Transcript Rolf Apweiler

Genome Annotation:
A Protein-centric Perspective
Protein data contributing to genome
annotation


Gene structure prediction
Gene function prediction
UniProt



Collaboration between EBI, SIB and PIR
Funded mainly by NIH
Based on the original work on PIR, Swiss-Prot
and TrEMBL
UniProt Goals




High level of annotation
Minimal redundancy
High level of integration with other databases
Complete and up-to-date
UniProt Non-Redundancy Concepts



UniProt Archive (UniParc):
 All sequences that are 100% identical over their
entire length are merged into a single entry,
regardless of species. UniParc represents each
protein sequence once and only once, assigning it a
unique Identifier. UniParc cross-references the
accession numbers of the source databases.
UniProt Knowledgebase:
 Aims to describe in a single record all protein
products derived from a certain gene (or genes if the
translation from different genes in a genome leads to
indistinguishable proteins) from a certain species.
UniProt Nref:
 Merges sequences automatically across different
UniParc 2.2. July 2004


3,913,916 unique sequences from 10,422,131
source records
Source databases are DDBJ/EMBL/GenBank,
UniProt/Swiss-Prot, UniProt/TrEMBL, PIR-PSD,
Ensembl, International Protein Index (IPI), PDB,
RefSeq, FlyBase, WormBase, H-Inv, TROME,
European Patent Office, United States Patent
and Trademark Office and Japan Patent Office
UniProt Protein DAS Reference
Server


Aristotle – Data Source for the Reference Server
Creating a Plugin for Thomas Down's DAZZLE
Servlet
DAS Infrastructure - Overview
DAS Client – Connects to Reference
Server and zero or more
Annotation Servers.
Merge duplicate features?
● Resolve version differences ?
●
EBI
Reference and
Annotation from the
EBI
Protein
DAS
Reference
Server
InterPro
UniProt
Download
every 2
weeks
Aristotle
Protein
DAS
Annotation
Server
Protein
DAS
Annotation
Server
Creating a Plugin for DAZZLE /
Aristotle


Involved linking the Aristotle Java API to the
BioJava & DAZZLE Java API's
Issues with enabling a useful entry_point
command

Creation of an 'artificial' hierarchy of entry points,
based upon sequence length
Possible Approach to Implementing
Local Annotation Servers





Use GFF Format as a simple and accessible primary
data source
Problem with this – not suitable for very large
numbers of records, so...
Load this into a relational database (sticking with
SQL-92 to ensure as cross-platform as possible)
Use a standard plugin that will allow the 'GFF' data
to be read from the relational database.
From the point of view of the data curators, this
process should be transparent, i.e. they should be
able to work with GFF files and not need to worry
about the database structure
UniProt Protein DAS Server



External Page: http://www.ebi.ac.uk/uniprot-das/
DAS server package downloadpage:
http://www.ebi.ac.uk/uniprot-das/gffDasApp.html
The UniProt DAS Server itself:
http://www.ebi.ac.uk/das-srv/uniprot/das