Baudhuin Linnea M, Lagerstedt Susan A, Klee Eric W, Fadra Numrah, Oglesbee Devin, Ferber Matthew J
Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota.
Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota.
J Mol Diagn. 2015 Jul;17(4):456-61. doi: 10.1016/j.jmoldx.2015.03.004. Epub 2015 May 8.
Current clinical laboratory practice guidelines for next-generation sequencing (NGS) do not provide definitive guidance on confirming NGS variants. Sanger confirmation of NGS results can be inefficient, redundant, and expensive. We evaluated the accuracy of NGS-detected single-nucleotide variants (SNVs) and insertion/deletion variants (indels) and the necessity of NGS variant confirmation using four NGS target-capture gene panels covering 117 genes, 568 Kbp, and 77 patient DNA samples. Unique NGS-detected variants (1080 SNVs and 124 indels) underwent Sanger confirmation and/or were compared to data from the 1000 Genomes Project (1000G). Recurrent variants in unrelated samples resulted in 919 comparisons between NGS and Sanger, with 100% concordance. In a second comparison, 762 unique NGS results (736 SNVs, 26 indels) from seven 1000G samples were found to have 97.1% concordance with 1000G phase 1 data. Sanger sequencing and 1000G phase 3 data confirmed the accuracy of the NGS results for all 1000G phase 1 discrepancies. In all samples, the depth of coverage exceeded 100× in >99.7% of bases in the target regions. In conclusion, confirmatory analysis by Sanger sequencing of SNVs detected via capture-based NGS testing that meets appropriate quality thresholds is unnecessarily redundant. In contrast, Sanger sequencing for indels may be required for defining the correct genomic location, and Sanger may be used for quality-assurance purposes.
当前关于下一代测序（NGS）的临床实验室操作指南并未就确认NGS变异提供明确指导。对NGS结果进行桑格测序确认可能效率低下、多余且昂贵。我们使用覆盖117个基因、568千碱基对的4个NGS靶向捕获基因panel以及77份患者DNA样本，评估了NGS检测到的单核苷酸变异（SNV）和插入/缺失变异（indel）的准确性以及NGS变异确认的必要性。通过NGS检测到的独特变异（1080个SNV和124个indel）进行了桑格测序确认和/或与千人基因组计划（1000G）的数据进行了比较。无关样本中的重复变异导致NGS与桑格测序之间进行了919次比较，一致性为100%。在第二次比较中，发现来自7个1000G样本的762个独特NGS结果（736个SNV、26个indel）与1000G一期数据的一致性为97.1%。桑格测序和1000G三期数据证实了所有1000G一期差异中NGS结果的准确性。在所有样本中，目标区域中>99.7%的碱基覆盖深度超过100倍。总之，对通过基于捕获的NGS检测且达到适当质量阈值所检测到的SNV进行桑格测序确认分析是不必要的多余操作。相比之下，可能需要对indel进行桑格测序以确定正确的基因组位置，并且桑格测序可用于质量保证目的。
