E-BioSci: Semantic networks of biological information

Abstract

Searching for specific pieces of information scattered across multiple articles in the full text literature or other databases, demands computer-based tools to aid the human expert. Such tools must cope with the complexities of natural language(s), be capable of unambiguous recognition of the idiosyncratic gene and other symbols devised by scientists and deal with a variety of other linguistic complications. As part of the response to this challenge, EMBO took the lead to create the E-BioSci network, a scientific information service that will interlink genomic and other factual or image data with the life sciences research literature. The service will be developed together with partners from different research institutions across Europe and will receive financial support from the European Commission for the coming three years. The prototype currently under construction facilitates intelligent searching of full text, document-neighbouring based on semantic content, cross-repository searching and cross-language searching. It permits recognition of gene and protein symbols in full text with an indication of database lookup services available. Future versions will provide graphical representation of semantic relationships between documents and entities retrieved in searches.