UniProt: the Universal Protein

Resource
About the UniProt
Consortium
The UniProt Consortium comprises
the European Bioinformatics
Institute (EBI), the Swiss Institute
of Bioinformatics (SIB) and the
Protein Information Resource
(PIR). EBI hosts a large resource
of bioinformatics databases and
services. SIB is the founding
centre of the Swiss-Prot group
and maintains the ExPASy (Expert
Protein Analysis System) server a
central resource for proteomics
databases and tools. PIR is heir
to the oldest protein sequence
database, Margaret Dayhoffs Atlas
of Protein Sequence and Structure,
and provides bioinformatics tools
for protein sequence analysis and
classification.
The mission of UniProt is to provide
the scientific community with a
comprehensive, high-quality and
freely accessible resource of protein
sequence and functional information.

www.uniprot.org

We can find out much more about the function of proteins by studying
properties such as the biological processes they are involved in, their
post-translational modifications, their interaction with other molecules,
and their location in cells and organisms, than we could ever learn by
studying the DNA that encodes them. As the number of completely
sequenced genomes increases, the research community is refocusing on
collecting information about all the proteins encoded in these genomes.
UniProt allows biologists to access and rationalise this wealth of data.

What is UniProt?
UniProt is produced by the UniProt Consortium, a collaboration between
the European Bioinformatics Institute (EBI), the Swiss Institute of
Bioinformatics (SIB) and the Protein Information Resource (PIR). UniProt
comprises four components:

The UniProt Knowledgebase (UniProtKB)

The UniProt Knowledgebase, the centrepiece of the UniProt Consortiums
activities, is an expertly and richly curated protein database, consisting of
two sections called UniProtKB/Swiss-Prot and UniProtKB/TrEMBL.

PRIDE

HAMAP

Others

Microbial
Protein
protein
identification
families
data
Gene
Ontology
InterPro
IntAct
Functional
Protein
information
Proteinprotein
families
interactions
and domains

Manual curation
Literature-based
annotation
Sequence analysis

IntEnz
Enzymes

UniProt Knowledgebase
Richly annotated protein sequences

UniProt Archive
All the protein sequences in the public domain
InterPro classification
(families, domains
and motifs)

Signal
prediction
Transmembrane
prediction

Other
predictions
Protein classification

Sources of annotation for the UniProt Knowledgebase.

Automated
annotation

UniProtKB/Swiss-Prot contains high-quality manually annotated and

non-redundant protein sequence records. Manual annotation consists
of analysis, comparison and merging of all available sequences for a
given protein, as well as a critical review of associated experimental and
predicted data. UniProt curators extract biological information from the
literature and perform numerous computational analyses. UniProtKB/
Swiss-Prot aims to provide all known relevant information about a
particular protein. It describes, in a single record, the different protein
products derived from a certain gene from a given species, including each
protein derived by alternative splicing, polymorphisms and/or posttranslational modifications. Protein families and groups are regularly
reviewed to keep up with current scientific findings. UniProt curation
priorities and processes are documented at: www.uniprot.org/help/
biocuration
UniProtKB/TrEMBL contains high-quality computationally analysed
records enriched with automatic annotation and classification. Records are
selected for full manual annotation and integration into UniProtKB/SwissProt according to defined annotation priorities.
The default raw sequence data for UniProtKB are:
DDBJ/ENA/GenBank coding sequence (CDS) translations,
the sequences of PDB structures,
sequences from Ensembl and RefSeq,
data derived from amino acid sequences that are directly submitted to
UniProtKB or scanned from the literature.
We exclude some types of data such as DDBJ/ENA/GenBank entries
that encode small fragments, synthetic sequences, most non-germline
immunoglobulins and T-cell receptors, most patent sequences and highly
over-represented data (e.g. viral antigens). However, these excluded data
are stored in the UniProt Archive (UniParc). We regularly review UniParc
data to ensure that no valuable data are missing.

UniProt Reference Clusters (UniRef)

Three UniRef databases UniRef100, UniRef90 and UniRef50 merge
sequences automatically across species. UniRef100 is based on all
UniProtKB records. It also contains selected UniParc records, including
Ensembl protein translations from chicken, cow, dog, fly, Fugu, human,
mouse, rat, Tetraodon, Xenopus and zebrafish. UniRef100 is produced by
clustering all these records by sequence identity. Identical sequences and
sub-fragments are presented as a single UniRef100 entry with accession
numbers of all the merged entries, the protein sequence, links to the
corresponding UniProtKB and archive records. UniRef90 and UniRef50
are built from UniRef100 to provide records with mutual sequence
identity of 90% or more, or 50% or more, respectively, with links to the
corresponding UniProtKB records. All the sequences in each cluster are
ranked to facilitate the selection of a representative sequence.

UniProt Archive (UniParc)

UniParc is designed to capture all publicly available protein sequence data
and contains all the protein sequences from the main publicly available
protein sequence databases. This makes UniParc the most comprehensive
publicly accessible non-redundant protein sequence database.
A protein sequence may exist in several databases and more than once in a
given database, thus creating redundant information. UniParc overcomes
this problem by storing each unique sequence only once, and assigning
it a unique UniParc identifier. UniParc handles all sequences simply as
text strings sequences that are 100% identical over their entire length are
merged regardless of whether they are from the same or different species.

UniParc data sources

Database(s)

Data type

DDBJ, ENA and Genbank

CDS translations

Coding sequences from the three public nucleotide sequence

databases

Ensembl and VEGA

Predicted coding sequences from vertebrate genomes

FlyBase

Coding sequence for species from the Drosophilidae family

H-Invitational Database
(H-Inv)

Human protein sequences

International Protein Index

(IPI)

Protein sequences of higher eukaryotes

Patent Offices in Europe,

US and Japan

Coding sequences associated with patents from the listed Patent

Offices

PIR-PSD

Curated protein sequences

Protein Data Bank (PDB)

Sequences of proteins whose 3D structures are in the PDB

Protein Research
Foundation (PRF)

Protein sequences from literature and

predictions

RefSeq

Coding sequences from the NCBIs set of genomic, transcript and

protein reference sequences

Saccharomyces Genome
database (SGD)

Coding sequences for Saccharomyces cerevisiae

The Arabidopsis
Information Resource
(TAIR)

Coding sequences for Arabidopsis thaliana

TROME

Predicted protein sequences

UniProtKB/Swiss-Prot

Manually curated protein sequences mostly derived from TrEMBL

UniProtKB/TrEMBL

Automatically curated protein sequences derived from coding

sequences in the nucleotide sequence databases

WormBase

Coding sequences for the nematode Caenorhabditis elegans

You can always trace the source database because UniParc cross-references
their accession numbers. UniParc also provides sequence versions, which
are incremented every time the underlying sequence changes. This allows
you to observe sequence changes in all the source databases.
UniParc records are not annotated because annotation is context
dependent: proteins with the same sequence can have different functions
depending on species, tissue, developmental stage or other variables. This
context-dependent information is the scope of UniProtKB.

UniProt Metagenomic and Environmental

Sequences (UniMES)
The availability of metagenomic data has necessitated the creation of
a separate database, UniMES, to store sequences which are recovered
directly from environmental samples. The predicted proteins from
this dataset are combined with automatic classification by InterPro, an
integrated resource for protein families, domains and functional sites, to
enhance the original information with further analysis.

Different databases for different uses

Why have we built four different databases? Each is optimised for a
different use:
The UniProt Knowledgebase, and in particular UniProtKB/Swiss-Prot,
is used to access functional information on proteins. Every UniProtKB
entry contains the amino acid sequence, protein name or description,
taxonomic data and citation information but in addition to this, we add
as much annotation as possible. This includes widely accepted biological
ontologies, classifications and cross-references, as well as clear indications

on the quality of annotation in the form of evidence attribution to

experimental and computational data.
The UniRef databases provide clustered sets of sequences from
UniProtKB and selected UniParc records to provide complete coverage
of sequence space at several resolutions. UniRef90 and UniRef50 yield
a database size reduction of approximately 40% and 65%, respectively,
providing significantly faster sequence searches.

Further reading
The UniProt Consortium. Ongoing
and future developments at the
Universal Protein Resource. Nucleic
Acids Res. 39, D214-D219 (2011)

Support

UniParc is the most comprehensive publicly accessible non-redundant

protein sequence database available, providing links to all underlying
sources and versions of these sequences. You can instantly find out
whether a sequence of interest is already in the public domain and, if not,
identify its closest relatives.

UniProt is funded by the European

Molecular Biology Laboratory
(EMBL), the US National Institutes of
Health, the European Union and the
Swiss Federal Government.

UniMES is a repository specifically for metagenomic and environmental

data.

Contacting UniProt

UniProtKB

help@uniprot.org

EMBL-EBI

www.uniprot.org

Protein Knowledgebase

UniProtKB/Swiss-Prot
Reviewed

Manual annotation

UniRef

UniMES

Sequence clusters
UniRef100
UniRef90
UniRef50

UniProtKB/TrEMBL
Unreviewed

Automatic annotation

Metagenomic
and environmental
sample sequences

EMBL-EBI
EMBL-EBI
Wellcome Trust Genome Campus
Cambridge, CB10 1SD, UK

EMBL-EBI

Phone: +44 (0)1223 494444

UniParc - Sequence archive

Current and obsolete sequences

ENA/GenBank/DDBJ, Ensembl, other sequence resources

Sources and flow of data for UniProts component databases.

Submitting data to the UniProt Knowledgebase

We provide accession numbers for proteins that have been directly
sequenced. We do not provide, in advance, accession numbers for
protein sequences that result from translation of nucleic acid sequences.
These translations are automatically forwarded to us from the DDBJ/
ENA/GenBank nucleotide sequence databases and are processed into
UniProtKB/TrEMBL. All the information you need to submit sequence or
annotation updates is available at www.uniprot.org/help/submissions

Retrieving data from UniProt databases

Browsing. UniProt offers a range of services at www.uniprot.org that
allow you to browse and analyze data. You can perform both simple and
complex text-based queries, run sequence-based searches of the UniProt
databases, perform multiple sequence alignments, retrieve multiple
entries and map identifiers from an external database to UniProtKB or
vice versa.
Downloading. If you need to download entire databases, the UniProtKB,
UniRef and UniMES databases are available at
www.uniprot.org/downloads
CD-ROM. The UniProt Knowledgebase full releases are distributed on
CD-ROM. If you would like to receive them, please send us an e-mail
using the query form at www.ebi.ac.uk/support/.

SIB Swiss Institute of Bioinformatics

CMU - 1 rue Michel Servet
CH-1211 Geneva 4
Switzerland
Phone: +41 22 379 50 50

Protein Information Resource

Georgetown University Medical
Center
3300 Whitehaven Street NW
Suite 1200
Washington, DC 20007
USA
Phone: +1 202 687 1432