Guo Junfu, Shi Chang, Chen Xi, Wang Ou, Liu Ping, Yang Huanming, Xu Xun, Zhang Wenwei, Zhu Hongmei
BGI-Tianjin, BGI-Shenzhen, Tianjin, China.
BGI-Shenzhen, Shenzhen, China.
Front Genet. 2021 Mar 18;12:636239. doi: 10.3389/fgene.2021.636239. eCollection 2021.
Co-barcoded reads originating from long DNA fragments (mean length >30 kbp) maintain both single base level accuracy and long-range genomic information. We propose a pipeline, stLFRsv, to detect structural variation using co-barcoded reads. stLFRsv identifies abnormal large gaps between co-barcoded reads to detect potential breakpoints and reconstruct complex structural variants (SVs). Haplotype phasing by co-barcoded reads increases the signal to noise ratio, and barcode sharing profiles are used to filter out false positives. We integrate the short read SV caller smoove for smaller variants with stLFRsv. The integrated pipeline was evaluated on the well-characterized genome HG002/NA24385, and 74.5% precision and a 22.4% recall rate were obtained for deletions. stLFRsv revealed some large variants not included in the benchmark set that were verified by long reads or assembly. For the HG001/NA12878 genome, stLFRsv also achieved the best performance for both resource usage and the detection of large variants. Our work indicates that co-barcoded read technology has the potential to improve genome completeness.
源自长DNA片段（平均长度>30 kbp）的共条形码读取既能保持单碱基水平的准确性，又能保留长程基因组信息。我们提出了一种流程stLFRsv，用于使用共条形码读取来检测结构变异。stLFRsv通过识别共条形码读取之间异常的大间隙来检测潜在的断点，并重建复杂的结构变异（SV）。共条形码读取的单倍型定相提高了信噪比，条形码共享图谱用于过滤假阳性。我们将短读长SV调用工具smoove与stLFRsv整合，用于检测较小的变异。在特征明确的基因组HG002/NA24385上对整合后的流程进行了评估，对于缺失变异，获得了74.5%的精确率和22.4%的召回率。stLFRsv揭示了一些未包含在基准集中的大变异，这些变异通过长读长或组装得到了验证。对于HG001/NA12878基因组，stLFRsv在资源使用和大变异检测方面也取得了最佳性能。我们的工作表明，共条形码读取技术有潜力提高基因组的完整性。
