All Title Author
Keywords Abstract

Communication and re-use of chemical information in bioscience

DOI: 10.1186/1471-2105-6-180

Full-Text   Cite this paper   Add to My Lib


In a previous article [1] we have argued the value of extracting the chemical information in bioscientific research, transforming it to XML and redisseminating it openly. The present article expands on the technical and cultural infrastructure required to support this. The technical aspects have been solved to proof-of-concept stage and we are starting to embark on experiments in the social domain. In this we thank BMC for inviting us to submit this and we present a model here which we believe could be attractive for bioscience publishers and their community.We concentrate on the current publication of chemistry in bioscience. This includes:1. mention of chemical compounds.2. details of synthesis (in vivo and in vitro) of compounds.3. proof of structure (spectra and analytical data).4. Methods and reagents in bioscience bio-protocols5. properties of compounds.6. reactions and their properties, both in enzymes and enzyme-free systems.This type of chemistry is very well understood and has a simple ontology which has not changed over decades[2]. Unlike much bioscience, where ontological tools are an essential part of reconciling the domain-dependent approaches, much chemistry has an implicitly agreed abstract description. The problems are primarily reconciling syntax and semantics. This is because chemists use abbreviated and lazy methods of communicating data, relying on trained readers to add information from the context. We have reviewed current problems of machine-understanding of chemistry[3] in a typical chemistry journal, many of which are perpetuated by the graphical orientation of conventional publishing houses. Here we take the view that a committed publishing house can create a cost-effective and human-tolerable system for authoring semantically correct chemistry in (bio)scientific documents.We know from experience that Utopian visions do not sell themselves. The enormous and accepted value of the sequence and structures databases arose not from the demands


comments powered by Disqus