"3D-Beacons is an open collaboration between providers of macromolecular structure models. The goal of this collaboration is to provide model coordinates and meta-information from all the contributing data resources in a standardized data format and on a unified platform." (3D-Beacons)
Till now large entities, such as SWISS-MODEL, PED, Genome3D, PDBe, and PDBe-KB provided structures via 3D-beacons. The emergence of cutting edge deep learning methods (e.g. AlphaFold, RoseTTAfold) makes it possible for smaller laboratories to generate high quality structure predictions supplemented with their experience in a given field of proteins. Therefore, an objective of the 3D-beacons consortia is to extend the 3D-beacons client framework for smaller teams. We both participate in developing the 3D-beacon client framework and serve as a test site.
Since we study transmembrane ABC proteins, we generated structural models for those human ABC proteins, which are half transporters, thus their functional and physiological structural unit (dimer) is not present in the AlphaFold Protein Structure Database. Some of our entries possess corresponding experimental structure. We provide these models as well, since they also contain regions previously not resolved experimentally. The structures and their metadata are served in a programmatic way (e.g. via the 3D-Beacon Hub) as
- cif (https://3dbeacon.hegelab.org/cif/hege-abc-0001.cif) and
- json (https://3dbeacon.hegelab.org/uniprot/Q9H222.json) files.
In order to provide an overview of these structures, we built this web application, which also contains additional information on the predicted structures (e.g. how the models were created).
