Quantitative cross-validation and content analysis of the 450k DNA methylation array from Illumina, Inc.

Quantitative cross-validation and content analysis of the 450k DNA methylation array from Illumina, Inc.
复制标题

DOI:
10.1186/1756-0500-5-210
复制
发表时间:
2012-04-30
期刊:
影响因子:
1.8
通讯作者:
Lehmann U
Lehmann U
中科院分区:
其他
文献类型:
--
作者:
Roessler J;Ammerpohl O;Gutwein J;Hasemeier B;Anwar SL;Kreipe H;Lehmann U

文献摘要

被引文献

相似文献

Illumina公司新发布的450k DNA甲基化阵列提供了以用户友好的标准化格式分析超过480,000个单个CpG位点的可能性。在本研究中,Illumina,Inc.分析每个单独的CpG二核苷酸的甲基化水平和通过焦磷酸测序获得的定量甲基化水平。此外,微RNA基因和印迹基因座在Illumina,Inc.对阵列进行了详细评估。应用450k DNA甲基化芯片分析了4种人乳腺癌细胞系(IPH-926、HCC 1937、MDA-MB-134、PMC42)和18例人乳腺癌标本以及4种正常乳腺上皮细胞的基因组DNA。使用常规定量焦磷酸测序交叉验证来自62个不同基因的692个单独CpG位点的β值。Illumina,Inc.如果在细胞系中分析相同的CpG位点,则显示与定量焦磷酸测序的高度一致性(斯皮尔曼r = 0.88,p <0.0001),这在原发性肿瘤标本中有所降低(斯皮尔曼r = 0.86,p <0.0001)。80.7%的CpG位点显示甲基化水平的绝对差异小于15个百分点。如果靶向相同CpG岛中的不同CpG位点,则一致性较低(在细胞系中r = 0.83,在原发性肿瘤中r = 0.7)。代表微小RNA基因和印迹基因座的CpG位点的数量是非常不均匀的(范围:对于微小RNA为1 - 70个CpG位点,对于印迹基因座为1 - 288个CpG位点)。Illumina,Inc.以方便的形式提供了DNA甲基化畸变的全基因组定量表示。总体而言,与焦磷酸测序数据的一致性非常好。然而,对于单个基因座,应小心将β值直接转化为甲基化水平百分比。
The newly released 450k DNA methylation array from Illumina, Inc. offers the possibility to analyze more than 480,000 individual CpG sites in a user friendly standardized format. In this study the relationship between the β-values provided by the Illumina, Inc. array for each individual CpG dinucleotide and the quantitative methylation levels obtained by pyrosequencing were analyzed. In addition, the representation of microRNA genes and imprinted loci on the Illumina, Inc. array was assessed in detail. Genomic DNA from 4 human breast cancer cell lines (IPH-926, HCC1937, MDA-MB-134, PMC42) and 18 human breast cancer specimens as well as 4 normal mammary epithelial fractions was analyzed on 450k DNA methylation arrays. The β-values for 692 individual CpG sites from 62 different genes were cross-validated using conventional quantitative pyrosequencing. The newly released 450k methylation array from Illumina, Inc. shows a high concordance with quantitative pyrosequencing if identical CpG sites are analyzed in cell lines (Spearman r = 0.88, p ≪ 0.0001), which is somewhat reduced in primary tumor specimens (Spearman r = 0.86, p ≪ 0.0001). 80.7% of the CpG sites show an absolute difference in methylation level of less than 15 percentage points. If different CpG sites in the same CpG islands are targeted the concordance is lower (r = 0.83 in cell lines and r = 0.7 in primary tumors). The number of CpG sites representing microRNA genes and imprinted loci is very heterogeneous (range: 1 – 70 CpG sites for microRNAs and 1 – 288 for imprinted loci). The newly released 450k methylation array from Illumina, Inc. provides a genome-wide quantitative representation of DNA methylation aberrations in a convenient format. Overall, the congruence with pyrosequencing data is very good. However, for individual loci one should be careful to translate the β-values directly into percent methylation levels.