Van Loo Peter, Nilsen Gro, Nordgard Silje H, Vollan Hans Kristian Moen, Børresen-Dale Anne-Lise, Kristensen Vessela N, Lingjærde Ole Christian
Cancer Genome Project, Wellcome Trust Sanger Institute, Hinxton, Cambridge, UK.
Methods Mol Biol. 2012;802:57-72. doi: 10.1007/978-1-61779-400-1_4.
Single nucleotide polymorphism (SNP) arrays are powerful tools to delineate genomic aberrations in cancer genomes. However, the analysis of these SNP array data of cancer samples is complicated by three phenomena: (a) aneuploidy: due to massive aberrations, the total DNA content of a cancer cell can differ significantly from its normal two copies; (b) nonaberrant cell admixture: samples from solid tumors do not exclusively contain aberrant tumor cells, but always contain some portion of nonaberrant cells; (c) intratumor heterogeneity: different cells in the tumor sample may have different aberrations. We describe here how these phenomena impact the SNP array profile, and how these can be accounted for in the analysis. In an extended practical example, we apply our recently developed and further improved ASCAT (allele-specific copy number analysis of tumors) suite of tools to analyze SNP array data using data from a series of breast carcinomas as an example. We first describe the structure of the data, how it can be plotted and interpreted, and how it can be segmented. The core ASCAT algorithm next determines the fraction of nonaberrant cells and the tumor ploidy (the average number of DNA copies), and calculates an ASCAT profile. We describe how these ASCAT profiles visualize both copy number aberrations as well as copy-number-neutral events. Finally, we touch upon regions showing intratumor heterogeneity, and how they can be detected in ASCAT profiles. All source code and data described here can be found at our ASCAT Web site ( http://www.ifi.uio.no/forskning/grupper/bioinf/Projects/ASCAT/).
单核苷酸多态性（SNP）阵列是描绘癌症基因组中基因组畸变的强大工具。然而，癌症样本的这些SNP阵列数据分析因三种现象而变得复杂：（a）非整倍体：由于大量畸变，癌细胞的总DNA含量可能与其正常的两份拷贝有显著差异；（b）非畸变细胞混合：实体瘤样本并非仅包含畸变的肿瘤细胞，而是总是包含一定比例的非畸变细胞；（c）肿瘤内异质性：肿瘤样本中的不同细胞可能有不同的畸变。我们在此描述这些现象如何影响SNP阵列图谱，以及在分析中如何考虑这些因素。在一个扩展的实际示例中，我们应用我们最近开发并进一步改进的ASCAT（肿瘤等位基因特异性拷贝数分析）工具套件，以一系列乳腺癌的数据为例来分析SNP阵列数据。我们首先描述数据的结构、如何绘制和解释数据以及如何进行分段。接下来，核心ASCAT算法确定非畸变细胞的比例和肿瘤倍性（DNA拷贝的平均数），并计算ASCAT图谱。我们描述这些ASCAT图谱如何可视化拷贝数畸变以及拷贝数中性事件。最后，我们提及显示肿瘤内异质性的区域，以及如何在ASCAT图谱中检测到它们。此处描述的所有源代码和数据均可在我们的ASCAT网站（http://www.ifi.uio.no/forskning/grupper/bioinf/Projects/ASCAT/）上找到。
