Genetic associations between human mitochondrial DNA (mtDNA) heteroplasmy and mitochondrial diseases, aging, and cancer have been elaborated, contributing a lot to the further understanding of mtDNA polymorphic spectrum in anthropology, population, and forensic genetics. In the past decade, heteroplasmy detection using Sanger sequencing and next generation sequencing (NGS) was hampered by the former’s inefficiency and the latter’s inherent bias due to amplification and mapping of short reads, respectively. Nanopore sequencing stands out for its ability to yield long contiguous segments of DNA, providing a new insight into heterogeneity authentication. In addition to MinION from Oxford Nanopore Technologies, an alternative nanopore sequencer QNome (Qitan Technology) has also been applied to various biological research and the forensic applicability of this platform has been proved recently. In this study, we evaluated the performance of four commonly used variant callers in the heterogeneity authentication of the control region of human mtDNA based on simulations of different ratios generated by mixing QNome nanopore sequencing reads of two synthetic sequences. Then, an open-source and python-based nanopore analytics pipeline, CmVCall was developed and incorporated multiple programs including reads filtering, removal of nuclear mitochondrial sequences (NUMTs), alignment, optional ‘Correction’ mode, and heterogeneity identification. CmVCall can achieve high precision, accuracy, and recall of 100%, 99.9%, and 92.3% with a 5% heteroplasmy level in ‘Correction’ mode. Moreover, blood, saliva, and hair shaft samples from monozygotic (MZ) twins were used for heterogeneity evaluation and comparison with the NGS data. Results of MZ twin samples showed that CmVCall could identify more point heteroplasmy sites, revealing significant levels of inter- and intra-individual mtDNA polymorphism. In conclusion, we believe that this analysis pipeline will lay a solid foundation for the development of a comprehensive nanopore analysis pipeline targeting the whole mitochondrial genome.

人类线粒体DNA（mtDNA）异质性与线粒体疾病、衰老和癌症之间的遗传关联已得到阐述，为人类学、人口和法医遗传学领域进一步了解mtDNA多态性谱做出了很大贡献。在过去的十年中，使用桑格测序和下一代测序（NGS）的异质性检测分别受到前者的低效率和后者由于短读长的扩增和定位而存在的固有偏差的阻碍。纳米孔测序因其能够产生长的连续 DNA 片段而脱颖而出，为异质性认证提供了新的见解。除了Oxford Nanopore Technologies的MinION之外，另一种纳米孔测序仪QNome（Qitan Technology）也已应用于各种生物研究，并且该平台的法医适用性最近得到了证明。在本研究中，我们基于对两个合成序列的 QNome 纳米孔测序读数混合产生的不同比率的模拟，评估了四种常用变体识别器在人类 mtDNA 控制区域异质性认证中的性能。然后，开发了一个基于 Python 的开源纳米孔分析管道 CmVCall，并整合了多个程序，包括读数过滤、核线粒体序列 (NUMT) 去除、比对、可选的“校正”模式和异质性识别。CmVCall 在“校正”模式下可实现 100%、99.9% 和 92.3% 的高精度、准确率和召回率，异质性水平为 5%。此外，来自同卵双胞胎（MZ）的血液、唾液和毛干样本被用于异质性评估并与NGS数据进行比较。同卵双胞胎样本的结果表明，CmVCall 可以识别更多的点异质性位点，揭示个体间和个体内 mtDNA 多态性的显着水平。总之，我们相信该分析流程将为开发针对整个线粒体基因组的综合纳米孔分析流程奠定坚实的基础。