Abstract
Current querying architectures for biological databases assume some degree of computational competence and significant structural awareness of the underlying linked open data network of the databases on the part of biologists. The non-compliance of FAIR standards by most of these biological databases also makes it difficult to find and conveniently query without significant effort, and requires investment in the form of expensive custom workflow construction efforts. In this paper, we introduce a more flexible and intelligent natural language query interface as an alternative platform called BioStar. In BioStar, we leverage a knowledgebase, called the schema graph, to map natural language queries to relevant databases and appropriate biological concepts to facilitate responding to queries accurately. We present the structure of the BioStar knowledgebase, and an algorithm, called Needle, to map queries to underlying resources.