Reading UniProt: a first look at protein sequences and functions

Reading UniProt protein sequences and functions — Nadoo Bio public database series

UniProt is a public database of protein sequences and functional information. A sequence is the order of amino acids, the small building blocks that make up a protein. Alongside sequences, UniProt records the organism a protein comes from, descriptions of what it does and supporting publications. [1]

Why might a search for one protein name return several results? Proteins with the same name may come from different species or occur in forms with different lengths or regions. UniProt makes those differences visible through sequences and functional descriptions.

UniProtKB: a profile for each protein

Proteins perform many tasks, including making substances and transmitting signals. Understanding those tasks begins with knowing which amino acids are connected, and in what order. UniProt represents that order as a sequence of letters.

UniProtKB is the central knowledgebase within UniProt. It connects each protein’s sequence with its name, organism, functional descriptions and relevant literature—a kind of protein profile. New research can lead to updated descriptions or sequence corrections.

A protein record connects its sequence, source organism and evidence for functional descriptions.

The sequence gives the amino-acid order, the organism identifies the protein’s source, and the evidence explains the basis for functional descriptions.

The same name can refer to different sequences

A human protein and a protein in another animal can share a name without having identical sequences. A study of a human protein’s function and a comparison across animals begin with different source materials. That is why the organism recorded in the entry matters.

An accession identifies the exact entry found through a name search. It is a unique identifier assigned to a UniProt entry. Because an entry can be updated, retaining the sequence used and the date it was checked makes later comparisons easier.

A single gene can also produce protein forms with different lengths or regions. These forms are called isoforms. [2] A UniProt entry has a representative, or canonical, sequence, but the research may require another isoform. The appropriate sequence depends on the form used in the experiment.

How Swiss-Prot and TrEMBL differ

Swiss-Prot and TrEMBL are names you will often encounter in UniProtKB. Swiss-Prot contains expert-reviewed entries, while TrEMBL contains entries organized primarily through automated analysis. [1] Descriptions attached to a protein’s function and features are called annotations.

Experts assemble annotations by reviewing publications and other sources. An annotation may reflect a direct experiment or knowledge inferred from similar proteins. Even within one Swiss-Prot entry, individual descriptions can therefore have different kinds of evidence. [3]

A paper establishing that a protein exists and a paper showing that it participates in a particular reaction support different claims. The sources attached to a functional description can indicate whether the protein itself was tested or whether the claim was inferred from similar sequences. It helps to read an entry’s overall review status separately from the evidence for each statement.

Preparing research inputs by checking the organism, sequence form and evidence behind the descriptions.

Choose the organism and sequence form appropriate to the research, then examine the evidence for the functional descriptions that matter.

The full sequence and the region used in an experiment

Parts of a protein can be removed after it is produced. A signal peptide is a region that helps direct the protein to its destination and may be removed during transport. The registered full sequence may therefore differ from the sequence of the mature protein that performs its function.

Researchers may also isolate a particular region for an experiment. A prediction made from the full sequence and an experiment using only part of it cover different material. Comparing them requires keeping the following information connected:

  • The organism and accession of the research target.

  • The canonical sequence or specific isoform.

  • Whether the input is the full sequence, the mature protein or a selected region.

  • The sequence range covered by the experimental or structural data being compared.

A selected sequence can be stored in FASTA, a text format with a name line followed by the amino-acid sequence. Computational tools receive that sequence or file as input, so the molecular types and sequence scope supported by the tool must match the material actually provided.

Connecting sequence and structure

A UniProt entry brings together information about a protein’s identity. The organism tells us where it comes from, the sequence specifies its form, and annotations and publications explain what is known about it.

Connecting this to a three-dimensional structure shows which shapes different parts of the sequence adopt. The PDB, a collection of experimentally determined structures, provides that next connection.

Terms of use

UniProt data can be used for both research and commercial purposes. Its CC BY 4.0 license permits copying, modification and redistribution with attribution. Users must credit the source, link to the license and indicate changes if they modify the data. [4] [5]

Separate rights associated with the data, such as patents, are not covered by this license. [4]

References

  1. UniProt. UniProtKB. Official help. Checked October 23, 2026.

  2. UniProt. Alternative products. Official help. Checked October 23, 2026.

  3. UniProt. Evidence. Official help. Checked October 23, 2026.

  4. UniProt. License & disclaimer. Usage policy. Checked October 23, 2026.

  5. Creative Commons. Attribution 4.0 International (CC BY 4.0). License deed, Korean version. Checked October 23, 2026.