[Obo-discuss] Problems with MESH obo file

[Richard C. <rc...@eb...>]

Hello all

I'm finding data inconsistencies between the mesh.obo file as submitted =
to
the OBO cvs repository and the raw ASCII file I downloaded from the NLM
(http://www.nlm.nih.gov/mesh/filelist.html). I'm trying to produce an =
OBO
file that is indexed by the MeSH unique identifiers (UI line) rather =
than
the mesh tree number (MN  line). I've come across a rather troubling =
fact:
there are 3467 terms in the mesh.obo file that I can't match back to the =
NLM
reference version.

Here's an example of one term.

This is from the reference MESH file:
*NEWRECORD
RECTYPE =3D D
MH =3D Ctenophora
MN =3D B01.500.325
UI =3D D037302

This is from the obo MESH file:
[Term]
id: MESH:B.01.325
name: ctenophora
is_a: MESH:B.01 ! Invertebrate (also, from the reference file, =
invertebrates
are B.01.500)

Please advise,
Rc

--
Richard Cote
Software Engineer - PRIDE Project Team (Sequence Database Group)
            =20
European Bioinformatics Institute            =20
Wellcome Trust Genome Campus            RC...@eb...     =20
Hinxton                                =20
Cambridge CB10 1SD                      Phone: (+44) 1223 492610 =20
United Kingdom                          Fax  : (+44) 1223 494468=20