Searching and visualising data from the Complex Portal

Searching for complexes

You can search complexes using:

  • any name(s) or identifier(s) of participants in the complex (for example, using ChEBI, RNACentral or UniProt IDs);
  • any name(s) or identifier(s) of the complex itself, including identifiers from external databases such as ChEMBL, The Gene Ontology, The PDB or Reactome;
  • the species.

A search can be performed in either the big box on the homepage, the small box on top of every other page, or in the URL [1]. The search is 'exact' but allows the use of Boolean operators and the wildcard (*).

To see how the search results will be presented (and what filters are available), you can click on the Examples listed on the right-hand-side of the screen [2]. 

For more detailed searches using the "Complex Query Language" go to the Help menu [3]. We also perform a search across all EMBL-EBI resources on your search terms [4].

Figure 1 Complex Search - showing the Complex Portal homepage and the search results page.

Search Results

Results are displayed as 10 hits per page.

Faceted filters are available in the left-hand margin, which allow you to search within the results [5].

Full EMBL-EBI search results are located in the drop-down table [4].

Figure 2 Complex portal results page, showing faceted filters.

Details View

Each complex is extensively annotated:

  • Each complex has a unique identifier, in the format: EBI-xxxxxxxxx (where x = {1-9}).
  • The recommended name is the name most commonly used in the literature – all alternatives are captured as synonyms.
  • We provide a systematic name, which is made up of concatenated gene names – these are based on the Reactome naming rules.*
  • Each complex is annotated with its function and, where applicable, further physical and biological properties.
  • Diseases, ligands and assembly terms are given, if applicable.
  • All participants of the complex are listed in a table, listing their unique identifiers (that link out to the relevant databases), names, descriptions, stoichiometry, molecule type and biological role (if applicable).
  • Cross references to the experimental or inferred evidence, the occurrence of the complex in another database, gene ontology and disease annotation are captured extensively.
  • The complex is visualised using an interactive viewer (see next page).
Figure 3 Details page for haemoglobin HbA complex.

Visualising complexes

The Complex Viewer depicts the participants in a complex as nodes and known binding regions as edges. The legend in the viewer shows the type of complex participant. If no specific binding sites are known, either no edges are displayed or the edges link molecules by their n-terminus (in the 'expanded view', see Figure 4).

The complex viewer is interactive:

  • Click and drag a node to change the layout [1]. Click <reset> to bring it all back together.
  • Click on a node to expand proteins to display binding regions in the sequence [2]. Hover over the c-terminus to swivel the protein bars around.
  • Press <shift> and click to expand the protein sequence to show the residue level [3].
  • If there are many instances of the same molecule, the stoichiometry for the whole complex will not be expanded, but it will be indicated in square brackets [4].

Feature views:

  1. MI Features: provides detailed binding site information. Hover your mouse over the binding edge to see binding site ranges [2].
  2. UniprotKB: displays all UniProt features. Hover your mouse over the sequene for more detail [5].
  3. SuperFamily: Hover your mouse over to reveal domain information [6].
  4. Interactor: shows unique interactors in the same colour [7].

The graphic can be exported as a .svg file.

Figure 4 Complex Viewer.