Matching references with MEDLINE via TCP/IP

Area of Science:

• Biomedical Informatics
• Database Management
• Bioinformatics

Background:

• Bibliographic references are crucial for clinicians and researchers, offering clerical benefits and linking disparate datasets.
• Integrating diverse databases is essential for comprehensive biomedical research.
• The Mouse Genome Informatics (MGI) effort aims to unify data for enhanced research capabilities.

Purpose of the Study:

• To describe a method for matching references from various databases to canonical MEDLINE references.
• To support a database unification project within the Mouse Genome Informatics (MGI) effort.
• To provide a model for similar data integration initiatives in biomedical research.

Main Methods:

• Developed custom software to interface with a commercial retrieval engine accessing MEDLINE on CD-ROM.
• Enabled client programs (UNIX/C) to access unabridged MEDLINE via TCP/IP networks.
• Implemented a reference matching process to link disparate database entries to MEDLINE.

Main Results:

• Successfully matched references from disparate databases to canonical MEDLINE entries.
• Facilitated the database unification project for the Mouse Genome Informatics (MGI) effort.
• Demonstrated the feasibility of the matching process across different environments.

Conclusions:

• The developed reference matching process is effective for unifying disparate biomedical databases.
• This methodology can serve as a model for future data integration projects in research and clinical settings.
• The approach enhances data accessibility and interoperability for biomedical researchers.