Databases and Protein Structures

Abstract

Bioinformatics databases contain biological data from scientific experiments most importantly DNA and protein sequences and protein structures. Databases of published literature, computational analysis of primary data, and metadata are also important. Primary and secondary databases refer to the type and source of stored data. Primary databases, such as GenBank and ENA, are also called archives or repositories, and they take information directly from the individual researcher, and data are owned by the submitter with privileges to change data. The nucleotide databases DDBJ, EMBL, and GenBank are automatically translated to the protein level if the DNA sequences are coding. The secondary databases (e.g., Swiss-Prot and PDB) are curated, and they perform a quality control and sorting of information before the information is made accessible to the public. These databases have better chances of reducing redundancy. They can also bypass the submitters of entries in the primary databases which are no longer updated. Domains are compact units of proteins that may behave independently and be associated with certain functions. Motifs are conserved regions of proteins which may be part of domains. The prediction of domains can be performed based on single motifs, multiple motifs, and full domains or using procedures mixing different methods. The function of a protein can be predicted by a rather low identity to other known proteins over rather short length of comparison and rather low similarity to protein structures. Proteomics is dealing with the prediction of proteins based on the measurements of mass-to-charge ratios (m/z). The prediction of proteins is then done with programs like Mascot where the m/z coordinates from an analysis are held up against the reviewed part of UniProt.