Reply to Zwaenepoel et al.: Meeting the Challenges of Detecting Polyploidy Events from Transcriptomic Data
Reply to Zwaenepoel et al.: Meeting the Challenges of Detecting Polyploidy Events from Transcriptomic Data
复制标题
回复 Zwaenepoel 等人：应对从转录组数据中检测多倍体事件的挑战
DOI：
10.1016/j.molp.2018.12.020
复制
发表时间：
2019
期刊：
影响因子：
27.5
通讯作者：
Qi Ji
中科院分区：
文献类型：
--
作者：
Wang Haifeng;Guo Chunce;Ma Hong;Qi Ji
转录组学数据已被广泛用于推断全基因组重复（WGD，或古多倍性）（Jiao et al.，2011; Barker等人，2016年; Huang等人，2016），通过使用大量具有相关基因树和KS分布的公认旁系同源物，具有确定此类多倍性事件日期的能力。然而，由于转录组数据集的限制，该分析可能具有挑战性，包括由于测序深度不足或组装序列不准确而导致的重复基因检测中的假阴性，以及由于组装错误、不同等位基因或可变剪接而从相同基因的转录物中调用旁系同源物的假阳性。需要对数据进行仔细考虑和彻底分析以得出可靠的结论，特别是对于必须考虑小规模基因重复（SSD）可能影响的谱系特异性WGD的追踪证据。最近，我们报道了来自100多个物种的基因组和转录组数据集的分析，为被子植物中广泛的WGD提供了支持（Ren et al.，2018年）。我们怀着极大的兴趣阅读了Zwaenepoel等人的题为“寻找全基因组复制的证据：重新评估”的信件。（2019）（本期分子植物），其中他们对Ren等人的研究提出了一些担忧。（2018），更一般地提供了一些关于从基因组和/或转录组数据推断WGD的关键讨论。我们欢迎可能有助于进一步阐明这一主题的重要讨论。为了解决Zwaenepoel等人的评论和担忧，（2019），在这里，我们提供了我们关于使用从转录组数据推断的信息分析WGD的想法，包括保留的基因重复（GD）数量，旁系同源物的KS分布，以及基于基因树与物种树比较的调和方法。我们研究中使用的基因组学方法，虽然作为基于同线性的算法，不能为谱系特异性WGD提供明确的信号，已广泛用于多倍性事件的大规模调查（Jiao et al.，2011; Barker等人，2016年）。因此，从转录组数据推断的许多WGD，包括在我们的研究中新发现的WGD，通过随后公布的基因组序列的分析进行了验证，表明该方法的有效性和效率。然而，从组装良好的基因组序列的同线性分析中得到的染色体中的保守基因顺序将可能为所发现的WGD候选者提供更强有力的支持。
Transcriptomic data have been widely employed to infer whole genome duplications (WGDs, or paleopolyploidy)(Jiao et al., 2011; Barker et al., 2016; Huang et al., 2016), through the use of large numbers of recognized paralogs with associated gene tree and KS distributions, with the ability to date such polyploidy events. However, the analysis can be challenging due to limitations of transcriptomic datasets, including false negatives in detection of duplicated genes due to insufficient depth in sequencing or inaccuracy in assembled sequences, and false positives in calling paralogs from transcripts of the same gene from assembly mistakes, different alleles, or alternative splicing. Careful consideration and thorough analyses of the data are needed to reach reliable conclusions, especially for tracing evidence of lineage-specific WGDs where possible effects of small-scale gene duplications (SSDs) must be considered. Recently, we reported analyses of genomic and transcriptomic datasets from over 100 species, providing support for widespread WGDs in angiosperms (Ren et al., 2018). We read with great interest of the Correspondence entitled ‘‘Finding evidence for whole genome duplications: a reappraisal’’by Zwaenepoel et al.(2019)(this issue of Molecular Plant), in which they raised a few concerns on our study by Ren et al.(2018) and more generally provided some critical discussions on the inferring of WGDs from genomic and/or transcriptomic data. We welcome critical discussions that may help elucidate this topic further. To address the comments and concerns by Zwaenepoel et al.(2019), here we provide our thoughts regarding the analyses of WGDs using information inferred from transcriptomic data including retained gene duplicate (GD) numbers, the KS distribution of paralogs, and the reconciliation method based on comparison of gene trees with species trees.The phylogenomic method employed in our study, though not providing definitive signals for lineage-specific WGDs as a synteny-based algorithm, has been widely adopted in large-scale surveys of polyploidy events (Jiao et al., 2011; Barker et al., 2016). Consequently, many WGDs inferred from transcriptomic data, including ones newly identified in our study, are verified by analyses of subsequently published genome sequences, suggesting the effectiveness and efficiency of the method. Nevertheless, conserved gene orders in chromosomes from synteny analysis of well-assembled genome sequences will likely provide even stronger support for the discovered WGD candidates.