Re: Searching Markush structures in Marpat vs. MMS

From: Robert Buntrock (buntrock2@earthlink.net)
Date: Fri Jan 28 2000 - 18:30:43 EST


Edlyn Simmons wrote:

> Markush DARC is a search system designed to provide topological indexing and
> retrieval of generic structures. Queries are entered by drawing graphical
> structures and defining nodes, bonds, variables, open sites, and attributes.
>
> MMS is the Merged Markush Service, the only Markush DARC database, which was
> created by merging the structure indexing of patents from what were previously
> two structure files relating to the bibliographic records in the Derwent World
> Patents Index and the PharmSearch file of the French Patent Office (INPI).

> Like STN's MARPAT file, MMS can be used to search generic/Markush structures;
> unlike MARPAT, MMS also contains indexing for specific compounds.

There other differences. MARPAT covers documents (patents) which
contain generic structures. MMS covers structures, both generic and
specific (in patents). The specific compounds in the documents covered
by MARPAT are covered in the CA files. Therefore, on STN one searches
CA Registry, MARPAT, and MARPAT Previews (and Beilstein for good
measure).

 
> To find specific structures in the Chemical Abstracts files, you have to search
> the Registry file, which is available on both STN and Questel-Orbit. STN has
> lots of bells and whistles that are unavailable to other search services.
>

Registry policies differ to some extent between CAS and Derwent (and
Beilstein). Bear this in mind when searching them.

Edlyn and other responders are correct. These "simple" questions have
profound answers. I'll take their word for it that a PERI course would
cover all of them (as much as can be done in one course), but training
in the components is essential to forming good searching practices.
Watch the vendor announcements -- you can often get training for free
(although you may have to travel to the course site).

-- Bob Buntrock
Buntrock Associates, Inc.



This archive was generated by hypermail 2b29 : Fri Aug 10 2001 - 15:58:07 EDT