I am developing specialized database on some desease microarray experiments. The initial data is retreived from NCBI and ArrayExpress databases. The problem is that sample attributes sometimes specify similar or same things but in different manners: as cell and tissue names, desease names, age, ethnicity of donor, ect. Where do I find conventions for naming experiment sample attributes?

  • 1
    $\begingroup$ I doubt there are many. There are some conventions regarding naming cell types corresponding to what surface and internal markers are expressed (or not), but to my knowledge those are in early stages. In papers, authors generally define their own naming conventions, which may or may not match up to others. The problem is there's a lot of argument about how to name things, with each lab or institution having their own preferences. As a simple example, tuberculosis is also known as Koch or Koch's Disease. You'll probably need to pull from a lot of individual databases, and just use synonyms. $\endgroup$ – MattDMo Aug 24 '16 at 16:27
  • $\begingroup$ Also there can be conventions without controlled vocabulary ( ncbi.nlm.nih.gov/geo/info/MIAME.html ) $\endgroup$ – tsttst Aug 25 '16 at 3:08

It turns out that ArrayExpress itself uses various ontologies (dictionaries structured as trees) available at Ontology Lookup Service when it processes users' search queries. Experimental Factor Ontology is the precisely one I was searching for.

Medical Subject Headings (MeSH) dictionary turned out to be handy as well.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.