidate cluster and its complement under a pair of complementary models. A quasi-likelihood procedure is developed for estimating the model parameters and identifying the clusters. An advantage of our approach over traditional spatial clustering methods is the identification of clusters that can have arbitrary shapes due to abrupt or noncontiguous changes while accounting for risk factors and spatial correlation. Asymptotic properties of the proposed methodology are established and a simulation study shows empirically sound finite-sample properties. The mapping and clustering of enterovirus 71 infections in Taiwan are carried out for illustration.     

An automatic method involving cluster analysis of secondary structures for the identification of domains in proteins.

With a growing number of structures available in the Brookhaven Protein Data Bank, automatic methods for domain identification are required for the construction of databases. Domains are considered to be clusters of secondary structure elements. Thus, helices and strands are first clustered using intersecondary structural distances between C alpha positions, and dendrograms based on this distance measure are used to identify domains. Individual domains are recognized by a disjoint factor, which enables the automatic identification and classification into disjoint, interacting, and conjoint domains. Application to a database of 83 protein families and 18 unique structures shows that the approach provides an effective delineation of boundaries and identifies those proteins that can be considered as a single domain. A quantitative estimate of the interaction between domains has been proposed. The database of protein domains is a useful tool for understanding protein folding, for recognizing protein folds, and for understanding structure-activity relationships.     

OptimClass: Using species‐to‐cluster fidelity to determine the optimal partition in classification of ecological communities

Question: Community ecologists are often confronted with multiple possible partitions of a single set of records of species composition and/or abundances from several sites. Different methods of numerical classification produce different results, and the question is which of them, and how many clusters, should be selected for interpretation. We demonstrate a new method for identifying the optimal partition from a series of partitions of the same set of sites, based on number of species with high fidelity to clusters in a partition (faithful species). Methods: The new method, OptimClass, has two variants. OptimClass 1 searches the partition with the maximum number of faithful species across all clusters, while OptimClass 2 searches the partition with the maximum number of clusters that contain at least a preselected minimum number of faithful species. Faithful species are determined based on the P value of the Fisher's exact test, as a measure of fidelity. OptimClass was tested on three vegetation datasets that varied in species richness and internal heterogeneity, using several classification algorithms, resemblance measures and cover transformations. Results: Results from both variants of OptimClass depended on the preselected threshold P value for faithful species: higher P gave higher probability that a partition with more clusters was selected as optimal. Good partitions, in terms of OptimClass criteria, involved flexible beta clustering, and also ordinal clustering. Good partitions were also obtained with TWINSPAN when the required number of clusters was small, or UPGMA when the required number of clusters was large. Poor partitions usually resulted from classifications that used resemblance measures and cover transformations emphasizing differences in species cover; this is not unexpected because OptimClass uses a presence/absence‐based fidelity measure. Conclusions: If the aim of a classification is to obtain clusters rich in faithful species, which can be subsequently used as diagnostic species for identification of community types, OptimClass is a suitable method for simultaneous choice of the optimal classification algorithm and optimal number of clusters. It can be computed in the JUICE program.     

An efficient method for measuring copy number variation applied to improvement of nematode resistance in soybean

Copy number variation (CNV) is implicated in important traits in multiple crop plants, but can be challenging to genotype using conventional methods. The Rhg1 locus of soybean, which confers resistance to soybean cyst nematode (SCN), is a CNV of multiple 31.2‐kb genomic units each containing four genes. Reliable, high‐throughput methods to quantify Rhg1 and other CNVs for selective breeding were developed. The CNV genotyping assay described here uses a homeologous gene copy within the paleopolyploid soybean genome to provide the internal control for a single‐tube TaqMan copy number assay. Using this assay, CNV in breeding populations can be tracked with high precision. We also show that extensive CNV exists within Fayette, a released, inbred SCN‐resistant soybean cultivar with a high copy number at Rhg1 derived from a single donor parent. Copy number at Rhg1 is therefore unstable within a released variety over a relatively small number of generations. Using this assay to select for individuals with altered copy number, plants were obtained with both increased copy number and increased SCN resistance relative to control plants. Thus, CNV genotyping technologies can be used as a new type of marker‐assisted selection to select for desirable traits in breeding populations, and to control for undesirable variation within cultivars.