Database Reference Fetching
Discovering Database References for Sequences
Database references are associated with a sequence are displayed as a
list in the tooltip shown when mousing over its sequence ID. Jalview
uses references for the retrieval of PDB structures and DAS
features, and for retrieving sequence cross-references such as the
protein products of a DNA sequence.
Initiating reference retrieval
The
application provides three ways to access the retrieval function.
Either:
Jalview discovers references for a sequence by generating a set of ID queries from the ID string of each sequence in the alignment. It then tries to query a subset of all the databases it can access in order to match the alignment sequence to any records retrieved from the database. If a match is found, then the sequence is annotated with that database's reference, and any cross-references that its records contain.
The Sequence Identification Process
The
method of accession id discovery is derived from the method which
earlier Jalview versions used for Uniprot sequence feature retrieval,
and was originally restricted to the identification of valid Uniprot
accessions.
Essentially, Jalview will try to retrieve records
from a subset of the databases accessible by the sequence fetcher using each sequence's ID string (or each string in
the ID separated by the '∣' symbol).
If a record (or set of records) is retrieved by any query derived from the ID string of a sequence, then the sequence is aligned to the ones retrieved to determine the correct start and end residue positions (which are displayed when the 'Show Full Sequence ID' option). This is important for the correct display of the location of any features associated with that database.
If the alignment reveals differences between the sequence in the alignment and the one in the record, then Jalview will assume that the aligned sequence is not the one in the retrieved record.
In some cases, the ID used to retrieve records may be out of date and a dialog box will be opened indicating that a 100% match between the sequence and the record was identified, but the sequence name is different. In this case, the can be manually changed (by right clicking on the sequence ID and selecting Sequence→Edit Name).