Hello
Are there any object models for storing information about SNPs and their annotations i.e. something equivalent to PAGE-OM that was designed for genotype to phenotype experiments for SNPs?
thanks
I know the NCBI model: ftp://ftp.ncbi.nlm.nih.gov/snp/specs/docsum_3.1.xsd
To store such information I would use RDF (but would it scale ?) or a document- oriented database (couchdb, mongodb etc... )
UPDATE: FLEXIBILITY: i would use a document oriented software because for such a database, your schema will change every time (new annotations, new way to structure your data...). For example it will be far more easier/straightforward to store your data as JSON like
{
"name":"rs25",
"mapping":
[
{
build:"build36",
positions:[{"chromosome":"chr1",position:987987},{"chromosome":"chr2",position:9879}]
},
{
build:"build37",
positions:[{"chromosome":"chr1",position:987997},{"chromosome":"chr2",position:9869}]
}
],
comments:[{"autor":"pierre","date":"2011-02-28","text":"Hello world"}]
}
instead of creating... how many SQL tables?
••RDF**: is a way to store your object as a set of triples: subject/predicate/object.
You can start by storing it using JENA or another simple RDF datastore and querying it using SPARQL. However, as I said before, i don't know if it would be a suitable way to store something like dbSNP.
I've never used RDF. Presumably an RDF schema can be mapped to a relational database? why would you consider a document oriented database? i know in gwas there are lots of documents you might want to store but not so much in snp annotation. What documents could there be?
Much of the PAGE-OM model is suitable for the information you require. It has classes to store variants, alleles, their genomic locations and their effects upon transcripts and translations. It also has classes to model the effect on protein function.
You may also want to look at
