A hybrid feature selection method for DNA microarray data

Gene expression profiles, which represent the state of a cell at a molecular level, have great potential as a medical diagnosis tool. In cancer classification, available training data sets are generally of a fairly small sample size compared to the number of genes involved. Along with training data limitations, this constitutes a challenge to certain classification methods. Feature (gene) selection can be used to successfully extract those genes that directly influence classification accuracy and to eliminate genes which have no influence on it. This significantly improves calculation performance and classification accuracy. In this paper, correlation-based feature selection (CFS) and the Taguchi-genetic algorithm (TGA) method were combined into a hybrid method, and the K-nearest neighbor (KNN) with the leave-one-out cross-validation (LOOCV) method served as a classifier for eleven classification profiles to calculate the classification accuracy. Experimental results show that the proposed method reduced redundant features effectively and achieved superior classification accuracy. The classification accuracy obtained by the proposed method was higher in ten out of the eleven gene expression data set test problems when compared to other classification methods from the literature.     

Ensemble gene selection by grouping for microarray data classification

Selecting relevant and discriminative genes for sample classification is a common and critical task in gene expression analysis (e.g. disease diagnostic). It is desirable that gene selection can improve classification performance of learning algorithm effectively. In general, for most gene selection methods widely used in reality, an individual gene subset will be chosen according to its discriminative power. One of deficiencies of individual gene subset is that its contribution to classification purpose is limited. This issue can be alleviated by ensemble gene selection based on random selection to some extend. However, the random one requires an unnecessary large number of candidate gene subsets and its reliability is a problem. In this study, we propose a new ensemble method, called ensemble gene selection by grouping (EGSG), to select multiple gene subsets for the classification purpose. Rather than selecting randomly, our method chooses salient gene subsets from microarray data by virtue of information theory and approximate Markov blanket. The effectiveness and accuracy of our method is validated by experiments on five publicly available microarray data sets. The experimental results show that our ensemble gene selection method has comparable classification performance to other gene selection methods, and is more stable than the random one.     

采用高通量基因芯片筛选乳腺癌分子诊断基因群的研究

早期诊断是提高乳腺癌治愈率和生存率的关键.肿瘤细胞的发生发展过程中涉及复杂的生物学过程.基因芯片技术可以在基因组范围内对组织细胞的基因表达谱进行研究,为肿瘤发生发展中多基因改变的分子机制研究提供了有力的工具[1].我们采用高通量基因表达谱芯片通过比较乳腺原发癌与其配对的癌旁正常乳腺组织基因表达差异筛选乳腺癌相关基因,并探讨其作为乳腺癌分子诊断基因标志群的临床意义.     

DNA microarray analysis reveals novel gene expression profiles in collagen-induced arthritis

Global gene expression was analyzed in early and late collagen-induced arthritis (CIA). Of 8734 cDNAs analyzed, 330 were induced and 55 downregulated greater than twofold in early or late disease. Hierarchical clustering of these 385 cDNAs demonstrated five distinct expression patterns differentiating early from late disease and correlating with histopathologic changes in the paw. Of the 385 cDNAs, 185 are known, characterized genes, the majority of which are not described as playing a role in arthritis. However, several of these genes are involved in pathological processes relating to arthritis, including apoptosis, inflammation, and cellular proliferation. One interesting gene, follistatin-like gene, is highly expressed along the margin of contact between inflammatory synovial pannus and eroding bone, suggesting a role in joint destruction. These results demonstrate that global gene expression profiles distinguish early and late CIA and reveal several genes novel to arthritis the further characterization of which will advance our understanding of arthritis.     

Analysis of Mycobacterium leprae gene expression using DNA microarray

Mycobacterium leprae, the causative agent of leprosy, does not grow under in vitro condition, making molecular analysis of this bacterium difficult. For this reason, bacteriological information regarding M. leprae gene function is limited compared with other mycobacterium species. In this study, we performed DNA microarray analysis to clarify the RNA expression profile of the Thai53 strain of M. leprae grown in footpads of hypertensive nude rats (SHR/NCrj-rnu). Of 1605 M. leprae genes, 315 showed signal intensity twofold higher than the median. These genes include Acyl-CoA metabolic enzymes and drug metabolic enzymes, which might be related to the virulence of M. leprae. In addition, consecutive RNA expression profile and in silico analyses enabled identification of possible operons within the M. leprae genome. The present results will shed light on M. leprae gene function and further our understanding of the pathogenesis of leprosy.     

DNA芯片检测乙型肝炎病毒基因多态性

目的　建立DNA芯片检测乙型肝炎病毒 (hepatitisBvirus,HBV)基因多态性的研究方法并对实验条件进行优化。方法　设计多条寡核苷酸探针 ,在硅烷化芯片的特定位置上 ,用点样仪将探针固定 ,并与PCR扩增的HBV基因相应区段杂交 ,杂交结果影印至硝酸纤维素膜 ,经BCIP NBT避光显色 ,用放大镜观察杂交信号呈暗紫色圆点 ,根据特定位置上杂交信号的有无和与之相应的探针序列来判定基因突变的类型。结果　通过 1次杂交反应可检测HBV前C C区 (nt 1896 1814 )、BCP区 (nt1762 1764)和P区 (nt 52 8 552 )等多个位点的变异 ,与测序分析结果完全一致 ,具有较好的检测灵敏度和重复性。结论　DNA芯片检测HBV基因常见突变位点多态性 ,操作简便易行 ,技术要求不高 ,具有临床推广应用价值 ,而且可以方便地通过向寡核苷酸探针阵列中添加相应探针 ,扩大基因芯片的检测应用范围 ,为临床检测提供了新的方法     

应用基因芯片技术筛查人子宫内膜容受性相关基因

目的探讨人子宫内膜容受性相关基因的差异表达。方法应用含14000条基因的cDNA表达谱基因芯片分析分泌早期与分泌中期子宫内膜基因的差异表达。结果所检测的14000个基因中,分泌早期子宫内膜与分泌中期子宫内膜之间存在显著差异表达基因313个。其中,分泌中期子宫内膜表达上调基因数为175个,下调基因数为138个。结论子宫内膜容受性的建立受许多因素的调控,应用基因芯片技术可以快速、高通量的筛查出相关基因,从而有可能找到合适的分子标志物作为子宫内膜容受性的临床诊断指标。     

Inflammatory gene expression patterns revealed by DNA microarray analysis in TNF-alpha-treated SGBS human adipocytes

We report here the use of human inflammation arrays to study the inflammatory gene expression profile of TNF-alpha- treated human SGBS adipocytes. Human preadipocytes (SGBS) were induced to differentiate in primary culture, and adipocyte differentiation was confirmed, using Oil Red O staining. We treated the differentiated adipocytes with TNF-alpha, and RNA from differentiated adipocytes with or without TNF-alpha treatment was hybridized to MWG human inflammation arrays to compare expression profiles. Eleven genes were up- or down-regulated in TNF-alpha-treated adipocytes. As revealed by array analysis, among 6 up-regulated genes, only eotaxin-1, monocyte chemoattractant protein-1 (MCP-1), and vascular cell adhesion molecule 1 isoform a precursor (VCAM1) were confirmed by real-time polymerase chain reaction (PCR). Similarly, among 5 down-regulated genes, only IL-1 family member 5 (IL1F5), a disintegrin and metalloprotease with thrombospondin motifs-1 preproprotein (ADAMTS1), fibronectin 1 isoform 1 preprotein (FN1), and matrix metalloproteinase 15 preprotein (MMP15) were confirmed by real-time PCR. There was a substantial increase (50-fold) in eotaxin-1 in response to TNF-alpha. Taken together, we have identified several inflammatory molecules expressed in SGBS adipocytes and discovered molecular factors explaining the relationship between obesity and atherosclerosis, focusing on inflammatory cytokines expressed in the TNF-alpha-treated SGBS cells. Further investigation into the role of these up- or down-regulated cytokine genes during the pathological processes leading to the development of atherosclerosis is warranted.     

Identification of genes differentially expressed in cultured human osteoblasts versus human fibroblasts by DNA microarray analysis

Little is known about patterns of gene expression from cells populating the connective tissues. This study investigated the possible variance of gene expression profile between human osteoblasts (HO) and human fibroblasts (HF) in vitro, using DNA microarray technology. Clustering identification was used to compare expression patterns between HO and HF for biological significance. Our results showed that genes encoding the extracellular matrix or apoptosis-related proteins tended to be expressed in greater abundance in HO, while more proteolysis-related proteins were expressed in higher level in HF. Significant differences in expression were also noted with genes related to signaling pathways. To confirm the array results, three genes (periostin, MFG-E8, MMP-10) were selected and analyzed independently by RT-PCR and northern blot. The results were found consistent with the array data in HO and HF. The present findings suggest that HO and HF differ not only phenotypically but in the expression level of tissue specific genes to assure the turnover and homeostasis of their respective tissues.     

DNA microarray applications in functional genomics

The successful completion of the Human Genome Project and the achievement of similar goals in other species have generated a huge amount of free available information about the genomic sequence of different organisms, opening the door to a postgenome era where new challenges arise. One of the most ambitious objectives of this new period, addressed by the emerging discipline of functional genomics, attempts to understand the genome and the products it encodes for, and how these gene products interact to produce complex living organisms. This new era is also characterized by the development of new technologies, which have produced genomic tools indispensable for understanding how gene products are regulated in normal and diseased conditions on a global genome scale. One of these technologies is DNA microarrays, turned into a very popular tool in the last years. Although the most common use of DNA microarrays is gene expression profiling, scientists have successfully used them for multiple applications, including genotyping, sequencing, DNA copy number analysis, and DNA-protein interactions, among others. In summary, DNA microarrays are changing the way biomedicine and other disciplines are addressing different biological questions and will allow the translation of genome research to the clinic.     

An integrated algorithm for gene selection and classification applied to microarray data of ovarian cancer

OBJECTIVE: The type of data in microarray provides unprecedented amount of data. A typical microarray data of ovarian cancer consists of the expressions of tens of thousands of genes on a genomic scale, and there is no systematic procedure to analyze this information instantaneously. To avoid higher computational complexity, it needs to select the most likely differentially expressed gene markers to explain the effects of ovarian cancer. Traditionally, gene markers are selected by ranking genes according to statistics or machine learning algorithms. In this paper, an integrated algorithm is derived for gene selection and classification in microarray data of ovarian cancer. METHODS: First, regression analysis is applied to find target genes. Genetic algorithm (GA), particle swarm optimization (PSO), support vector machine (SVM), and analysis of variance (ANOVA) are hybridized to select gene markers from target genes. Finally, the improved fuzzy model is applied to classify cancer tissues. RESULTS: The microarray data of ovarian cancer, obtained from China Medical University Hospital, is used to test the performance of the proposed algorithm. In simulation, 200 target genes are obtained after regression analysis and six gene markers are selected from the hybrid process of GA, PCO, SVM and ANOVA. Additionally, these gene markers are used to classify cancer tissues. CONCLUSIONS: The proposed algorithm can be used to analyze gene expressions and has superior performance in microarray data of ovarian cancer, and it can be performed on other studies for cancer diagnosis.     

Novel function of ovarian growth factors: combined studies by DNA microarray, biochemical and physiological approaches

Owing to the development of the DNA microarray technique, modulation of gene function can be studied systematically. Considerable attention has been focused on members of the growth factor family to elucidate the main regulators of oocyte maturation and ovarian follicle rupture. Among these growth factors, it was found both in rodents and in humans that amphiregulin (Ar) and epiregulin (Ep) of the epidermal growth factor (EGF) family were dramatically up-regulated by gonadotrophins in the intact ovary and in primary granulosa cells, respectively. Their role in cumulus expansion and oocyte maturation was established in rodents, and their formation under LH stimulation in granulosa cells was demonstrated in humans. To be activated, Ar and Ep must be cleaved by A Disintegrin And Metalloproteinases (ADAMs) family. However, the precise processing of Ar and Ep by the cumulus cells is still obscure. Future investigations using DNA microarray technique may reveal the repertoire of genes activated in Ar- and Ep-stimulated cumulus cells and may help elucidate the molecular basis of ovulation.     

Dystrophin基因缺失检测微阵列构建及应用中技术优化的研究

目的建立Dystrophin基因缺失检测的DNA微阵列技术，并优化其构建及应用中的关键技术条件。方法提取样品基因组DNA，Klenow法随机引物扩增同时FITC荧光标记，并比较生物素标记法。将FITC荧光标记的核酸点样于不同方法活化处理的玻片上并用不同浓度的盐溶液洗脱，测试对DNA固定效率的影响。以分子克隆法获得的Dystrophin基因18个易缺失外显子cDNA片段为探针、APES和poly-Lys联合处理的玻片为基片，制备简易微阵列。分别与DMD/BMD患者及健康人荧光标记的基因组DNA杂交，检测微阵列的质量和评估结果的可靠性。结果APES和poly-Lys联合处理的玻片对核酸固定效率最高，核酸在洗涤过程中的脱落程度随洗液盐浓度的增加而增大；FITC标记步骤简便，生物素标记相对烦琐，但生物素标记杂交后的荧光强度明显高于FITC标记；微阵列杂交信号信噪比较好，各种对照结果满意，检测结果与PCR一致。结论优化了DNA微阵列构建及应用过程中基片表面活化、探针固定、靶基因DNA扩增标记、杂交条件、杂交信号检测分析等方法，为更好地应用微阵列进行Dystrophin基因诊断奠定基础。     

Evaluation of differential gene expression in Leishmania major Friedlin procyclics and metacyclics using DNA microarray analysis

The various Leishmania species are flagellated protozoans, responsible for a wide spectrum of human diseases. The sequence of the L. major genome is nearing completion and a large proportion of the identified genes have yet to be ascribed functions. DNA microarrays containing PCR-amplified DNA from a random amplified genomic library of L. major Friedlin (LmjF) [Mol. Biochem. Parasitol. 113 (2001) 337] were hybridized with fluorescent probes made from L. major Friedlin RNA from five time-points during differentiation from procyclics to metacyclics. The data were normalized for background and probe intensity and the relative abundance of RNA for each spot was calculated. Almost 15% (1387/9282) of the DNAs showed statistically significant (P<0.01) changes in expression (1.1-5-fold) during the transition, with 1.16% (108) showing up-regulation at two or more time-points and 0.14% (13) showing down-regulation. Northern blot analyses of selected genes confirmed these results. These studies confirmed the stage-specific expression of several known genes, as well as identifying a number of novel genes that are up-regulated in either procyclics or metacyclics.     

DNA微阵列技术及其在生物医学中的应用

DNA微阵列技术是一种新型的有力工具 ,它可以同步、快速而有效地分析大量核苷酸杂交试验。DNA微阵列技术的应用包括 :基因表达分析、多基因突变和多态性检测、疾病诊断和分型、研究疾病分子机制以及药物筛选和疫苗开发等 ,本文简要介绍了 DNA微阵列技术并综述了其在这些方面的应用。