Browse Prior Art Database

Accessing Medline citations using DiscoveryLink and joining the search results with other relational and non-relational databases Disclosure Number: IPCOM000016189D
Original Publication Date: 2002-Sep-09
Included in the Prior Art Database: 2003-Jun-21
Document File: 3 page(s) / 98K

Publishing Venue



The Problem:

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 45% of the total text.

Page 1 of 3

  Accessing Medline citations using DiscoveryLink and joining the search results with other relational and non-relational databases

The Problem:

MEDLINE® (Medical Literature, Analysis, and Retrieval System Online) is the U.S. National Library of Medicine's (NLM) premier bibliographic database that contains over 11 million references to journal articles in life sciences with a concentration on biomedicine. Scientists routinely use Medline to search for publications. They would either use NCBI's Entrez PubMed interface or other interfaces from a library or local database. However, if they are to join the Medline search results with other databases, they have to do it manually. This is very time consuming, and could be impossible if joining with another very large database. We describe a solution for accessing Medline citations from Documentum and DiscoveryLink.

The Solution

MEDLine is available in XML format to subscribers. The database and its weekly updates can be loaded to the Documentum Docbase. Each citation is loaded as a document in the Docbase, and the meta data (author, publication date, keywords, etc.) are saved as attributes for the document. The user can search the documents by attributes or by words using Documentum's word search capability.

DiscoveryLink is a middleware data integration product from IBM (1). DiscoveryLink allows user to access multiple heterogeneous data sources in a single SQL query. DiscoveryLink permits a user to access and search documents in a Docbase using SQL by means of its Documentum wrapper. It also permits users to join Documentum documents with other relational and non-relational databases, using other wrappers. The solution is illustrated in the following diagram.


Medline Citations on

Other documents onDocumentumDocumentum

  All Relational Databases (such as DB2, Oracle, Sybase, MS SQL


 All non relational data sources (such as flat files, EXCEL spreadsheets, etc.)

Applications (such as BLAST)



[This page contains 5 pictures or other non-text objects]

Page 2 of 3

1. Advantages over using NCBI's PubMed web interface:

NCBI's web site allows full-text and meta data searching, and hypertext linking to other data sources, but not SQL querying across Medline and other relational and non-relational databases. Instead, users must do multiple searches and make the connections between databases manually, using cut and paste. In addition, any queries run against the web interface are public and potentially may be intercepted by competitors. The same is true for the TINet system (2) of GlaxoSmithKline and Gene Logic, Inc. which allows searches of Medline publication data with other data sources in a single query, but also accesses Medline via the NCBI internet web site. The proposed solution works on a local Documentum database; therefore no queries go out over the internet, and the users have greater privacy.

2. Advantages over importing Medline into a relational database...