Liu Xiao, Zhang Wei, Zeng Xiaojing, Zhang Ruifang, Du Yuanping, Hong Xueyu, Cao Hongzhi, Su Zheng, Wang Changxi, Wu Jinghua, Nie Chao, Xu Xun, Kristiansen Karsten
BGI-Shenzhen, Shenzhen, 518083, China.
Department of Biology, University of Copenhagen, Copenhagen, Denmark.
PLoS One. 2016 Mar 28;11(3):e0152464. doi: 10.1371/journal.pone.0152464. eCollection 2016.
High-throughput sequencing has recently been applied to profile the high diversity of antibodyome/B cell receptors (BCRs) and T cell receptors (TCRs) among immune cells. To date, Multiplex PCR (MPCR) and 5'RACE are predominately used to enrich rearranged BCRs and TCRs. Both approaches have advantages and disadvantages; however, a systematic evaluation and direct comparison of them would benefit researchers in the selection of the most suitable method. In this study, we used both pooled control plasmids and spiked-in cells to benchmark the MPCR bias. RNA from three healthy donors was subsequently processed with the two methods to perform a comparative evaluation of the TCR β chain sequences. Both approaches demonstrated high reproducibility (R2 = 0.9958 and 0.9878, respectively). No differences in gene usage were identified for most V/J genes (>60%), and an average of 52.03% of the CDR3 amino acid sequences overlapped. MPCR exhibited a certain degree of bias, in which the usage of several genes deviated from 5'RACE, and some V-J pairings were lost. In contrast, there was a smaller rate of effective data from 5'RACE (11.25% less compared with MPCR). Nevertheless, the methodological variability was smaller compared with the biological variability. Through direct comparison, these findings provide novel insights into the two experimental methods, which will prove to be valuable in immune repertoire research and its interpretation.
高通量测序最近已被应用于分析免疫细胞中抗体组/B细胞受体（BCR）和T细胞受体（TCR）的高度多样性。迄今为止，多重PCR（MPCR）和5'RACE主要用于富集重排的BCR和TCR。这两种方法都有优缺点；然而，对它们进行系统评估和直接比较将有助于研究人员选择最合适的方法。在本研究中，我们使用混合对照质粒和掺入细胞来评估MPCR偏差。随后用这两种方法处理来自三名健康供体的RNA，以对TCRβ链序列进行比较评估。两种方法均显示出高重现性（R2分别为0.9958和0.9878）。大多数V/J基因（>60%）在基因使用上未发现差异，并且平均52.03%的CDR3氨基酸序列重叠。MPCR表现出一定程度的偏差，其中几个基因的使用与5'RACE不同，并且一些V-J配对丢失。相比之下，5'RACE的有效数据率较低（比MPCR少11.25%）。然而，与生物学变异性相比，方法学变异性较小。通过直接比较，这些发现为这两种实验方法提供了新的见解，这将在免疫组库研究及其解释中被证明是有价值的。
