Improved visualisation and more at the aggregated views of proteins from PDBe-KB

PDBe-KB aggregated views for proteins - improved visualisation
21 February 2020

We are always looking for innovative and helpful ways to display the data on PDBe-KB’s protein-centric pages. We have now introduced a number of visualisation improvements, including adding the new Mol* viewer, emphasising key interaction sites, highlighting residue conflicts between PDB and UniProt, and integration of the PDBe search bar into these PDBe-KB pages.

 

The Mol* /'molstar/ browser-based molecular viewer (https://molstar.org) has recently replaced LiteMol on the PDBe search and entry pages. This new, improved viewer, developed in open collaboration between PDBe and RCSB PDB and CEITEC, has now also been added to the PDBe-KB protein pages, and therefore allows you to view protein structures and interaction sites in even better clarity. The PDBe search bar has now been integrated into these pages, therefore better connecting PDBe-KB with the powerful querying tools at PDBe and helping users of both resources more easily find the data that they need.

 

Mol* viewer displays ligand interactions in PDBe-KB aggregated views

Mol* viewer displaying the ligand environment for a haem molecule bound to Prostaglandin G/H synthase 2. The viewer has now been integrated into all sections of the PDBe-KB aggregated views for proteins.

 

We made further visualisation improvements to the ligand and macromolecule interactions, displayed on the ProtVista viewer. More common interacting residues are now displayed in a darker colour for each molecule, therefore allowing clearer identification of key interacting residues, both between macromolecular interaction partners and small molecules that interact with your protein of interest. This helps filter out the less prevalent and therefore less relevant interactions, ensuring that the most interesting interacting residues and interaction partners are emphasised.

 

For example, small molecules such as EDO (1,2-Ethanediol) tend to bind indiscriminately to the surface of proteins, and will therefore show up at the bottom of the list of bound small molecules, and will have light colours. In contrast, drug-like molecules tend to interact with well-defined residues consistently, and will therefore be displayed higher on the list and with darker colours (https://pdbe-kb.org/proteins/Q92793)

 

Protvista in PDBe-KB aggregated views, highlighting more common interactions more clearly

Protvista component on the PDBe-KB aggregated view for CREB-binding protein. The non-specific interactions, such as EDO and DMS, are reduced in intensity to emphasise the more specific and likely more relevant interactions, such as the naphthalene-containing ligands 99E and 99N.

 

We continue to integrate annotations for macromolecules and small molecules into PDBe-KB to add more depth and context to the data displayed. For example, the “Structures and Domains” section now displays conflicts between the amino acids in each individual PDB structure with the standard Uniprot sequence, therefore engineered mutations and other variants are clearly highlighted. This allows clear identification of specific PDB entries that have structural variation, compared to the standard protein structure. We also now display Interpro domains (https://www.ebi.ac.uk/interpro/) in the structures section, therefore increasing the coverage of identified sequence domains for the selected protein and improving the biological context.

 

We are continuously reviewing and improving our services and are always grateful for your feedback. You can provide this by clicking the feedback button on any PDBe or PDBe-KB page, or by emailing us at pdbekb_help@ebi.ac.uk.