Sometimes it would be good to have an ontology for bioinformatics itself. It would for instance help structuring the tags (and thus questions) we have here on BioStar and various tool libraries. Is any of you aware of such an ontology?
If not would it be worth to develop something like that? We could for instance use the existing Biostar tags to get a start. Of course for BioStar itself that would only be really useful if a new code base would also be aware of that ontology. Although people that structure tags could keep it in mind when they improve the tags used.
•
written
by
Chris Evelo
Heard about EDAM ontology today. EDAM (EMBRACE Data and Methods) is an ontology of general bioinformatics concepts, including topics and data types, formats, identifiers and operations. http://edamontology.sourceforge.net/
Thanks to @gawbul and @phidias51 for this tweet.
EDAM (EMBRACE Data and Methods) is an ontology of general bioinformatics concepts, including topics and data types, formats, identifiers and operations. EDAM provides a controlled vocabulary for description in semantic terms of things such as:
- Web services e.g. WSDL files
- Applications
- Tool collections and packages
- Work-benches and workflow software
- Databases and ontologies
- XSD data schema and data objects
- Data syntax and file formats
- Web portals and pages
- Resource catalogues
- Documents such as scientific publications
The goal is to provide, at a coarse level at least, a vocabulary for all the general (non-specialised) topics, data types, formats, identifiers and operations that are of common use in bioinformatics.
edit my Michael: added a quote from the page to merge our two answers & delete mine.
•
written
by
Khader Shameer
See the subcategories of wikipedia for Bioinformatics:
http://en.wikipedia.org/wiki/Category:Bioinformatics
(edit: or dbpedia.org: http://dbpedia.org/page/Category:Bioinformatics)
the subcategories of MESH for Computational+Biology:
http://www.nlm.nih.gov/cgi/mesh/2011/MB_cgi?mode=&term=Computational+Biology&field=entry
I also found some entries in the bioportal: http://bioportal.bioontology.org/
But, as far as I can see, there is no complete ontology for bioinformatics. One could start a new one (or try to add some entries to a broader ontology of Sciences)
EDIT: regarding a 'broader' ontology of Sciences:
searching for "bioinformatics geology filetype:rdf astronomy" returns this SKOS/RDF ontology: http://namespace.adfi.usq.edu.au/anzsrc/for/
and http://code.google.com/p/ccoldev/source/browse/ontowiki/src/ontologies/lab2go.rdf
•
written
by
Pierre Lindenbaum
Pierre, about your last two suggestions. What would you think would be better? Create a new one or add to an existing broader one? Ad which one would that be?
Wikipedia categories seem like a good starting point. As a bonus, they would be instantly resolvable to Wikipedia URLs.
I don't know what would be the 'better'. May be the one(s) with a open license, available on github, etc...
and I'm looking for a 'broader' ontology: https://twitter.com/#!/yokofakun/status/78214406722437120 :-)
•
written
by
Pierre Lindenbaum
While it does not capture all of bioinformatics, there is the Chemical Information Ontology, which can be used for various areas of bioinformatics that involve small molecules, at least. It's development is coordinated by Janna from the ChEBI team.
Other things that spring to mind
There is a software ontology, but I haven't used it: http://bioportal.bioontology.org/ontologies/45349
For reference of what already exists, the "The Open Biological and Biomedical Ontologies" (there are quite a lot!), systems biology modelling and simulation.
NextBio seems to be yet another attempt at ontologizing biodata. http://en.wikipedia.org/wiki/Nextbio
Log in to answer this question.
Basically, putting the tags in a tree? :-)
Basically, putting the tags in a tree? :-) ..apologies if this sounds snotty, I actually think that hierarchical tags might be of great value.
Yes... That might be a way to do it if there is no existing ontology (But see e.g. Pierre's answer below). Building the tree in that way we would probably encounter some missing branches. I think you are right that it would be relatively easy (but that goes for the basic concept of many ontologies).
I think "hierarchical tags" cannot be implemented as a tree because a tag can have multiple parents... I used to be a great fun of hierarchical tags, but they quickly get impractical when you get many tags with multiple relationships.
added answer; just leaving a comment so you get a notification because it might be of interest to you
Thanks Michael, I actually for a notification about the answer as well.
Ok, I'll keep that in mind for next time. However, I just saw that Khader answered the same thing on July 5th, so I'll delete my answer again.