PROTEIN-PROTEIN DOCKING USING LOCAL SHAPE INVARIANTS

点击翻译按钮获取中文摘要

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. The aim of the project is to design and develop algorithms to predict protein-protein interactions using docking. The structure of the docked complex is useful in elucidating the biological function of the protein. Since solving a protein structure experimentally is time consuming and technically challenging, development of computational docking has become an urgent task in protein bioinformatics. The starting point in this approach are the crystallized structures of the interacting proteins which are treated as rigid, and the conformational space generated by the two interacting proteins is explored extensively. To help guide the search, a set of moment invariants are used. The underlying idea is to capture local shape properties defined over a small neighborhood of a set of interest points, in terms of a set of numbers called Zernike moments. These local descriptors are invariant to rotation transform and robust to noise. The rotational and translational space (6D space) is sampled using a geometric hashing algorithm. The predicted models are then ranked using a suitable energy function. This step is followed by the subsequent refinement of the top ranking models using molecular dynamics simulations. Proteins are inherently flexible which means that its 3-D structure may change under different conditions. Accounting for this is very challenging as in addition to the sampling of the rigid-body orientations, due consideration needs to be given to the folding of the protein. In order to avoid the heavily time consuming search through the entire flexible conformational space of two proteins during the docking or refinement process, an ensemble approach is adopted wherein, a pre-generated set of different feasible conformations are cross-docked.