After completing this section you will:
Know the most important datatypes in bioinformatics: sequencese (protein and DNA) and structures.
Know the 3 major data collections in bioinformatics: SwissProt, EMBL and PDB.
Know the 5 data elements all good databases minimally contain.

Figure 10. SwissProt is a well-curated database of protein sequences.

Figure 11. PDB is more a databank than a database. The PDB was started by people at the Brookhave national lab. Nowadays the PDB is kept at Rutgers; with mirror systems at the EBI and in Japan. It holds macromolecular structures solved mainly by X-ray or NMR. These are mainly protein structures but also DNA, RNA, and all kinds of complexes.

Figure 12. EMBL is not only the name of a research institute, but it is also the name of the international depository for nucleic acid sequences, the EMBL database.

The OMIM databank is the brainchild of Victor A McKusick. This databank holds description of phenotypes for a whole series of disease causing SNPs / mutations in the human genome. OMIM stands for On-line Mendelian Inheritance in Man.

The Prosite database holds information about sequence patterns that indicate potential post-translational modification sites, cleavage sites, active sites, etc.

Figure 13. UniProt is a much larger protein sequence database that one normally should use if SwissProt doesn't hold what you are looking for. During this course SwissProt alone will always be enough to answer the questions.

Figure 14. Unique identifier, or accession code
The unique identifier, or accession code is needed to track data through the years. When the PDB started all entries were kept on punched cards, and much of the annotation was in a filing cabinet. One had to call the PDB to ask for a photocopy of the additional information. Thanks to the fact that the PDB uses the four-letter-code as unique identifier (in the PDB the file name and the unique identifier are the same), we can now still retrieve these very old data, and all information around it.

Figure 15. Name of depositor
The name of the depositor is needed to give proper credit, but also to know who to contact in case of problems, or questions.

Figure 16. Literature references
Literature references help track down the nitty-gritty data that normally is only available in the materials and methods section of the article.

Figure 17. Deposition date
The deposition date is useful for patent claims, but also provides some help to the users. E.g., the first refinement program became available in the early eighties. So, coordinates deposited before 1980 are surely unrefined, and that is important knowledge.

Figure 18. The real data
It seems trivial that database entries contain data. But it isn't. There is a conflict between industrial secrets and patents the one side and the dissemination of science on the other. In the past it was permissible to publish articles about a sequence or structure without depositing the data in a database. It was also possible to deposit an entry that contained no data. This practice is no longer allowed.