一种基于ZBDD结构的Top—k挖掘算法

研究微阵列数据中挖掘Top—k频繁闭合项集问题,并设计挖掘算法ZDtoP。算法采用ZBDD结构压缩存储数据集,使用自顶向下深度优先搜索策略挖掘项集长度不小于给定值min_l的Top—k频繁闭合项集,并对搜索空间进行有效修剪。通过实例证明该算法是正确有效的。     

A genetic algorithm-based method for feature subset selection

As a commonly used technique in data preprocessing, feature selection selects a subset of informative attributes or variables to build models describing data. By removing redundant and irrelevant or noise features, feature selection can improve the predictive accuracy and the comprehensibility of the predictors or classifiers. Many feature selection algorithms with different selection criteria has been introduced by researchers. However, it is discovered that no single criterion is best for all applications. In this paper, we propose a framework based on a genetic algorithm (GA) for feature subset selection that combines various existing feature selection methods. The advantages of this approach include the ability to accommodate multiple feature selection criteria and find small subsets of features that perform well for a particular inductive learning algorithm of interest to build the classifier. We conducted experiments using three data sets and three existing feature selection methods. The experimental results demonstrate that our approach is a robust and effective approach to find subsets of features with higher classification accuracy and/or smaller size compared to each individual feature selection algorithm.     

基于遗传算法的多任务学习

机器学习中冗余特征会降低学习器的性能,而特征选择方法可以去掉一些冗余特征.然而,冗余特征也包含有用信息,因此可以利用多任务学习的概念,通过重复利用冗余特征提高预测精度.但是,如何确定哪些特征作为输入和输出仍然是一个待解决的问题.之前的工作是在多任务学习当中,运用遗传算法来确定哪些特征作为输入,哪些作为输出,取得了较好的效果,但是该算法不足之处是没有考虑到不相关特征.现将特征分为三部分:输入的特征、输出的特征和不相关特征,提出了对一个特征进行双位编码的遗传算法搜索策略.在基因芯片数据上的实验结果表明,提出的新算法e-GA-MTL比已有基于遗传算法的GA-M-TL和其它启发式方法效果更好.     

Characterization of Optimal Aptamer‐Microarray Binding Chemistry and Spacer Design

Compared to conventional protein microarrays, the aptamer microarray is a relatively new and competitive method for analytical applications. The three‐dimensional folded aptamers show excellent binding specificity and affinity, and thus, can be used as an alternative to antibodies. Immobilization and binding effects of aptamers under different combinations of surfaces and spacers for aptamer‐microarray applications are investigated. In addition, the effects of spacers integrated at the terminal position of aptamers on made in‐house and commercially available microarray supports are explored.     

演化超网络在多类型癌症分子分型中的应用

该文提出一种用于多类型癌症分子分型的演化超网络模式识别方法。首先采用“一对多”方法,将一个多类分型问题转化为多个二类分型问题;然后利用信噪比方法对DNA微阵列数据进行信息基因选择;经过超网络对训练集的演化学习,构造一系列二类分类器并进行集成,最终构建一个多类型癌症分型系统并对待测样本进行分类。对急性白血病、儿童小圆蓝细胞肿瘤和GCM数据集实验结果表明：演化超网络留一交叉验证(LOOCV)识别率分别为：98.61%,100%和85.35%。演化超网络有利于挖掘癌症相关基因,具有良好的学习结果可读性。     

Development of a DNA microarray for detection and identification of Legionella pneumophila and ten other pathogens in drinking water

The safety and accessibility of drinking water are major concerns throughout the world. Consumption of water contaminated with infectious agents, toxic chemicals or radiological hazards represents a significant health risk and is strongly associated with mortality. Therefore, we have developed an oligonucleotide-based microarray using the sequences of 16S-23S rDNA internal transcribed spacer regions (ITS) and the gyrase subunit B gene (gyrB) found in the most prevalent and devastating waterborne pathogenic agents. This new diagnostic contains 26 specific probes and can simultaneously detect Aeromonas hydrophila, Klebsiella pneumoniae, Legionella pneumophila, Pseudomonas aeruginosa, Salmonella spp., Shigella spp., Staphylococcus aureus, Vibrio choleraeo, Vibrio parahaemolyticus, Yersinia enterocolitica and Leptospira interrogans. Testing was carried out against a total of 218 bacterial strains, including 53 representative strains, 103 clinical isolates and 62 strains of other bacterial species belonging to 10 genera and 48 species. The results were specific and reproducible, with a detection sensitivity of 0.1 ng DNA or 10⁴ CFU/ml achieved for pure cultures of each target organism. The diluted cultures and real drinking water samples were tested by the microarray with 100% accuracy. This novel diagnostic method is superior in time- and labor-efficiency to conventional bacterial culture and antiserum agglutination, and can be readily applied to epidemiological surveillance and other food safety applications.     

Gene expression profiling of HepG2 cells after treatment with black tea polyphenols

This study aimed to determine the effects of black tea polyphenols on gene expression in hepatocellular cancer cells. The total RNA from HepG2 hepatocellular cancer cells treated with black tea polyphenols was subjected to Human 14K cDNA microarray analysis. Real-time PCR and Western blot analysis were conducted to verify microarray data. Black tea polyphenols treatment at the dose of 20 mg/L, 40 mg/L or 80 mg/L for one to three days inhibited the growth of HepG2 cells in a dose and time dependent manner. A total of 48 genes showed more than two-fold change after black tea polyphenols treatment, including 17 upregulated genes and 31 downregulated genes, and they were involved in the regulation of cell growth, cell cycle, apoptosis, signaling, angiogenesis, tumor invasion and metastasis. Real-time PCR analysis of the selected genes showed that their mRNA expression changes were consistent with the microarray data. In addition, Western blot analysis of the selected genes showed that their protein expression changes were consistent with mRNA expression. In conclusion, gene expression profiles provide comprehensive molecular mechanisms by which black tea polyphenols exerts growth inhibition effects on cancer cells. The novel molecular targets identified in this study may be further exploited as therapeutic strategies for hepatocellular cancer.     

MicroClAn: Microarray clustering analysis

Evaluating clustering results is a fundamental task in microarray data analysis, due to the lack of enough biological knowledge to know in advance the true partition of genes. Many quality indexes for gene clustering evaluation have been proposed. A critical issue in this domain is to compare and aggregate quality indexes to select the best clustering algorithm and the optimal parameter setting for a dataset. Furthermore, due to the huge amount of data generated by microarray experiments and the requirement of external resources such as ontologies to compute biological indexes, another critical issue is the performance decline in term of execution time. Thus, the distributed computation of algorithms and quality indexes becomes essential. Addressing these issues, this paper presents the MicroClAn framework, a distributed system to evaluate and compare clustering algorithms using the most exploited quality indexes. The best solution is selected through a two-step ranking aggregation of the ranks produced by quality indexes. A new index oriented to the biological validation of microarray clustering results is also introduced. Several scheduling strategies integrated in the framework allow to distribute tasks in the grid environment to optimize the completion time. Experimental results show the effectiveness of our aggregation strategy in identifying the best rank among different clustering algorithms. Moreover, our framework achieves good performance in terms of completion time with few computational resources.     

Simultaneous detection of sulfamethazine,streptomycin, and tylosin in milk by microplate-array based SMM–FIA

This paper presents an approach to simultaneously detect sulfamethazine, streptomycin, and tylosin in milk by indirect competitive multianalyte Fluorescence immunoassay (FIA). Microscope glass slides modified with agarose were used for the preparation of small molecule microarrays (SMMs). Bovine serum albumin (BSA) conjugates of the haptens were immobilized on glass slides. The system consists of four glass slides containing 96 wells formed by an enclosing hydrophobic mask, which precisely matches a standard microplate. All liquid handling and sample processing were fully automated as 96-wells ELISA format. Monoclonal antibodies against sulfamethazine, streptomycin, and tylosin allowed the simultaneous detection of the respective analytes. Antibody binding was detected by a second antibody labeled with Cy5 generating fluorescence, which was scanned with chip scanner. The detection limits for three analytes were 3.26 ng/ml (sulfamethazine), 2.01 ng/ml (streptomycin), and 6.37 ng/ml (tylosin), being far below the respective MRLs. The system proved to be the first SMM–FIA platform having the potential to test for numerous antibiotics in parallel, such being of considerable interest for the control of safety in the food industry.