StemNet: An Evolving Service for Knowledge Networking in the Life Sciences

Up until now, crucial life science information resources, whether bibliographic or factual databases, are isolated from each other. Moreover, semantic metadata intended to structure their contents is supplied in a manual form only. In the StemNet project we aim at developing a framework for semantic interoperability for these resources. This will facilitate the extraction of relevant information from textual sources and the generation of semantic metadata in a fully automatic manner. In this way, (from a computational perspective) unstructured life science documents are linked to structured biological fact databases, in particular to the identifiers of genes, proteins, etc. Thus, life scientists will be able to seamlessly access information from a homogeneous platform, despite the fact that the original information was unlinked and scattered over the whole variety of heterogeneous life science information resources and, therefore, almost inaccessible for integrated systematic search by academic, clinical, or industrial users