Huber Wolfgang, Toedling Joern, Steinmetz Lars M
European Molecular Biology Laboratory, European Bioinformatics Institute, Cambridge CB10 1SD, UK.
Bioinformatics. 2006 Aug 15;22(16):1963-70. doi: 10.1093/bioinformatics/btl289. Epub 2006 Jun 20.
High-density DNA tiling microarrays are a powerful tool for the characterization of complete transcriptomes. The two major analytical challenges are the segmentation of the hybridization signal along genomic coordinates to accurately determine transcript boundaries and the adjustment of the sequence-dependent response of the oligonucleotide probes to achieve quantitative comparability of the signal between different probes.
We describe a dynamic programming algorithm for finding a globally optimal fit of a piecewise constant expression profile along genomic coordinates. We developed a probe-specific background correction and scaling method that employs empirical probe response parameters determined from reference hybridizations with no need for paired mismatch probes. This combined analysis approach allows the accurate determination of dynamical changes in transcription architectures from hybridization data and will help to study the biological significance of complex transcriptional phenomena in eukaryotic genomes.
R package tilingArray at http://www.bioconductor.org.
高密度DNA平铺微阵列是表征完整转录组的强大工具。两个主要的分析挑战是沿着基因组坐标对杂交信号进行分割，以准确确定转录本边界，以及调整寡核苷酸探针的序列依赖性响应，以实现不同探针之间信号的定量可比性。
我们描述了一种动态规划算法，用于沿着基因组坐标找到分段常数表达谱的全局最优拟合。我们开发了一种探针特异性背景校正和缩放方法，该方法采用从参考杂交中确定的经验性探针响应参数，无需配对错配探针。这种联合分析方法能够从杂交数据中准确确定转录结构的动态变化，并将有助于研究真核基因组中复杂转录现象的生物学意义。
R包tilingArray，网址为http://www.bioconductor.org 。
