The GlycoQL interface was designed to perform glycan (sub)structure search on the GlyConnect collection of full glycan structures. This tool takes as input a glycan structure represented as an SFNG cartoon that can be drawn or in a text form corresponding to a structure GlycoCT encoding, widely used in glycoinformatics resources. With this graphic or text query, the list of glycans matching structures stored in GlyConnect is output with direct links to database entries.
A fully defined query can match both fully defined or ambiguous structures as illustrated here: 
Conversely, an ambiguous query can match both ambiguous or fully defined structures as illustrated here: 
Technically, the (sub)structure search relies on the GlySTreeM glycan structure triple store that can be queried without expert knowledge of the SPARQL query language. A data mapping algorithm used in GlySTreeM takes GlycoCT strings and creates an RDF representation of the glycan tree. The main distinction between the output of the GlySTreeM pipeline and that of GlycoQL is in the handling of undefined values in the GlycoCT strings. See the corresponding published paper for more details.
A Sankey diagram is an interactive graph, here, connecting the different types of GlyConnect metadata associated with a glycan structure. Two Sankey versions may co-exist for each glycan, depending on experimental evidence stored in the database.
Legend:
| Taxonomy | Protein | Tissue Source | Cell Type | Cell Line | Disease |
|---|