Jointly Integrating VCF-Based Variants and OWL-Based Biomedical Ontologies in MongoDB
Jointly Integrating VCF-Based Variants and OWL-Based Biomedical Ontologies in MongoDB
复制标题
在 MongoDB 中联合集成基于 VCF 的变体和 2 个基于 OWL 的生物医学本体
DOI：
10.1109/tcbb.2019.2951137
复制
发表时间：
2020-09-01
影响因子：
4.5
通讯作者：
Zhang，Lei
中科院分区：
文献类型：
--
作者：
Liu，Jian;Qu，Zhi;Zhang，Lei
下一代测序（NGS）技术的发展导致了大量的VCF（变体调用格式）文件，其已成为1000个基因组计划开发的标准格式。与此同时，随着生物医学本体在生物医学领域的广泛应用，越来越多的应用已经接受Web本体语言（OWL）作为生物医学本体描述规范的主导数据格式，导致基于OWL的生物医学本体规模迅速增长。在本文中，我们试图探索一种有效的方法来管理基于VCF的遗传变异和基于OWL的生物本体使用MongoDB数据库。考虑到当前的许多应用（例如短遗传变异数据库dbSNP等），正在过渡到使用JSON（JavaScript Object Notation）的新设计，以支持未来的海量数据扩展和交换。我们首先提出了一系列的规则，从VCF和OWL文件到JSON文件的映射，然后提出了基于规则的算法转换基于VCF的遗传变异和基于OWL的生物本体到JSON对象。在此基础上，介绍了在MongoDB中集成JSON映射文件的有效方法。最后，我们用一组实验来补充这项工作，以显示我们所提出的方法的性能。拟议办法的源代码可在https://github.com/lyotvincent/AJIA上免费查阅。
The development of the next-generation sequencing (NGS) technologies has led to massive amounts of VCF (Variant Call Format) files, which have been the standard formats developed with 1000 Genomes Project. At the same time, with the widespread use of biomedical ontologies in the biomedical community, more and more applications have accepted the Web Ontology Language (OWL) as the dominant data format for the specifications of biomedical ontology descriptions, leading to the rapid growth of OWL-based biomedical ontology scale. In this paper, we seek to explore an effective method for the management of VCF-based genetic variants and OWL-based biological ontologies using the MongoDB database. Considering many current applications (such as the short genetic variations database dbSNP, etc.) are transitioning to the new design by using JSON (JavaScript Object Notation) to support future massive data expansion and interchanges. We firstly propose a series of rules for the mapping from VCF and OWL files to JSON files, and then present rule-based algorithms for transforming VCF-based genetic variants and OWL-based biological ontologies into JSON objects. On this basis, we introduce effective approaches of integrating the mapped JSON files in MongoDB. Finally, we complement this work with a set of experiments to show the performance of our proposed approaches. The source code of the proposed approaches could be freely available at https://github.com/lyotvincent/AJIA.