| At the meeting of April 20, I had an action item AlanR to review http://esw.w3.org/topic/HCLS/ParkinsonUseCase and see if BioRDF task for translating data to RDF provide answers to questions being asked I did a quick review of this and came up with the following connections: 1. In the systems physiology view, the question What proteins (receptors) recognize those chemicals (neurotransmitters) and what are the responses that they elicit in the neuron? Concepts of interest: Receptors, Proteins (subtype of Receptors), Recognizes, Response_Elicited Antibodies are a tools for probing the state of proteins. To the extent that the responses to neurotransmitters might change the state or _expression_ of a protein, antibodies can be used to measure this via, e.g. one of the in situ methods. You will be able to link from protein/gene to antibodies against that protein, and, probably to specific antibodies based on the specific modification of a protein you are expecting, and specifically to those antibodies annotated as appropriate for use e.g. FISH. 2. Cellular and Molecular Biologist View Staining by the use of antibodies is a fundamental method for the visualization of proteins in cells. Most experimentalists who want to work with model systems will use them. So being able to quickly link from protein of interest to the antibodies against that protein will speed up the process of designing experiments. In addition, although the initial antibody database is curated by people interested in Alzheimer's research, many of the same proteins are of interest in studying other neurological diseases. In the same way that the antibodies in this database will be annotated as "being of interest to someone studying alzheimers", I expect that other such resources will be encoded as "being of interest to someone studying parkinsons", once can imagine a situation in which, eventually, researchers will be able to query for reagents that have been validated for use in their specific area of interest. Such an association could eventually mined from research articles, where, using NLP one could extract the disease of interest, and from the methods sections, any information about antibodies used. 3. Clinical Researcher View Can a certain diagnostic test (e.g., a blood test for a biomarker or an imaging study) provide an approach to diagnosing Parkinson’s disease that is superior to or can complement existing diagnostic approaches? Concepts of interest: Diagnostic test, blood test (subtype of diagnostic test), imaging study (subtype of diagnostic test), Test Associated With Disease Antibody tests, particularly ELISA (an antibody assay) are commonly used as medical diagnostics. One of the properties annotated for antibodies what methods it is useful for and ELISA (and other assay types with potential clinical relevance) are in the glossary of such methods: see http://www.alzforum.org/res/com/ant/glossary.asp. Or go to http://www.alzforum.org/res/com/ant/default.asp and search for ELISA as a keyword. 4. Neurologist View What non-human models exist for Parkinson's Disease? (Last week I heard a talk by Leo Pallanck on fruit fly research on Parkinson's disease!) A small piece of this might be the species reactivity annotations in the antibody database. So if a protein is determined as being of interest, and an antibody is known against that protein, then there might be a suggestion that any species against which the antibody is known to be reactive might be candidates for non-human model systems. That's all for now. Regards, Alan |
