Towards FAIRer Biological Knowledge Networks Using a Hybrid Linked Data and Graph Database Approach.
Towards FAIRer Biological Knowledge Networks Using a Hybrid Linked Data and Graph Database Approach.
复制标题
DOI：
10.1515/jib-2018-0023
复制
发表时间：
2018-08-07
影响因子：
1.9
通讯作者：
Hassani-Pak K
中科院分区：
文献类型：
--
作者：
Brandizi M;Singh A;Rawlings C;Hassani-Pak K
无论是人类还是人工智能，新科学发现的速度和准确性都取决于基础数据的质量以及有效连接、搜索和共享数据的技术。近年来，我们看到了图形数据库和半正式数据模型（如知识图）的兴起，以促进科学发现的软件方法。这些方法扩展了基于形式化模型的工作，例如语义网。在本文中，我们介绍了我们的发展，以连接，搜索和共享数据的基因组规模的知识网络（GSKN）。我们已经开发了一个简单的应用本体的基础上OWL/RDF与映射到标准模式。我们正在使用本体来支持数据访问服务，如可解析的URI，SPARQL端点，JSON-LD Web API和基于Neo4j的知识图。我们演示了如何建议的本体和图形数据库大大提高搜索和访问互操作性和可重用的生物知识（即公平的数据原则）。
The speed and accuracy of new scientific discoveries – be it by humans or artificial intelligence – depends on the quality of the underlying data and on the technology to connect, search and share the data efficiently. In recent years, we have seen the rise of graph databases and semi-formal data models such as knowledge graphs to facilitate software approaches to scientific discovery. These approaches extend work based on formalised models, such as the Semantic Web. In this paper, we present our developments to connect, search and share data about genome-scale knowledge networks (GSKN). We have developed a simple application ontology based on OWL/RDF with mappings to standard schemas. We are employing the ontology to power data access services like resolvable URIs, SPARQL endpoints, JSON-LD web APIs and Neo4j-based knowledge graphs. We demonstrate how the proposed ontology and graph databases considerably improve search and access to interoperable and reusable biological knowledge (i.e. the FAIRness data principles).