Biomedical entities

Identification of Relationships that Link Biomedical Entities with... 

Recognition of biomedical entities in text (genes, proteins, allelic variants, clinical phenotype descriptions, etc.)... 

The concept-based identification of biomedical entities is now being offered by most commercial vendors of text mining solutions and by an increasing number of academic tools.31 Biomedical concept search is mainly based on publicly available thesauri such as MeSH or GO (see below). 

NLP techniques provide the basis to extract this kind of more detailed information. In the past years approaches have been developed that are mainly focused on the biomedical domain to extract protein-protein interactions60 and gene-disease relations.61 They are based upon the correct recognition of the named entities taking part in the relationship. Dedicated patterns have to be developed to identify all the phrasal constructs that are indicative of relationships between chemical and biomedical entities available in text being of interest for academic researchers and the pharmaceutical industry. 

Chemical structural information is one of the missing pieces in the great effort to bring biomedical research into the realm of twenty-first century information extraction and knowledge discovery paradigms. Proteins, genes, diseases, and chemical compounds constitute the major entities extracted in the biomedical domain. The ability to read structure information and substructure information and their association to other entities could have a major impact on toxicity information in particular and ADMET data in general. 

Smith, D., Sadagopan, N Zientek, M Reddy, A. and Cohen, L. (2007) Analytical approaches to determine cytochrome P450 inhibitory potential of new chemical entities in drug discovery. Journal of Chromatography. B, Analytical Technologies in the Biomedical and Life Sciences, 850, 455 63. 

Various types of relationships between biomedical and chemical entities can be distinguished by the grade of information they provide. We give some examples of the most important and—in our opinion—most informative classes of relationships that occur in biomedical text in the sense of their meaning. They can be divided into... 

BioTeKS thus is composed of almost all functionalities to analyze biomedical literature including patents for mentions in text of chemical and biological as well as medical entities and the extraction of their relationships. The advantage of the implementation of this system as a UIMA-based, service-oriented architecture is that annotators can be added or replaced depending on the needs of the user of the system. In essence, the only service that is missing from the complex BioTeKS is an analysis service for the annotation of chemical structure depictions. 

During the last part of the 1960s, the Retina Foundation went through a major transformation. The trustees of the Foundation decided that the future of scientific research would be best served if the Foundation were divided into two new entities. One of these was to continue as a research institute devoted entirely to research on the eye and medical problems in ophthalmology. The other was to continue as a biomedical research institute devoted to a broad program of research and not limited to any organ or medical specialty. The first institute is known today as The Charles Schepens Eye Institute, and the second as the Boston Biomedical Research Institute (BBRI). As a co-founder of BBRI I became its first Director. 

Entity extraction is the process of tagging things in the text as specific items such as genes, cell lines, people, chemical compounds, and so on. Two specific requirements for entity extraction are unique and critical to biomedical applications of text mining. Gene name disambiguation and chemical compound name tagging require different approaches. 

---