ABSTRACT: Interatomic contact areas as means for the analysis and modeling of protein structures and interactions

We would like to present the abstract of our keynote speaker Justas Dapkūnas, who is a researcher at the Department of Bioinformatics of Vilnius University Institute of Biotechnology.

Interatomic contact areas as means for the analysis and modeling of protein structures and interactions

Structural data on proteins and their interactions is indispensable for understanding the molecular details of biological processes. This data may also guide the design of proteins having desired new features or forming novel interactions. Unfortunately, experimental structures are available only for a small part of known proteins and for an even smaller part of known protein interactions. Nowadays protein structures can be often predicted using computational modeling. However, the analysis of the structures, either experimental or predicted, might be also challenging. As a result, various computational methods are widely used in structural biology.

Here we present a novel approach to protein structure analysis that is based on interatomic contacts. The contacts between atoms are defined using the Voronoi tessellation of macromolecular structures. This method allows straightforward estimation of the contact areas between atoms that reflect the physical interactions in the protein molecule.

There are multiple applications of contact area-based methods for protein structure analysis that clearly demonstrate the usefulness of this concept. For example, contact area differences may be used to compare protein structures and their interaction interfaces using our novel method CAD-score. Clustering of interaction interfaces according to CAD-score was implemented in the PPI3D web server for searching and analyzing of structural data on protein interactions, where it reduces the redundancy of available experimental data and thus simplifies the analysis. This clustering was one of the main reasons why PPI3D was highly effective in finding templates for comparative modeling of protein complexes in recent CAPRI experiment.