PubGene BioWeb Search and Analysis Engine

PubGene AS is a leading provider of databases with gene and protein information harvested and refined using text mining (statistical and natural language processing) on high-quality data sources. PubGene was the first to extract knowledge from Medline by use of text-mining (Nature Genetics 2001 28(1): 21-28). The PubGene product provides bioinformatics compilation, sorting and mapping to high-end specialist users in the academic and industrial environment internationally. In this project we will use the P ubGene technology, extend it, and apply it on the Worldwide Web (WWW) in order to develop a domain specific search and analysis engine with focus on the biology and medicine domain. We will apply advanced text-mining to extract information from all releva nt biomedical documents on the WWW, such as information about biomedical entities, e.g. genes, proteins, drugs, and diseases. Then we will integrate this information with data in publicly available data repositories of sequences, genomes and other large a ggregated biological information sources, such as GenBank, ENSEMBL, etc. Combined, this will constitute the most comprehensive resource of biomedical information. The BioWeb will be the first domain specific search engine with knowledge layers generated b y analyses of the contents in the data sources. This is also the main and essential distinction between BioWeb and current search-engines, such as Google. The information will be much more accessible and directly useful for a broad range of users interest ed in molecular biology, medicine, diseases and pharmacogenomics. The BioWeb will serve the expanding demand for information about biology, diseases, biological processes, medicine and the effects of drugs. The market grows fast, and there is a critical n eed for bioinformatics solutions that can handle both the large amount of data published or stored in proprietary databases, and public data on WWW.