Protein Sequence Databases: Swiss-Prot, TrEMBL, UniProt, UniProtKB, UniParc, UniRef and UniMES
Protein sequence databases are essential resources in bioinformatics, molecular biology, biotechnology, genomics, proteomics and computational biology. They help researchers find protein sequences, protein names, functions, domains, active sites, biological pathways, post-translational modifications, disease associations and many other types of information.
Among the most important resources, Swiss-Prot, TrEMBL, UniProt, UniProtKB, UniParc and UniRef are frequently discussed in biotechnology and bioinformatics courses. UniMES is also found in older textbooks and study material, but it is important to know that UniMES has now been retired.
This article explains these resources in simple language, with comparison tables, diagrams, important links, exam points, FAQs and revision questions. It is especially useful for CSIR NET, GATE Biotechnology, DBT BET, ICAR, ICMR, CUET PG, JRF and biotechnology entrance examinations.
📚 Index / Table of Contents
- Introduction to Protein Sequence Databases
- What is a Protein Sequence Database?
- Why Protein Databases Are Important
- What is UniProt?
- Swiss-Prot
- TrEMBL
- UniProtKB
- UniParc
- UniRef
- UniMES
- Relationship Between Swiss-Prot, TrEMBL and UniProtKB
- How Protein Data Flows Through UniProt
- Swiss-Prot vs TrEMBL vs UniProtKB vs UniParc vs UniRef
- What Information is Found in a UniProt Protein Record?
- UniProt Accession Number
- Reviewed vs Unreviewed Protein Entries
- UniRef100, UniRef90 and UniRef50
- Applications in Biotechnology
- How to Search UniProt
- Important Official Websites and Resources
- Important Exam Points
- MCQs for Quick Revision
- Frequently Asked Questions
- Conclusion
1. Introduction to Protein Sequence Databases
Proteins are one of the most important biological molecules in living organisms. They act as enzymes, receptors, transporters, structural components, transcription factors, antibodies and signalling molecules. The function of a protein is strongly connected with its amino acid sequence.
When scientists determine or predict the amino acid sequence of a protein, that information can be stored in a protein sequence database. Over time, millions of protein sequences have become available through genome sequencing, transcriptome projects, proteomics experiments and other biological studies.
A protein sequence is normally represented using the one-letter amino acid code. For example, a short sequence might look like:
MKTLLLTLVVVTIVCLDLGY...
However, a protein database does not simply store this sequence. A useful database record may also contain information about the protein's name, gene, organism, function, catalytic activity, domains, subcellular location, interactions, disease relevance, sequence variants, post-translational modifications and links to other databases.
This is why protein sequence databases are much more than simple collections of amino acid sequences.
2. What is a Protein Sequence Database?
A protein sequence database is an organized collection of protein sequences and associated biological information.
The basic sequence is written using the 20 standard amino acids, represented by one-letter symbols such as A, R, N, D, C, E, Q, G, H, I, L, K, M, F, P, S, T, W, Y and V.
Depending on the database, a protein record may include:
- Protein amino acid sequence
- Protein name
- Gene name
- Organism
- Protein function
- Enzyme activity
- Subcellular location
- Protein domains
- Active sites
- Binding sites
- Post-translational modifications
- Sequence variants
- Protein-protein interactions
- Literature references
- Cross-references to other databases
- Evidence information
These annotations make a protein sequence much more useful than a sequence alone.
Simple example
Imagine that a researcher has identified a protein sequence from a bacterial genome. The sequence itself may tell us its amino acid composition, but a database search can reveal that a similar protein is an enzyme involved in a particular metabolic pathway.
The researcher can then investigate conserved amino acids, domains, catalytic residues and related proteins.
3. Why Are Protein Sequence Databases Important?
Protein databases are important because modern biological research produces far more sequence data than scientists can manually interpret one sequence at a time.
3.1 Protein function prediction
If an unknown protein sequence resembles a well-characterized protein, sequence similarity can provide clues about its possible function.
3.2 Protein family identification
Related proteins often share conserved sequence regions. Protein databases help researchers identify members of the same protein family.
3.3 Domain analysis
Protein sequences can be examined for conserved domains and functional regions. This is useful for predicting the biological role of uncharacterized proteins.
3.4 Evolutionary studies
Protein sequences are frequently used to study evolutionary relationships between organisms. Conserved protein sequences can provide evidence of common ancestry.
3.5 Drug discovery
Researchers can use protein information to study drug targets, enzymes, receptors and other medically important proteins.
3.6 Structural biology
Protein sequence information can be connected with experimentally determined structures in resources such as the Protein Data Bank.
3.7 Proteomics
Mass-spectrometry-based proteomics relies heavily on reference protein sequences for peptide identification.
3.8 Biotechnology
Protein sequence databases are useful for identifying industrial enzymes, metabolic proteins, transporters and proteins with potential biotechnology applications.
4. What is UniProt?
UniProt stands for Universal Protein Resource. It is one of the most widely used resources for protein sequence and functional information.
UniProt is not simply one single database. It is better understood as a collection of related protein resources, with UniProtKB, UniParc and UniRef being major components.
The UniProt Knowledgebase, or UniProtKB, is the central resource for functional protein information. UniProtKB contains two major sections:
- UniProtKB/Swiss-Prot – reviewed and manually annotated records.
- UniProtKB/TrEMBL – unreviewed records with computational annotation.
UniProt's official documentation describes UniProtKB as the central hub for functional protein information and explains that Swiss-Prot is the reviewed section while TrEMBL is the unreviewed section.
Figure: Simplified organization of major UniProt protein resources.
5. Swiss-Prot
Swiss-Prot is the reviewed section of UniProtKB and is famous for its high-quality manual annotation.
In modern terminology, the correct name is UniProtKB/Swiss-Prot.
Swiss-Prot records are reviewed by expert curators. Information from scientific literature, experimental evidence and computational analysis is evaluated and incorporated into the record.
Important features of Swiss-Prot
- Reviewed protein entries
- Manual annotation
- High-quality functional information
- Protein sequence information
- Literature references
- Functional descriptions
- Protein domains and sites
- Cross-references to other resources
- Evidence information
Swiss-Prot is particularly useful when you need reliable functional information about a protein.
For example, if you are studying a known human enzyme and want information about its catalytic function, active site, subcellular location or known variants, a reviewed Swiss-Prot record is often a very useful starting point.
Swiss-Prot and redundancy
Swiss-Prot aims to provide a high-quality, non-redundant view of protein information. UniProt documentation explains that UniProtKB/Swiss-Prot brings together experimental results, computed features and scientific conclusions in manually reviewed records.
6. TrEMBL
TrEMBL is the unreviewed section of UniProtKB. The name is traditionally associated with translated EMBL nucleotide sequence data, although the current UniProt system integrates protein sequences from several sources.
In current terminology, the resource is called UniProtKB/TrEMBL.
Unlike Swiss-Prot, TrEMBL entries are primarily annotated computationally and have not yet undergone full manual review.
Why is TrEMBL needed?
Modern genome sequencing generates an enormous number of predicted proteins. It would be impossible for experts to manually review every new protein immediately.
TrEMBL provides a way to make computationally annotated protein information available while allowing selected records to receive additional manual curation later.
Main characteristics
- Unreviewed protein records
- Automatic/computational annotation
- Large-scale protein sequence coverage
- Useful for newly predicted proteins
- Records can later receive manual annotation
Swiss-Prot = reviewed
TrEMBL = unreviewed
UniProt's documentation states that UniProtKB/TrEMBL contains computationally annotated, unreviewed records, while UniProtKB/Swiss-Prot contains manually reviewed entries.
7. UniProtKB – UniProt Knowledgebase
UniProtKB stands for UniProt Knowledgebase. It is the central hub of UniProt for protein sequence and functional information.
UniProtKB combines the two major sections discussed above:
| Section | Status | Annotation | Main purpose |
|---|---|---|---|
| UniProtKB/Swiss-Prot | Reviewed | Manual curation | High-quality functional information |
| UniProtKB/TrEMBL | Unreviewed | Automatic/computational annotation | Large-scale protein coverage |
A UniProtKB entry generally contains the amino acid sequence, protein name or description, taxonomic information, citations and many additional annotations.
What can you find in UniProtKB?
- Protein function
- Protein sequence
- Gene name
- Organism
- Subcellular location
- Protein family
- Domains
- Active sites
- Binding sites
- Post-translational modifications
- Sequence variants
- Diseases
- Pathways
- Protein interactions
- Publications
- Cross-references
UniProt accession number
Every UniProtKB entry has an accession identifier. UniProt recommends using the primary accession number when linking to or citing a UniProtKB entry because it provides an unambiguous reference to the entry.
8. UniParc – UniProt Archive
UniParc stands for UniProt Archive. It is designed to provide a comprehensive archive of protein sequences from public sequence resources.
UniParc is different from UniProtKB because its primary focus is sequence archiving rather than detailed functional annotation.
According to UniProt, each unique sequence is stored once in UniParc and receives a stable identifier called a UPI. The UPI is not reassigned to a different sequence. UniParc also keeps cross-references to source database records and tracks sequence history.
Main features of UniParc
- Protein sequence archive
- Non-redundant sequence collection
- Stable UPI identifiers
- Tracks sequence history
- Contains cross-references to source databases
- Useful for tracing sequence changes
- Includes sequences from multiple public resources
What does UniParc focus on?
A simple way to understand UniParc is:
UniProtKB asks: What do we know about this protein?
UniParc asks: What protein sequences have existed in public sequence resources, and where have they appeared?
This distinction is extremely useful when comparing UniProt databases.
9. UniRef – UniProt Reference Clusters
UniRef stands for UniProt Reference Clusters. It groups related protein sequences into clusters so that redundant sequence information can be reduced during large-scale searches.
UniRef is particularly useful for sequence similarity searching because a query can be searched against representative clusters instead of returning huge numbers of highly similar sequences.
UniProt describes UniRef as clustered sets of sequences from UniProtKB, including isoforms, and selected UniParc records.
The three major UniRef levels
- UniRef100
- UniRef90
- UniRef50
These levels represent different degrees of sequence clustering.
UniRef100 combines identical sequences and qualifying subfragments. UniRef90 is constructed by clustering UniRef100 sequences at approximately 90% sequence identity with an overlap requirement, while UniRef50 is built at approximately 50% sequence identity.
Figure: Simplified concept of UniRef sequence clustering.
10. UniMES – UniProt Metagenomic and Environmental Sequences
UniMES stands for UniProt Metagenomic and Environmental Sequences.
It was developed as a resource for protein sequences originating from metagenomic and environmental sequencing projects. This was particularly relevant when large environmental sequencing projects were producing huge numbers of predicted protein sequences that did not fit easily into traditional organism-based protein resources.
However, UniProt has retired UniMES. The official UniProt documentation now recommends the EMBL-EBI Metagenomics portal, MGnify for metagenomics research.
Why is UniMES still mentioned in textbooks?
Older bioinformatics textbooks and study material may list UniMES together with Swiss-Prot, TrEMBL, UniProtKB, UniParc and UniRef. This is because UniMES was historically an important UniProt resource for metagenomic and environmental protein sequences.
Therefore, students should know both facts:
- UniMES was designed for metagenomic and environmental protein sequence data.
- UniMES has now been retired and is not a current active UniProt database.
For current metagenomics work, researchers should look toward MGnify, the EMBL-EBI metagenomics resource.
11. Relationship Between Swiss-Prot, TrEMBL and UniProtKB
One of the most common sources of confusion for students is the relationship between these names.
The easiest way to understand them is:
UniProtKB = UniProt Knowledgebase
Inside UniProtKB:
- UniProtKB/Swiss-Prot = reviewed and manually annotated.
- UniProtKB/TrEMBL = unreviewed and computationally annotated.
Figure: UniProtKB consists of reviewed Swiss-Prot and unreviewed TrEMBL sections.
12. How Protein Data Flows Through UniProt
Protein sequence information can enter UniProt from several sources. A major source is the translation of coding sequences submitted to nucleotide sequence resources such as GenBank, ENA and DDBJ.
UniProt states that more than 95% of protein sequences provided by UniProtKB are derived from translations of coding sequences submitted to these public nucleotide sequence resources. These sequences are automatically incorporated into the unreviewed TrEMBL section and selected records can subsequently receive manual annotation in Swiss-Prot.
Figure: Simplified view of how nucleotide-derived protein sequences can enter UniProt resources.
13. Comparison of Swiss-Prot, TrEMBL, UniProtKB, UniParc, UniRef and UniMES
| Resource | Main role | Annotation | Important point |
|---|---|---|---|
| Swiss-Prot | Reviewed protein knowledge | Manual | High-quality reviewed entries |
| TrEMBL | Unreviewed protein knowledge | Automatic/computational | Large-scale protein coverage |
| UniProtKB | Protein knowledgebase | Manual + automatic | Contains Swiss-Prot and TrEMBL |
| UniParc | Protein sequence archive | Not primarily functional annotation | Unique sequences and sequence history |
| UniRef | Protein sequence clustering | Cluster-based | UniRef100, UniRef90 and UniRef50 |
| UniMES | Metagenomic/environmental sequences | Historical resource | Retired/deprecated |
Swiss-Prot = reviewed knowledge
TrEMBL = unreviewed knowledge
UniProtKB = Swiss-Prot + TrEMBL
UniParc = sequence archive
UniRef = sequence clusters
UniMES = retired metagenomic/environmental resource
14. What Information is Found in a UniProt Protein Record?
A UniProtKB protein record can contain a large amount of biological information. Learning the major sections is useful for both research and examinations.
14.1 Protein name
The record provides the recommended protein name and, when available, alternative names.
14.2 Gene
Gene names and gene-related information can be included in the record.
14.3 Organism
The taxonomic source of the protein is given so that the protein can be connected with a particular organism or strain.
14.4 Function
This section explains what the protein does when reliable information is available.
14.5 Catalytic activity
For enzymes, the record may describe the biochemical reaction catalysed by the protein.
14.6 Subcellular location
Information about where a protein is located inside a cell can be provided when evidence is available.
14.7 Sequence features
Important sequence features may include active sites, binding sites, domains, transmembrane regions, signal peptides and post-translational modifications.
14.8 Variants
Some records contain information about naturally occurring or disease-associated sequence variants.
14.9 Cross-references
UniProtKB entries can link to many other biological databases. This allows researchers to move between sequence, structure, pathway, genome and literature resources.
15. UniProt Accession Number
An accession number is one of the most important identifiers associated with a UniProtKB entry.
For example, a UniProtKB accession may look like:
P99999
The exact accession depends on the protein record.
Accession numbers are useful because they provide a stable way to identify a UniProtKB entry. UniProt recommends using accession numbers rather than relying only on entry names when linking to specific records.
Why accession numbers matter
- Easy protein record retrieval
- Useful in research publications
- Useful for database cross-referencing
- Useful for bioinformatics pipelines
- Useful for ID mapping
- Helps avoid ambiguity between protein names
16. Reviewed vs Unreviewed Protein Entries
This is one of the most frequently tested concepts in protein databases.
| Feature | Reviewed – Swiss-Prot | Unreviewed – TrEMBL |
|---|---|---|
| Review status | Reviewed | Unreviewed |
| Annotation | Manual curation | Automatic/computational |
| Expert review | Yes | Not fully manually reviewed |
| Functional confidence | Generally higher when evidence supports it | Varies and may be computationally inferred |
| Typical use | Reliable functional interpretation | Large-scale sequence coverage |
It is important to understand that unreviewed does not mean useless. TrEMBL is extremely valuable because it provides access to large numbers of protein sequences and computational annotations. The distinction is about the annotation review status.
17. UniRef100, UniRef90 and UniRef50
UniRef reduces redundancy by grouping related protein sequences. The three main cluster levels are UniRef100, UniRef90 and UniRef50.
UniRef100
UniRef100 combines identical sequences and qualifying sequence fragments into a single cluster. It provides broad coverage of the protein sequence space while reducing exact duplication.
UniRef90
UniRef90 is built from UniRef100 seed sequences and groups sequences at approximately 90% sequence identity with an overlap criterion.
UniRef50
UniRef50 is built from UniRef90 seed sequences and provides a broader grouping at approximately 50% sequence identity with an overlap criterion.
UniProt notes that UniRef90 and UniRef50 reduce database size substantially, making sequence similarity searches faster and results less redundant.
UniRef100 → identical sequences
UniRef90 → about 90% identity clustering
UniRef50 → about 50% identity clustering
18. Applications of Protein Sequence Databases
18.1 Protein function prediction
Unknown proteins can be compared with characterized proteins to obtain clues about their possible function.
18.2 Enzyme discovery
Protein sequence databases are useful for identifying enzymes with potentially useful catalytic properties.
18.3 Drug target identification
Researchers can investigate proteins involved in diseases and biological pathways to identify potential therapeutic targets.
18.4 Vaccine research
Protein sequences can be examined to identify conserved or antigenically relevant regions.
18.5 Evolutionary analysis
Protein sequence alignment can be used to study homologous proteins and evolutionary relationships.
18.6 Proteomics
Reference protein sequences are important for identifying peptides generated during mass spectrometry experiments.
18.7 Protein engineering
Researchers can compare homologous proteins to identify amino acid positions that may influence stability, activity or specificity.
18.8 Industrial biotechnology
Protein databases can help researchers identify enzymes involved in food biotechnology, biofuel production, bioremediation and industrial biocatalysis.
18.9 Structural bioinformatics
Protein sequence records can be connected to experimentally determined structures, predicted structures and structural databases.
18.10 Comparative genomics
Predicted proteins from multiple organisms can be compared to investigate conserved biological processes and organism-specific adaptations.
19. How to Search UniProt
Searching UniProt is relatively simple, but getting useful results requires a good search strategy.
Step 1: Start with the protein or gene name
For example, you can search for a protein name such as hemoglobin, catalase or DNA polymerase.
Step 2: Add the organism
If you know the organism, include it in your search. Searching for a protein from Homo sapiens is more precise than searching only for the protein name.
Step 3: Use reviewed filters
If you specifically want manually reviewed records, filter the results for reviewed entries.
Step 4: Check the accession number
Always record the accession number when using a specific protein record for research.
Step 5: Check the evidence
Do not assume that every annotation has the same level of experimental support. Look at the evidence associated with the information.
Step 6: Explore cross-references
UniProt records can connect to other resources such as structure databases, pathway databases, genome resources and protein family databases.
UniProt also provides advanced search and programmatic access through its REST API.
20. Important Official Websites and Resources
For study and research, it is better to use the official database websites rather than random third-party websites.
🔗 UniProt
🔗 UniProtKB
UniProt Knowledgebase – UniProtKB
🔗 UniParc
🔗 UniRef
UniProt Reference Clusters – UniRef
🔗 UniProt Help
UniProt Documentation and Help
🔗 UniProt REST API
🔗 EMBL-EBI
EMBL-EBI – European Bioinformatics Institute
🔗 MGnify – Current Metagenomics Resource
UniProt's official documentation provides direct information about UniProtKB, UniParc and UniRef, while the former UniMES resource has been retired and MGnify is recommended for current metagenomics work.
21. Important Exam Points for CSIR NET, GATE BT, DBT BET and JRF
- Swiss-Prot = reviewed protein database section.
- TrEMBL = unreviewed protein database section.
- UniProtKB = UniProt Knowledgebase.
- UniProtKB consists of Swiss-Prot and TrEMBL.
- UniParc = UniProt Archive.
- UniParc focuses on protein sequence archiving and sequence history.
- UPI = stable UniParc identifier.
- UniRef = UniProt Reference Clusters.
- UniRef100 groups identical sequences and qualifying fragments.
- UniRef90 represents approximately 90% identity clustering.
- UniRef50 represents approximately 50% identity clustering.
- UniMES = UniProt Metagenomic and Environmental Sequences.
- UniMES is retired/deprecated.
- MGnify is the current EMBL-EBI metagenomics resource recommended by UniProt.
Easy Memory Trick
Swiss-Prot → Reviewed TrEMBL → Unreviewed UniProtKB → Knowledgebase UniParc → Archive UniRef → Clusters UniMES → Retired
One-line revision
Swiss-Prot = Manual review | TrEMBL = Automatic annotation | UniProtKB = Knowledgebase | UniParc = Archive | UniRef = Clusters | UniMES = Historical metagenomic resource.
22. Quick Revision MCQs
Q1. Which section of UniProtKB contains reviewed protein entries?
Answer: UniProtKB/Swiss-Prot.
Q2. Which section contains unreviewed protein entries?
Answer: UniProtKB/TrEMBL.
Q3. What does UniProtKB stand for?
Answer: UniProt Knowledgebase.
Q4. What is UniParc?
Answer: UniProt Archive, a comprehensive protein sequence archive.
Q5. What is the identifier used by UniParc?
Answer: UPI or UniParc identifier.
Q6. What is UniRef used for?
Answer: Clustering related protein sequences to reduce redundancy and facilitate sequence searches.
Q7. What are the three major UniRef levels?
Answer: UniRef100, UniRef90 and UniRef50.
Q8. Which database was designed for metagenomic and environmental protein sequences?
Answer: UniMES.
Q9. Is UniMES currently active?
Answer: No. UniMES has been retired/deprecated.
Q10. What resource does UniProt recommend for current metagenomics research?
Answer: EMBL-EBI MGnify.
Q11. Which is generally manually annotated?
Answer: Swiss-Prot.
Q12. Which is automatically annotated?
Answer: TrEMBL.
Q13. What is the major difference between UniProtKB and UniParc?
Answer: UniProtKB focuses on protein knowledge and functional annotation, whereas UniParc focuses primarily on archiving unique protein sequences and tracking their history.
23. Frequently Asked Questions
UniProt is one of the most widely used resources for protein sequence and functional information.
Yes. In current terminology, Swiss-Prot is the reviewed section of UniProtKB.
Yes. UniProtKB/TrEMBL is the unreviewed, computationally annotated section of UniProtKB.
Swiss-Prot contains reviewed, manually curated entries, while TrEMBL contains unreviewed entries with computational annotation.
UniParc is used to archive protein sequences and maintain a record of their occurrence and history across source databases.
UniRef clusters related protein sequences to reduce redundancy and make large-scale similarity searching more efficient.
UniMES has been retired by UniProt. UniProt recommends the EMBL-EBI MGnify resource for current metagenomics research.
UniProtKB/Swiss-Prot is the reviewed section and is a strong starting point for high-quality manually curated protein information.
UPI is the stable unique identifier used for a UniParc sequence entry.
No. They are different but interconnected UniProt resources. UniProtKB focuses on protein knowledge, UniParc on sequence archiving and UniRef on sequence clustering.
24. Conclusion
Protein sequence databases are an essential part of modern biotechnology and bioinformatics. They provide researchers with access to protein sequences and the biological information needed to interpret those sequences.
The most important resource in this area is UniProt. Within UniProt, UniProtKB serves as the central knowledgebase. Its two major sections are Swiss-Prot, which contains reviewed and manually curated entries, and TrEMBL, which contains unreviewed computationally annotated entries.
UniParc has a different purpose. It acts as a comprehensive protein sequence archive, storing unique sequences and tracking their occurrence across source databases. UniRef groups protein sequences into clusters at different levels of sequence identity, helping reduce redundancy and improve the efficiency of similarity searches.
UniMES is important from a historical and examination perspective because it was developed for metagenomic and environmental protein sequences. However, it has been retired, and current metagenomics users are directed toward EMBL-EBI MGnify.
Figure: Quick visual revision of major protein sequence resources.
Final exam revision:
Swiss-Prot → Reviewed
TrEMBL → Unreviewed
UniProtKB → Knowledgebase
UniParc → Sequence Archive
UniRef → Sequence Clusters
UniMES → Retired Metagenomic/Environmental Resource
Understanding these differences makes many bioinformatics questions much easier. Once you know the basic roles of these databases, you can move on to related topics such as BLAST, FASTA, protein sequence alignment, PROSITE, Pfam, InterPro, PDB, protein structure prediction and molecular phylogeny.
📌 Recommended Official Resources
Biotech Notes Hub – Biotechnology, Bioinformatics, CSIR NET, GATE BT, DBT BET, JRF and Life Science study resources in simple language.
No comments:
Post a Comment