BioTAGME: A comprehensive platform for biological knowledge network analysis

<strong>A Knowledge Graph containing logical or physical relationships among biological elements</strong>. <br> This Network was built through BioTAGME, a system that combines TAGME, an entity-annotation framework based on Wikipedia corpus with a network-based inference methodology (i.e., DT-Hybrid).<br> <em>BioTAGME</em> exploits several Biological Ontologies as "ground truth" of significant bio-entities, such as <em>DisGeNET, DrugBank, STRING</em> and many more. <br> We Deployed <em>BioTAGME</em> on <em>PubMed</em>, where the aim was on extracting biological entities and their relations from titles and abstracts.<br> Biological entities are the <em>nodes</em> of our graph, while <em>edges</em> are the relations between them.<br> Edges are of three categories: Literature edges: interactions derived from publications. STRING: protein-protein associations stored in the STRING database. BioTAGME: interactions predicted by our tool. The <strong>network</strong> is released (<strong>BiotagmeNetwork.zip</strong>) in a neo4j compatible format. Such archive contains: <strong>Edges.csv and Nodes.csv</strong> that contain the nodes and edges of our network, rispectively. <strong>Name_Aliases.csv</strong>: contains the synonyms list of each biological entity. <strong>Wiki_Titles.csv</strong>: contains the wikipedia pages title associated with the annotated entities. <strong>wid1_wid2_pmid.csv</strong>: contains the associations between pairs of entities and pubmed id <strong>BioIDs_WikiIDs.csv</strong>: contains the association between the biological entities annotated by BioTAGME, and the wikipedia pages that are associated.