Library for NCBI Entrez (eutils API?)

[David Winter]

Hi all,

I'll guess I'll make a start on this mailing list then?

I was wondering if there is any interest in an ROpenSci library to
talk to the NCBI entrez databases.

I have a "start level" package I've put together in trying to make
some of my "run once" scrips into something more general (the readme
shows an example of the process as it works at present):

https://github.com/dwinter/rentrez

If there is interest in having such a library as part of the ROpenSci
project, I'd be keen for some advice from people that might use these
functions - how specific or generic you'd expect the functions to be
and what you'd expect to get back from them.

Cheers,
David

[Scott Chamberlain]

Hi David, 

Good idea. 

There are four packages I know of that have something to do with NCBI (RNCBIAxis2Libs, RNCBIEUtilsLibs, NCBI2R, RNCBI), but I am not sure if these do what you are looking for (or do it well?).  I just bring these up as we don't want to duplicate efforts in the R community. 

However, that doesn't mean another pkg can't be developed interacting with NCBI data.  

I don't know much about NCBI data.  Thoughts from other folks?

Scott

[David Winter]

Hi Scott et al,

Wow, I've never seen those. Looking into them RNCBI* are all part of
one super-package and NCBIR2R is for one specific job (annotating
SNPs)

I had envisaged something at a "lower level" than either of these,
akin to Biopython's Bio.Entrez which I've used quite a lot:
http://biopython.org/DIST/docs/api/Bio.Entrez-module.html

The idea being you can pass as many arguments as you want to the
relevant function, and it will make the url needed to talk to the
database.
Presuming everything works, for most databases you get an xml file
which can be parsed to get the things people are most likely to want
to use from it (along with the original file for people that want to
go deeper).

Of course, it's all down to the sorts of things ropensci wants to do -
and the level at which you want the packages to operate.

Thoughts?

David