The main purpose of this server is to allow individual and comparative analysis of protein interfaces at atomic level.

Protein interactions have been classified at family level following the SCOP hierarchical scheme (CL: class, CF: class fold and SF: super family) that we extended by including three interacting levels (FA: family, BR: binding region, IF: interface. The binding regions are defined as distinctive surface regions of a protein family used to recognize other molecules. The interfaces distinguish the different partners or ligands that a specific region can recognize. Ligands include proteins, peptides, nucleic acids and saccharides.

1) Query

A protein family of interest might be searched by:

  • name (or part) of a class, classfold, superfamily or family of SCOP e.g. "globin"
  • SCOP id (e.g. "46463"); SCOP hierarchy code (e.g. "a.1.1.1")
  • A PDB structure might be searched by PDB id (e.g. "1aui") or sequence (e.g. "[SRVVTNADSKAAY]")

2) Search options

Ligand type: A protein may interact with ligands of different molecular nature (other proteins, peptides, DNA, RNA or saccharides). The user can select binding regions that can recognize more than one type of ligands at the same time (or/and command).

Complex type: Two interacting domains may belong either to the same (homo) or different (hetero) protein families.

Interaction type: Two interacting domains can either belong to the same (intra) or to different (inter) protein chains. Saccharides can also be observed covalently bond to the protein (intra).

Binding region type: By default all binding regions per family are shown: observed as those inferred at family level, and predicted as those inferred from other protein families.

Clustering cutoff: All protein complexes per family have been clustered at different binding similarity cutoffs (see Introduction to read about the methodology)

3) SCOWLP navigation: Binding regions page (BR)

SCOWLP web application facilitates the navigation through the SCOP hierarchy until family level, where the binding regions are shown. This page is useful to analyze the different regions used by a protein family to recognize other molecules.

The page consists in 3 interrelated frames: 1) Alignment, 2) 3D Viewer 3) Control

3.1) ALIGNMENT frame

The structure-based sequence alignment uses the representative sequences to visualize the binding regions by highlighting the interacting residues. Different patterns should be observed for each BR.

Binding regions are represented by a SCOWLP id that can be click to obtain the interfaces formed with different ligands (IF level)

Click the blue arrow to download a text file with the PDB id and chains that belong to the binding region or interface analyzed. In the Interface and PDB pages, the arrow is colored red and is used to download a text file with the Residue Interactions shown in the CONTROL frame.

Click to visualize the binding region in the VIEWER and to obtain more information about the binding region in the CONTROL.

3.2) 3D VIEWER frame

A Jmol plugin is used to visualize a representative structure of the protein family showing the surface region used for binding upon selection of a BR_id in the ALIGNMENT. It corresponds to a schematic representation of the spatial locations of the binding regions.

3.3) CONTROL frame

It gives information about the selected binding region in the ALIGNMENT, such as the representative structure shown in the VIEWER and the interacting partners.

4) SCOWLP navigation: Interfaces page (IF)

Once a binding region is selected by clicking on the BR ID (i.g. BR 18798) the different protein interfaces known can be analyzed in the interface page. Each interface corresponds to a protein complex with a common receptor, where the ligand differs. The ligand type filter can be useful at this stage to select protein families with regions able to recognize different molecules automatically; otherwise it would be an extremely tedious job.

The interacting residues highlighted in the ALIGNMENT should have a similar pattern since it corresponds to the same binding region. The VIEWER shows a representative protein complex for each interface (i.g. IF_18799), where differences at the structure of the ligand should be appreciated.

5) SCOWLP navigation: Interacting domains page (ID)

Once an interface is selected by clicking on the IF ID (i.g. IF_32) all interacting domains are shown in this page. Note, that all complexes share common receptors and ligands (have the same receptor-ligand composition). Analysis of the recognition properties can be done at this level combining the ALIGNMENT, VIEWER and CONTROL frames through two extra features:

  • Coloring interacting residues by water mediation and chemical interaction types

  • Connectivity of each interacting residue (ALIGNMENT) with the position in the structure (VIEWER) and its description (CONTROL: Residue interactions).

NOTE: When alignment is colored by "water mediation", water molecules that mediate residue-residue interactions are shown in the VIEWER and can also be independently highlighted by clicking on their ids located in CONTROL : Residue interactions : water mediated : wet.

6) SCOWLP navigation: PDB page (PDB)

By clicking or querying a specific PDB code, SCOWLP shows all existing domain interactions. The chain-chain order corresponds to receptor-ligand, i.e interaction AB is not the same as BA since the first chain is taken as reference to show the interacting residues with the second. The information is sorted by receptor's family and by binding region, where interacting patterns should be very similar.

Once the user selects any of the family binding regions, the analysis of the interacting information can be performed on the same way as explained above.

7) Predicted novel binding regions and similar interfaces

The BR page contains not only family binding regions but sometimes predicted ones inferred from similar structures of other protein families (see methodology). This information is hidden for simplicity under the text "There are also predicted binding regions" and can be shown by "Click here to show them".

The aligned sequences correspond to a family representative, where the interacting residues from the inferred binding region are mapped on top. The structural icon allows to show the spatial location of the predicted regions in the 3D VIEWER and the list of receptors/ligands used for the inference in the CONTROL. By clicking the structural icon, the user obtains the list of families used to infer this binding region.