SAGA

SAGA (Substructure Index-based Approximate Graph Alignment) is a tool for querying a biological graph database to retrieve matches between subgraphs of molecular interactions that scientists select and biological networks. SAGA implements an efficient approximate subgraph matching algorithm that can be used for a variety of biological graph matching problems. One application is: pathway matching, In which SAGA is used to compare pathways in KEGG and Reactome. .

You can also use SAGA to find matches in literature databases that have been parsed into semantic graphs. In this use of SAGA, portions of PubMed have been parsed into graphs that have nodes representing gene names. A link is drawn between two genes if they are discussed in the same sentence (indicating there is potential association between the two genes). SAGA let you match graphs to these two different databases even though the content is distinct and the databases organize pathways in different ways . This is achieved by SAGA’s flexible approximate subgraph matching model which computes graph similarity, and allows for node gaps, node mismatches, and graph structural differences. Note comparing pathways from different databases can be a precursor to pathway data integration.

Intended Audiences

Developers, analysts

Tool Category

Information Retrieval , Integration/ Mapping , NLP , Pathways/Networks