A Method to Evaluate the Quality of Clinical Gene-Panel Sequencing Data for Single-Nucleotide Variant Detection

作者：Lee Chung; Bae Joon S; Ryu Gyu H; Kim Nayoung K D; Park Donghyun; Chung Jongsuk; Kyung Sungkyu; Joung Je Gun; Shin Hyun Tae; Shin Seung Ho; Kim Younglan; Kim Byung S; Lee Hojun; Kim Kyoung Mee; Kim Jung Sun; Park Woong Yang^*; Son Dae Soon^*

来源：Journal of Molecular Diagnostics, 2017, 19(5): 651-658.

DOI：10.1016/j.jmoldx.2017.06.001

Customized gene-panel tests, based on next-generation sequencing, have demonstrated their usefulness in a plethora of clinical settings. As with other clinical diagnostic techniques, gene-panel sequencing for clinical purposes requires precise quality control (QC) measures to ensure its reliability. Only detected variants are currently recorded in clinical reports; however, identifying whether a nondetected variant is a true or false negative is regarded essential in a clinical setting and, thus, a comprehensive QC measure is in demand. Conventional QC metrics, such as mean coverage and uniformity, are considered inadequate for such an evaluation. As such, a more specific measure focused on clinically important variants is herein proposed. In this study, we suggest a new scoring method for assessing the quality of clinical gene-panel sequencing data, specifically for the detection of a set of single-nucleotide variants. The performance of the method was analyzed using 2295 clinical samples (1012 formalin-fixed, paraffin-embedded and 1283 fresh-frozen tissues), and was shown to provide additional information that conventional methods do not show, such as mean depth and uniformity. Customized sequencing protocols, which include QC criteria, have been optimized by each genomic laboratory. The pass rate scoring method proposed in this study provides an appropriate QC response variable for the customized panel, which strengthens the reliability of calls on clinically relevant variants implicated in clinical reports.

• 出版日期2017-9