Amselem Joëlle, Cornut Guillaume, Choisne Nathalie, Alaux Michael, Alfama-Depauw Françoise, Jamilloux Véronique, Maumus Florian, Letellier Thomas, Luyten Isabelle, Pommier Cyril, Adam-Blondon Anne-Françoise, Quesneville Hadi
URGI, INRA, Université Paris-Saclay, 78026 Versailles, France.
Mob DNA. 2019 Jan 29;10:6. doi: 10.1186/s13100-019-0150-y. eCollection 2019.
Thanks to their ability to move around and replicate within genomes, transposable elements (TEs) are perhaps the most important contributors to genome plasticity and evolution. Their detection and annotation are considered essential in any genome sequencing project. The number of fully sequenced genomes is rapidly increasing with improvements in high-throughput sequencing technologies. A fully automated de novo annotation process for TEs is therefore required to cope with the deluge of sequence data.However, all automated procedures are error-prone, and an automated procedure for TE identification and classification would be no exception. It is therefore crucial to provide not only the TE reference sequences, but also evidence justifying their classification, at the scale of the whole genome. A few TE databases already exist, but none provides evidence to justify TE classification. Moreover, biological information about the sequences remains globally poor.
We present here the RepetDB database developed in the framework of GnpIS, a genetic and genomic information system. RepetDB is designed to store and retrieve detected, classified and annotated TEs in a standardized manner. RepetDB is an implementation with extensions of InterMine, an open-source data warehouse framework used here to store, search, browse, analyze and compare all the data recorded for each TE reference sequence. InterMine can display diverse information for each sequence and allows simple to very complex queries. Finally, TE data are displayed via a worldwide data discovery portal. RepetDB is accessible at urgi.versailles.inra.fr/repetdb.
RepetDB is designed to be a TE knowledge base populated with full de novo TE annotations of complete (or near-complete) genome sequences. Indeed, the description and classification of TEs facilitates the exploration of specific TE families, superfamilies or orders across a large range of species. It also makes possible cross-species searches and comparisons of TE family content between genomes.
由于转座元件（TEs）能够在基因组内移动和复制，它们可能是基因组可塑性和进化的最重要贡献者。在任何基因组测序项目中，对它们的检测和注释都被认为是必不可少的。随着高通量测序技术的改进，全基因组测序的数量正在迅速增加。因此，需要一个完全自动化的TEs从头注释流程来应对海量的序列数据。然而，所有自动化程序都容易出错，TEs识别和分类的自动化程序也不例外。因此，不仅要提供TEs参考序列，还要在全基因组范围内提供证明其分类合理性的证据，这一点至关重要。已经存在一些TEs数据库，但没有一个能提供证明TEs分类合理性的证据。此外，关于这些序列的生物学信息总体上仍然很少。
我们在此展示了在GnpIS（一个遗传和基因组信息系统）框架下开发的RepetDB数据库。RepetDB旨在以标准化方式存储和检索检测到的、分类的和注释的TEs。RepetDB是InterMine的一个扩展实现，InterMine是一个开源数据仓库框架，在此用于存储、搜索、浏览、分析和比较为每个TEs参考序列记录的所有数据。InterMine可以为每个序列显示各种信息，并允许进行简单到非常复杂的查询。最后，TEs数据通过一个全球数据发现门户进行展示。可通过urgi.versailles.inra.fr/repetdb访问RepetDB。
RepetDB旨在成为一个TEs知识库，其中填充了完整（或接近完整）基因组序列的全从头TEs注释。实际上，TEs的描述和分类有助于在广泛的物种范围内探索特定的TE家族、超家族或目。它还使跨物种搜索和比较基因组之间的TE家族内容成为可能。
