Lecture 18 Jan 2001 Per Kraulis

Databases in bioinformatics

2. The different types of databases

One may characterize the available biological databases by several different properties:

• Type of data
  • nucleotide sequences
  • protein sequences
  • proteins sequence patterns or motifs
  • macromolecular 3D structure
  • gene expression data
  • metabolic pathways
  
  
• Data entry and quality control
  • Scientists (teams) deposit data directly
  • Appointed curators add and update data
  • Are erroneous data removed or marked?
  • Type and degree of error checking
  • Consistency, redundancy, conflicts, updates
  
  
• Primary or derived data
  • Primary databases: experimental results directly into database
  • Secondary databases: results of analysis of primary databases
  • Aggregate of many databases
    • Links to other data items
    • Combination of data
    • Consolidation of data
  
  
• Technical design
  • Flat-files
  • Relational database (SQL)
  • Object-oriented database
  • Exchange/publication technologies (FTP, HTML, CORBA, XML,...)
  
  
• Maintainer status
  • Large, public institution funded by government (EMBL, NCBI)
  • Quasi-academic institute (Swiss Institute of Bioinformatics, TIGR)
  • Academic group or scientist
  • Commercial company
  
  
• Availability
  • Publicly available, no restrictions
  • Available, but with copyright
  • Accessible, but not downloadable
  • Academic, but not freely available
  • Proprietary, commercial; possibly free for academics