生物
分子生物学
分数(化学)
遗传学
色谱法
DNA
计算生物学
化学
作者
Ping Song,Sherry X. Chen,Yan Helen Yan,Alessandro Pinto,Lauren Y. Cheng,Peng Dai,Abhijit A. Patel,David Y. Zhang
标识
DOI:10.1038/s41551-021-00713-0
摘要
DNA sequence variants with allele fractions below 1% are difficult to detect and quantify by sequencing owing to intrinsic errors in sequencing-by-synthesis methods. Although molecular-identifier barcodes can detect mutations with a variant-allele frequency (VAF) as low as 0.1% using next-generation sequencing (NGS), sequencing depths of over 25,000× are required, thus hampering the detection of mutations at high sensitivity in patient samples and in most samples used in research. Here we show that low-frequency DNA variants can be detected via low-depth multiplexed NGS after their amplification, by a median of 300-fold, using polymerase chain reaction and rationally designed ‘blocker’ oligonucleotides that bind to the variants. Using an 80-plex NGS panel and a sequencing depth of 250×, we detected single nucleotide polymorphisms with a VAF of 0.019% and contamination in human cell lines at a VAF as low as 0.07%. With a 16-plex NGS panel covering 145 mutations across 9 genes involved in melanoma, we detected low-VAF mutations (0.2–5%) in 7 out of the 19 samples of freshly frozen tumour biopsies, suggesting that tumour heterogeneity could be notably higher than previously recognized. Low-frequency DNA variants can be detected via low-depth multiplexed sequencing after their amplification using polymerase chain reaction and rationally designed ‘blocker’ oligonucleotides.
科研通智能强力驱动
Strongly Powered by AbleSci AI