基于遗传算法的基因表达数据的K-均值聚类分析

聚类算法在基因表达数据的分析处理过程中得到日益广泛的应用.本文通过把 K-均值聚类算法引入到遗传算法中,结合基因微阵列的特点,来讨论一种基于遗传算法的K-均值聚类模型,目的是利用遗传算法的全局性来提高聚类算法找到全局最优的可能性,实验结果证明,该算法可以很好地解决某些基因表达数据的聚类分析问题.     

GeneSifter在基因表达谱芯片数据挖掘中的应用

介绍一款基因芯片数据分析工具——GeneSifter软件。具有快速、直观、便捷等特点，尤其适用于基因表达谱的数据挖掘。芯片数据一般以格式化文本文档形式上载，根据实验目的、设计不同，总共有4种上载向导工具，数据分析从控制台Analysis项目下的Pairwise或Projects进入，需要设置滤过、阈值和统计分析等参数，Pairwise可获得的结果有：差异显著性基因列表、基因本体报告和KEGG通路报告等，Projects有更为强大的功能，可获取聚类等6种结果。GeneSifter独特的一站式单击设置，可获得相关基因的11个数据库最新链接。GeneSifcer适用于基因芯片数据挖掘的生物研究人员。     

肿瘤基因表达谱——肿瘤免疫学研究的新策略

基因表达谱代表了细胞中基因表达的状况。通过比较肿瘤细胞和相应正常组织细胞的基因表达谱所获得的信息 ,就可获得在肿瘤和正常细胞中差异表达的基因 ,进一步研究这些基因的结构和功能 ,对研究肿瘤的发生发展和肿瘤的临床诊断和治疗都具有重要的意义。本文着重介绍了目前常用的研究肿瘤基因表达谱的各种方法 ,包括 m RNA水平和蛋白质水平肿瘤基因表达谱的研究方法 ,国内外的最新研究进展及应用前景     

类风湿关节炎大鼠滑膜细胞的类肿瘤样增生与相关基因表达的关系

目的:探讨佐剂性关节炎(AA)大鼠滑膜细胞的类肿瘤样增生和相关基因表达的机制。方法:将Wistar大鼠随机分两组,每组10只,并应用弗氏完全佐剂诱发大鼠AA实验性模型,定期观测踝关节肿胀度至第40天。分别取AA模型组和正常对照组大鼠踝关节进行组织切片,用HE染色后观察病理学变化。取大鼠膝关节滑膜组织体外进行原代细胞培养,用MTT比色法检测大鼠滑膜细胞的增生。同时采用RTPCR法,检测模型组及正常对照组大鼠滑膜组织中相关基因Cmyc和ODCmRNA的转录水平。结果:①体外培养的AA模型组大鼠关节滑膜细胞的增殖显著高于正常组大鼠(P<0.01),并呈类肿瘤样增生。②AA模型组大鼠关节滑膜组织中,Cmyc和ODCmRNA的转录水平明显高于正常对照大鼠(P<0.01)。结论:AA模型组大鼠关节滑膜细胞的类肿瘤样增生,可能与其相关基因Cmyc和ODCmRNA的转录水平增高有关。     

基因芯片数据的聚类分析方法在疾病诊断中的应用

基因芯片技术是后基因组时代功能基因组研究的主要工具,如何处理和分析这些数据并从中提取出有价值的生物学信息是一个极为重要的问题。聚类分析是基因芯片数据分析中广泛使用的一类方法，本文就基因芯片数据的聚类分析方法及其在疾病诊断中的应用作一综述。     

神经网络对卵巢粘液性囊腺肿瘤细胞识别的研究

目的：探讨人工神经网络在识别卵巢粘液性囊腺肿瘤细胞的价值。方法：使用MATALB软件中的神经网络工具箱（Neural Network Toolbox）,分别设计两种神经网络并训练。网络以五个形态参数（细胞核面积、周长、最大直径、等效圆直径、似圆度）为输入,输出为正常、良性、交界和恶性四种形态类型。结果：训练好的神经网络可以准确的对卵巢肿瘤细胞形态分类。结论：人工神经网络对卵巢粘液性囊腺肿瘤的病理学鉴别有很好的应用前景;选择适合的网络,不仅看网络自身,还要看训练样本。     

结合GO体系信息与芯片数据构建肿瘤特征基因网络

我们提出了一种新的肿瘤特征基因调控网络构建方法,首先用P-tree方法快速筛选肿瘤特征基因集;然后结合GO体系中对基因的注释信息与芯片数据信息对特征基因进行聚类;最后通过文献挖掘方法,以聚类获得的每个功能类中的基因为核心,建立肿瘤特征基因功能类网络模块.实验结果表明:本研究的方法明显提高了特征基因筛选速度,网络构建基于生物功能过程进一步细化,同时通过文献挖掘方法在网络中补充入大量与肿瘤特征基因发生直接或间接关系的基因,网络内容更加丰富与条理化.通过结合芯片数据,GO体系与相关文献中信息构建肿瘤特征基因调控网络能够构建更为细致、丰富,并针对具体调控过程的网络模块,为肿瘤的产生,发展与转移过程分析提供有效的指导.     

基于支持向量聚类的肿瘤表达谱分型识别算法

对肿瘤样本进行准确的分型识别是有效治疗肿瘤的前提。首先,利用方差滤波方法选择肿瘤表达谱中具有最大方差的部分基因作为识别特征集,然后,利用支持向量聚类对肿瘤表达谱进行分型识别。针对多类型样本情况和支持向量聚类中出现的孤立点聚类问题,分别提出了有效的解决办法。对两个肿瘤表达谱数据的测试结果显示,基于支持向量聚类的方法能够准确地对肿瘤样本进行分型识别,同时能够自动发现肿瘤样本真实的亚型数量。     

乳腺肿瘤超声图像中感兴趣区域的自动检测

由于斑点噪声、伪影以及病灶形状多变的影响,乳腺肿瘤超声图像中肿瘤区域的自动检测以及病灶的边缘提取比较困难,已有的方法主要是由医生先手工提取感兴趣区域(ROI)。本研究提出一种乳腺肿瘤超声图像中感兴趣区域自动检测的方法,选用超声图像的局部纹理、局部灰度共生矩阵以及位置信息作为特征,采用自组织映射神经网络进行分类,自动识别乳腺肿瘤区域。对包含168幅乳腺肿瘤超声图像的数据库进行识别的结果表明:该方法自动识别ROI的准确率达到86.9%,可辅助医生提取肿瘤的实际边缘以及进一步诊断。     

基于自组织神经网络的超声心脏图象分割

0　引言多维超声心脏图象能够提供大量高质量的心脏结构和功能信息 ,是一种十分有效的诊断工具。然而 ,困难的图象获取、较差的图象质量 ,特别是分割中的交互式操作 (手动分割 )限制了它的发展和临床应用[1] 。图象分割是图象处理与模式识别等领域中十分重要且又十分困难的问题 ,它包括特征提取与模式特征分类两部分。传统的分类方法是采用Ｋ ｍｅａｎｓ聚类方法[2 ] ,分割结果受初始聚类中心和样本空间的分布影响较大 ,很难取得较好的结果 ,往往需要根据经验选取初始聚类中心 ,不具有自适应性。本文采用Ｔ .Ｋｏｈｏｎｅｎ自组织特征映射神…     

基因表达数据分析

基因芯片或称微阵列 (gene chip或 microarrays)是最近分子生物学实验技术的一个突破 ,利用该技术可以对成千上万个基因的表达进行平行分析 ,已经产生了总量巨大的有用数据 ,分析与整理这些数据成为利用这一技术的一个主要瓶颈问题。原始的微阵列数据是图像 ,必须把它们转化成为基因表达矩阵表 ,矩阵的行代表基因 ,列代表各样本及条件(如 :组织、实验条件、处理因素等 ) ,每个格子的数据表示特定的基因在特定的样本中的表达水平。这些矩阵必须经过进一步的分析才能得到它们潜在生物学过程的信息。本文着重讨论了生物信息学方法在该领域的应用 ,主要探讨了监督与非监督数据分析方法在基因表达数据分析中的应用 ,如预测基因功能分类及恶性肿瘤分级。本文还讨论了如何利用基因表达矩阵预测序列中假想调控信号。     

Improving cluster visualization in self-organizing maps: application in gene expression data analysis

Cluster analysis is one of the crucial steps in gene expression pattern (GEP) analysis. It leads to the discovery or identification of temporal patterns and coexpressed genes. GEP analysis involves highly dimensional multivariate data which demand appropriate tools. A good alternative for grouping many multidimensional objects is self-organizing maps (SOM), an unsupervised neural network algorithm able to find relationships among data. SOM groups and maps them topologically. However, it may be difficult to identify clusters with the usual visualization tools for SOM. We propose a simple algorithm to identify and visualize clusters in SOM (the RP-Q method). The RP is a new node-adaptive attribute that moves in a two dimensional virtual space imitating the movement of the codebooks vectors of the SOM net into the input space. The Q statistic evaluates the SOM structure providing an estimation of the number of clusters underlying the data set. The SOM-RP-Q algorithm permits the visualization of clusters in the SOM and their node patterns. The algorithm was evaluated in several simulated and real GEP data sets. Results show that the proposed algorithm successfully displays the underlying cluster structure directly from the SOM and is robust to different net sizes.     

Knowledge-assisted recognition of cluster boundaries in gene expression data

BACKGROUND AND MOTIVATION: DNA microarray technology has made it possible to determine the expression levels of thousands of genes in parallel under multiple experimental conditions. Genome-wide analyses using DNA microarrays make a great contribution to the exploration of the dynamic state of genetic networks, and further lead to the development of new disease diagnosis technologies. An important step in the analysis of gene expression data is to classify genes with similar expression patterns into the same groups. To this end, hierarchical clustering algorithms have been widely used. Major advantages of hierarchical clustering algorithms are that investigators do not need to specify the number of clusters in advance and results are presented visually in the form of a dendrogram. However, since traditional hierarchical clustering methods simply provide results on the statistical characteristics of expression data, biological interpretations of the resulting clusters are not easy, and it requires laborious tasks to unveil hidden biological processes regulated by members in the clusters. Therefore, it has been a very difficult routine for experts. OBJECTIVE: Here, we propose a novel algorithm in which cluster boundaries are determined by referring to functional annotations stored in genome databases. MATERIALS AND METHODS: The algorithm first performs hierarchical clustering of gene expression profiles. Then, the cluster boundaries are determined by the Variance Inflation Factor among the Gene Function Vectors, which represents distributions of gene functions in each cluster. Our algorithm automatically specifies a cutoff that leads to functionally independent agglomerations of genes on the dendrogram derived from similarities among gene expression patterns. Finally, each cluster is annotated according to dominant gene functions within the respective cluster. RESULTS AND CONCLUSIONS: In this paper, we apply our algorithm to two gene expression datasets related to cell cycle and cold stress response in budding yeast Saccharomyces cerevisiae. As a result, we show that the algorithm enables us to recognize cluster boundaries characterizing fundamental biological processes such as the Early G1, Late G1, S, G2 and M phases in cell cycles, and also provides novel annotation information that has not been obtained by traditional hierarchical clustering methods. In addition, using formal cluster validity indices, high validity of our algorithm is verified by the comparison through other popular clustering algorithms, K-means, self-organizing map and AutoClass.     

Insights into TREM2 biology by network analysis of human brain gene expression data

Rare variants in TREM2 cause susceptibility to late-onset Alzheimer's disease. Here we use microarray-based expression data generated from 101 neuropathologically normal individuals and covering 10 brain regions, including the hippocampus, to understand TREM2 biology in human brain. Using network analysis, we detect a highly preserved TREM2-containing module in human brain, show that it relates to microglia, and demonstrate that TREM2 is a hub gene in 5 brain regions, including the hippocampus, suggesting that it can drive module function. Using enrichment analysis we show significant overrepresentation of genes implicated in the adaptive and innate immune system. Inspection of genes with the highest connectivity to TREM2 suggests that it plays a key role in mediating changes in the microglial cytoskeleton necessary not only for phagocytosis, but also migration. Most importantly, we show that the TREM2-containing module is significantly enriched for genes genetically implicated in Alzheimer's disease, multiple sclerosis, and motor neuron disease, implying that these diseases share common pathways centered on microglia and that among the genes identified are possible new disease-relevant genes.     

Cancer adjuvant chemotherapy strategic classification by artificial neural network with gene expression data: An example for non-small cell lung cancer

PurposeAdjuvant chemotherapy (ACT) is used after surgery to prevent recurrence or metastases. However, ACT for non-small cell lung cancer (NSCLC) is still controversial. This study aimed to develop prediction models to distinguish who is suitable for ACT (ACT-benefit) and who should avoid ACT (ACT-futile) in NSCLC.MethodsWe identified the ACT correlated gene signatures and performed several types of ANN algorithms to construct the optimal ANN architecture for ACT benefit classification. Reliability was assessed by cross-data set validation.ResultsWe obtained 2 probes (2 genes) with T-stage clinical data combination can get good prediction result. These genes included 208893_s_at (DUSP6) and 204891_s_at (LCK). The 10-fold cross validation classification accuracy was 65.71%. The best result of ANN models is MLP14-8-2 with logistic activation function.ConclusionsUsing gene signature profiles to predict ACT benefit in NSCLC is feasible. The key to this analysis was identifying the pertinent genes and classification. This study maybe helps reduce the ineffective medical practices to avoid the waste of medical resources.     

TA-clustering: cluster analysis of gene expression profiles through Temporal Abstractions

This paper describes a new technique for clustering short time series of gene expression data. The technique is a generalization of the template-based clustering and is based on a qualitative representation of profiles which are labelled using trend Temporal Abstractions (TAs); clusters are then dynamically identified on the basis of this qualitative representation. Clustering is performed in an efficient way at three different levels of aggregation of qualitative labels, each level corresponding to a distinct degree of qualitative representation. The developed TA-clustering algorithm provides an innovative way to cluster gene profiles. We show the developed method to be robust, efficient and to perform better than the standard hierarchical agglomerative clustering approach when dealing with temporal dislocations of time series. Results of the TA-clustering algorithm can be visualized as a three-level hierarchical tree of qualitative representations and as such easy to interpret. We demonstrate the utility of the proposed algorithm on a set of two simulated data sets and on a study of gene expression data from S. cerevisiae.