Predict protein interaction surface with the ARDOCK server ========================================================== Scientific Purposes ------------------- ArDock is a structural bioinformatics web server for the prediction and the visualization of potential interaction regions at protein surfaces. Many biological function rely on the formation of protein-protein complexes. The identification protein interaction region is therefore of critical interest for our understanding of biology. The service is accessible through a web interface, both for submission and the reporting of results. How it works ? -------------- ArDock ranks the surface residues of a protein according to their tendency to form interfaces in a set of predefined docking experiments between the query protein and a set of arbitrary protein probes. The ArDock methodology is derived from large scale cross-docking studies where it was observed that randomly chosen proteins tend to dock in a non-random way at protein surfaces. This method is implemented on a computational cluster, where docking calculations run in parallel and results are aggregated and returned to the web client. Inputs ------ - A protein structure in `PDB format `_ - Tokens obtained in previous analysis can be given to the server to restore their results Outputs ------- - Interactive and 3D vizualisation of the results on the protein structure - Interactive scoreboard of the protein residue scores - Result files available for download in CSV and PDB formats Targeted audience ----------------- All scientists working in biology related areas where protein study is relevant with a focus on structural biologists and biochemists. Additional resources --------------------- * A `tutorial `_ was set up by the Ardock team * Publications describing `methods `_ and `implementations `_ * The public database of protein structures is avaible at `the PDB web site `_