Once a collection of visual queries (or patterns) is present (see How to define visual queries and How to use analysis contexts?), each pattern can be used in 4 distinct modes. We illustrate this through a naif example. Advanced usages are shwon in How to use inference.
http://www.sgtp.net/AndreaSplendiani/ext/rdfscape/VisualQueryWorking1.jpg
1 Add patterns. Add all the matches of the entire pattern in teh knowledge base to the current graph. An example of a result is shown here:
http://www.sgtp.net/AndreaSplendiani/ext/rdfscape/VisualQueryWorking2.jpg
2 Add variables. As the above, but add only variables, and not all the patterns. The corresponding result is illustrated below:
http://www.sgtp.net/AndreaSplendiani/ext/rdfscape/VisualQueryWorking3.jpg
3 Search Variables. Search (and select) in the current graph the variables for which there is a specific match in teh ontology. In the above examples, it searches in the current graph all elements that in the ontology are known to of rdf:type bp-exp-2:protein.
When the current graph is the result of 1, this obviously leads to the fallowing selection:
http://www.sgtp.net/AndreaSplendiani/ext/rdfscape/VisualQueryWorking5.jpg
Actual variables matches are also shown in a pattern panel:
http://www.sgtp.net/AndreaSplendiani/ext/rdfscape/VisualQueryWorking6.jpg
4 Search patterns. As for 3, but search for whole patterns instead of varibles only. A pattern searched on the graph generated through itself (1) obviously selects all the graph:
http://www.sgtp.net/AndreaSplendiani/ext/rdfscape/VisualQueryWorking4.jpg