|Téma:||Hledání míst pro vstup ligandu do proteinových tunelů|
|Vedoucí:||Ing. Vojtěch Vonásek, Ph.D.|
|Vypsáno jako:||Diplomová práce, Bakalářská práce, Semestrální projekt|
|Popis:||Proteins are complex macromolecules essential for processes in living cells and organisms.
Protein function is often manifested through interactions with other molecules: either other
proteins or small molecules, called ligands. The interaction between protein and ligand can
undergo in the protein's active site, which can be located either on the protein surface or buried
inside the protein. In the second case, the ligand needs to be transported to the active site from
the outer environment. Since the wet lab experiments for molecular docking are very costly and time consuming, the protein-ligand binding is typically investigated using in-silico simulations. In cases when the
ligand has to pass through a tunnel or cavity into the active site, its transportation becomes as
relevant as its final pose Methods based on molecular dynamics (MD) are computationally expensive and difficult to set-up. Docking-based tools sample ligand’s trajectory by iterative docking of the ligand along the tunnel. They are faster than MD and easier to set-up. However, they cannot observe the most appropriate positions of the ligand entering the tunnel at the tunnel mouth. Motion planning approaches can be used to
generate the entrance trajectories even in these situations.
The goal of this project is to investigate how to use sampling-based motion planning for finding ligand trajectories for entering protein tunnels. The search will be performed in configuration space considering the binding energy between the protein and the ligand. Monte-carlo search can be used to boost the search towards local optima. Necessary supporting libraries will be provided by the supervisor.
Protein (and ligand) are represented by a hard-sphere model. The task requires a (good) knowledge of c/c++, knowledge of python is a big advantage; knowledge of Linux is necessary.