Dear Researchers,
I am interested to measure the hinge motion around the hinge region of a protein and visualise the motion as well. By normal mode analysis and all-atom simulation, I generated pdb files of the protein. When distance between two atoms, one atom in domain-1 and another atom in the domain-2, was measured in the generated pdb files, the variation in the distance was small ( below 5 angstroms). In some videos of you-tube, a large motion of domains around the hinge regions were seen. I was expecting similar motion in my case too. Are there other ways to measure hinge motion?
Thank you for help,
Mahendra Thapa
Hinge motion allows two (adjacent) domains to move and interact in a better manner during protein-protein or protein-ligand interactions. The extent of movement of the hinge depends on several factors such as the type and nature of residues in the immediate or close neighborhood (about 4 A) with respect to the residue of interest. For instance if the proportion of gly is high in that region then flexibility would be more. It is not advisable to compare your data with what u see on you tube unless its the same proteins
How many residues are contained in the "hinge regions" you mention? The question has different answers if the 'hinge' is a single rotatable bond versus a disordered linker.
The visualization of normal modes as molecular movies is fairly biased towards the observation of hinge motions, because normal modes have well-defined nodes and anti-nodes.
Have you tried to project your all-atom simulation along your hinge-like normal mode's basis vector (eigenvector)? If it's one of the first modes, it should account for much of the trajectory's dynamics, but a higher mode may not be meaningful on its own (it may only be observed as something superimposed on other modes).
If your "hinge region" is many residues (like a disordered linker between two ordered domains) then the details of the normal mode analysis matter (like exactly how the initial coordinates are selected). Really, such a region cannot be rightfully assumed to have harmonic motions in the first place.
Dear Ryan, Leandro and Manan
Thank you for helpful suggestions.
(i) At first, I used web-server 'HingePro' which located hinge residue and also generated trajectories in first two slowest modes.
I then measured distances between a set of two atoms across the hinge residue in the pdb files generated in these two slowest modes and compared with corresponding distances in the starting pdb. There were small variation ( 1 to 2 angstrom) in distance between corresponding two atoms. I was expecting large variation as the server uses normal mode analyis (GNM and ANM) for global dynamics.
(ii) I also run 100ns all-atom MD simulation using AMBER suit, but a little improved results than as of (i) were seen.
I am now thinking to run aMD .
Thank you,
Mahendra
If you're looking for larger amplitude motions, definitely consider aMD or coarse-graining. For coarse-graining, I'm a fan of CafeMol by the Takada group (linked below.) You can also get a lot of mileage out of GROMACS+SMOG (linked below.)
http://www.cafemol.org/index.php
http://smog-server.org/
- Badges
- Science topic