基于主曲线的微阵列数据分类

提出了一种基于主曲线(principal curves)的微阵列数据分类方法(PC)。主曲线是第一主成分的非线性推广,它是数据集合的“骨架”,数据集合是主曲线的“云”。基于主曲线的微阵列数据分类方法,首先利用专门设计的算法在训练数据集上计算出每类样本的主曲线,然后根据测试样本与各类样本主曲线距离的期望方差来确定测试样本所属的类别。实验结果表明,该分类方法在进行小样本微阵列数据分类时性能优于现有的方法。     

Transcriptional and functional responses of Escherichia coli O157:H7 growing in the lettuce rhizoplane

Lettuce and spinach are increasingly implicated in foodborne illness outbreaks due to contamination by Escherichia coli O157:H7. While this bacterium has been shown to colonize and survive on lettuce leaf surfaces, little is known about its interaction with the roots of growing lettuce plants. In these studies, a microarray analyses, mutant construction and confocal microscopy were used to gain an understanding of structure and function of bacterial genes involved in the colonization and growth of E. coli O157:H7 on lettuce roots. After three days of interaction with lettuce roots, 94 and 109 E. coli O157:H7 genes were significantly up- and down-regulated at least 1.5 fold, respectively. While genes involved in biofilm modulation (ycfR and ybiM) were significantly up-regulated, 40 of 109 (37%) of genes involved in protein synthesis were significantly repressed. E. coli O157:H7 was 2 logs less efficient in lettuce root colonization than was E. coli K12. We also unambiguously showed that a ΔycfR mutant of E. coli O157:H7 was unable to attach to or colonize lettuce roots. Taken together these results indicate that bacterial genes involved in attachment and biofilm formation are likely important for contamination of lettuce plants with Shiga toxin-producing E. coli strains.     

A Quick and Parallel Analytical Method Based on Quantum Dots Labeling for ToRCH-Related Antibodies

Quantum dot is a special kind of nanomaterial composed of periodic groups of II–VI, III–V or IV–VI materials. Their high quantum yield, broad absorption with narrow photoluminescence spectra and high resistance to photobleaching, make them become a promising labeling substance in biological analysis. Here, we report a quick and parallel analytical method based on quantum dots for ToRCH-related antibodies including Toxoplasma gondii, Rubella virus, Cytomegalovirus and Herpes simplex virus type 1 (HSV1) and 2 (HSV2). Firstly, we fabricated the microarrays with the five kinds of ToRCH-related antigens and used CdTe quantum dots to label secondary antibody and then analyzed 100 specimens of randomly selected clinical sera from obstetric outpatients. The currently prevalent enzyme-linked immunosorbent assay (ELISA) kits were considered as “golden standard” for comparison. The results show that the quantum dots labeling-based ToRCH microarrays have comparable sensitivity and specificity with ELISA. Besides, the microarrays hold distinct advantages over ELISA test format in detection time, cost, operation and signal stability. Validated by the clinical assay, our quantum dots-based ToRCH microarrays have great potential in the detection of ToRCH-related pathogens.     

Detecting Dense Subgraphs in Complex Networks Based on Edge Density Coefficient

Densely connected patterns in biological networks can help biologists to elucidate meaningful insights. How to detect dense subgraphs effectively and quickly has been an urgent challenge in recent years. In this paper, we proposed a local measure named the edge density coefficient, which could indicate whether an edge locates a dense subgraph or not. Simulation results showed that this measure could improve both the accuracy and speed in detecting dense subgraphs. Thus, the G-N algorithm can be extended to large biological networks by this local measure. Finally, we applied this algorithm to microarray data sets of Saccharomyces cerevisiae and performed the gene ontology analysis of the result by the GOEAST.     

Optimization of a parallel permutation testing function for the SPRINT R package

The statistical language R and its Bioconductor package are favoured by many biostatisticians for processing microarray data. The amount of data produced by some analyses has reached the limits of many common bioinformatics computing infrastructures. High Performance Computing systems offer a solution to this issue. The Simple Parallel R Interface (SPRINT) is a package that provides biostatisticians with easy access to High Performance Computing systems and allows the addition of parallelized functions to R. Previous work has established that the SPRINT implementation of an R permutation testing function has close to optimal scaling on up to 512 processors on a supercomputer. Access to supercomputers, however, is not always possible, and so the work presented here compares the performance of the SPRINT implementation on a supercomputer with benchmarks on a range of platforms including cloud resources and a common desktop machine with multiprocessing capabilities. Copyright © 2011 John Wiley & Sons, Ltd.     

采用喷墨打印机进行低成本微阵列印刷的可行性研究

根据网点形成技术,微阵列制作方法可归结为接触式印刷和非接触式印刷.接触式印刷中,如印刷生物样本时,设备与基底是直接接触.而非接触式印刷中,设备与基底之间不会直接接触.在非接触式印刷中,喷墨印刷技术是相对灵活、简便、便宜的一种方式,同时还减少了污染且避免了更高的数据吞吐量.但是,喷墨印刷需要有能喷射儿微微升墨滴的喷嘴和能够精确喷射墨滴的喷嘴位置控制单元.若使用喷墨方法制作微阵列就需要复杂的机构系统,以致成本增加.因此,本论文提出使用已具有喷墨系统的商用喷墨打印机来制作微阵列.     

Complex regulation of the aflatoxin biosynthesis gene cluster of Aspergillus flavus in relation to various combinations of water activity and temperature

A microarray analysis was performed to study the effect of varying combinations of water activity and temperature on the activation of aflatoxin biosynthesis genes in Aspergillus flavus grown on YES medium. Generally A. flavus showed expression of the aflatoxin biosynthetic genes at all parameter combinations tested. Certain combinations of a_w and temperature, especially combinations which imposed stress on the fungus resulted in a significant reduction of the growth rate. At these conditions induction of the whole aflatoxin biosynthesis gene cluster occurred, however the produced aflatoxin B₁ was low. At all other combinations (25 °C/0.95 and 0.99; 30 °C/0.95 and 0.99; 35 °C/0.95 and 0.99) a reduced basal level of cluster gene expression occurred. At these combinations a high growth rate was obtained as well as high aflatoxin production. When single genes were compared, two groups with different expression profiles in relation to water activity/temperature combinations occurred. These two groups were co-ordinately localized within the aflatoxin gene cluster. The ratio of aflR/aflJ expression was correlated with increased aflatoxin biosynthesis.     

Validation of the performance of a GMO multiplex screening assay based on microarray detection

A new screening method for the detection and identification of GMO, based on the use of multiplex PCR followed by microarray, has been developed and is presented. The technology is based on the identification of quite ubiquitous GMO genetic target elements first amplified by PCR, followed by direct hybridisation of the amplicons on a predefined microarray (DualChip^® GMO, Eppendorf, Germany). The validation was performed within the framework of a European project (Co-Extra, contract no 007158) and in collaboration with 12 laboratories specialised in GMO detection. The present study reports the strategy and the results of an ISO complying validation of the method carried out through an inter-laboratory study. Sets of blind samples were provided consisting of DNA reference materials covering all the elements detectable by specific probes present on the array. The GMO concentrations varied from 1% down to 0.045%. In addition, a mixture of two GMO events (0.1% RRS diluted in 100% TOPAS19/2) was incorporated in the study to test the robustness of the assay in extreme conditions. Data were processed according to ISO 5725 standard. The method was evaluated with predefined performance criteria with respect to the EC CRL method acceptance criteria. The overall method performance met the acceptance criteria; in particular, the results showed that the method is suitable for the detection of the different target elements at 0.1% concentration of GMO with a 95% accuracy rate. This collaborative trial showed that the method can be considered as fit for the purpose of screening with respect to its intra- and inter-laboratory accuracy. The results demonstrated the validity of combining multiplex PCR with array detection as provided by the DualChip^® GMO (Eppendorf, Germany) for the screening of GMO. The results showed that the technology is robust, practical and suitable as a screening tool.     

一种基于聚类和统计分析DNA基因芯片图像处理算法

DNA基因芯片可以同时监控成千上万个基因的表达信息。图像分析是基因芯片试验中一个重要的环节,直接影响到其后续的处理、分析和研究,比如鉴别预测具有不同表达信息的基因功能。基因芯片图像分析包括三个步骤:图像网格化,图像分割以及信息抽取。该文主要研究分割和信息抽取问题。首先基于K-Means聚类技术提出了一种新的分割方法;其次基于统计分析文章建议了一种新的背景和前景分割校正方法用于更准确的信息抽取。新方法的优点是对于基因芯片中spot图像没有任何形状限制。实际图像分析结果与目前最流行的基因芯片图像分析软件GenePix对比研究表明该文算法是精确有效的。     

Microarray analysis of high light intensity stress on hydrogen production metabolism of Rhodobacter capsulatus

Biohydrogen obtained from purple non sulfur bacteria (PNSB) is an environmentally friendly alternative for hydrogen production. PNSB can be employed in large scale outdoor photobioreactors to produce hydrogen by photofermentation with sunlight as the light source. In external environmental conditions, however, bacteria can experience stress due to high light intensities, which can inhibit or slow down hydrogen production. Previous studies with other PNSB showed varying responses to light intensities (above 4000 lux), in some cases improving, and in others adversely affecting hydrogen production.In this study, Rhodobacter capsulatus, a PNSB species that produce hydrogen efficiently from dark fermenter effluents containing acetate, was used to investigate the effects of high light intensity stress on the hydrogen production metabolism at the gene expression level. A microarray analysis was carried out using a custom-design Affymetrix GeneChip TR_RCH2a520699F. R. capsulatus DSM1710 was grown under a cyclic illumination of 2000 and 7000 lux (12 h light/12 h dark) in a hydrogen production medium having 30 mM acetate and 2 mM glutamate, and was exposed to a high light intensity (10,000 lux) for 1 h in the middle of a light period. The results reveal that photosynthetic reaction center genes were down-regulated in order to protect the photosynthetic membrane from damage. On the other hand, the expression of nitrogenase and electron transport system genes were enhanced by high light intensity. These results show that a high light intensity stress drives R. capsulatus to direct gene expression towards hydrogen production, which supports the hypothesis that hydrogen production is a way for the disposal of excess reducing equivalents to maintain the internal redox balance.