UniCarbKB : a glycobioinformatics infrastructure for data discovery using semantics

Glycobioinformatics databases and tools are now co-operatively adopting semantic technologies for managing data content. Recently, an international team of developers have published standard/controlled ontology for representing glycan structure, annotation information and analytical data collections using RDF (Resource Description Framework). The availability of GlycoRDF and access to RDFized databases is opening new and exciting avenues for connecting and interrogating large volumes of publicly accessible data. UniCarbKB has implemented this standard prototype and generated RDF versions for all curated glycan structures and glycoprotein (site-specific and global) information. Here, we will give examples of the capability of UniCarbKB-RDF for exploring and correlating glycomics structural and experimental data. Furthermore, the utility of semantic technologies to connect glycan-related knowledge bases with other omics resources (UniProtKB and NeXtProt) to enhance data discovery and inference of protein and glycan biological function will be highlighted. In addition, we will introduce innovative semantic concepts for describing glycan structures (fully defined, linkage ambiguity and residue substitutions) and strategies for efficiently data mining GlycoRDF affiliated databases. To achieve this ambitious goal a coordinated biocuration program focuses on extracting glycan structure and glycoprotein (site-specific and global) information from the literature. This is supported by the development of workflows designed to improve data capture whilst ensuring consistency and accuracy. An overview of the strategy will be presented including efforts to enhance community-driven curation partnered with UniProtKB to increase content and value of data provided by UniCarbKB. Especially, our efforts to align data capture with the GlycoRDF, MIRAGE and international glycan repository initiatives.1 page(s