Department of Anatomy, University of Otago, Dunedin, New Zealand.
Eisai Inc., Cambridge, MA, USA.
Methods Mol Biol. 2021;2272:29-44. doi: 10.1007/978-1-0716-1294-1_3.
Whole-genome bisulfite sequencing (WGBS) is a popular method for characterizing cytosine methylation because it is fully quantitative and has base-pair resolution. While WGBS is prohibitively expensive for experiments involving many samples, low-coverage WGBS can accurately determine global methylation and erasure at similar cost to high-performance liquid chromatography (HPLC) or enzyme-linked immunosorbent assays (ELISA). Moreover, low-coverage WGBS has the capacity to distinguish between methylation in different cytosine contexts (e.g., CG, CHH, and CHG), can tolerate low-input material (<100 cells), and can detect the presence of overrepresented DNA originating from mitochondria or amplified ribosomal DNA. In addition to describing a WGBS library construction and quantitation approach, here we detail computational methods to predict the accuracy of low-coverage WGBS using empirical bootstrap samplers and theoretical estimators similar to those used in election polling. Using examples, we further demonstrate how non-independent sampling of cytosines can alter the precision of error calculation and provide methods to improve this.
全基因组亚硫酸氢盐测序（WGBS）是一种用于描述胞嘧啶甲基化的常用方法，因为它是完全定量的，并且具有碱基分辨率。虽然 WGBS 对于涉及许多样本的实验来说过于昂贵，但低覆盖度的 WGBS 可以以与高效液相色谱（HPLC）或酶联免疫吸附测定（ELISA）相似的成本准确确定全局甲基化和擦除。此外，低覆盖度的 WGBS 能够区分不同胞嘧啶环境中的甲基化（例如 CG、CHH 和 CHG），可以耐受低输入材料（<100 个细胞），并且可以检测来自线粒体或扩增核糖体 DNA 的过表达 DNA 的存在。除了描述 WGBS 文库构建和定量方法外，我们还详细介绍了使用类似于选举民意调查中使用的经验引导采样器和理论估计器来预测低覆盖度 WGBS 准确性的计算方法。通过示例，我们进一步展示了非独立采样的胞嘧啶如何改变误差计算的精度，并提供了改进方法。
