Hi guys,
I want to perform some MD simulations of a protein that has been resolved by NMR spectroscopy (thus it has multiple structure models). Should I start with an average structure or should I randomly choose one of the model instead? Do you know how to calculate that structure using Gromacs? Any advice would be welcome.
Regards,
Xabi
The primary data yielded by NMR analysis is mostly local and more recently global geometric information about atoms within the structure. Typically, these include distance between pairs of atoms, dihedral angles (typically backbone φ angles and some side-chain χ1 angles) and sometimes global information such as the orientation of a given bond with respect to a fixed axis of the molecule. These data are used as "restraints" to reconstruct 3D models which are compatible with the NMR data. All calculations are performed directly in the physical space, starting with a random conformation of the macromolecule, which is progressively folded to satisfy the restraints. Typically, several runs are performed, starting from different initial conformations, in order to check that the calculation converges onto a single solution. The result is thus an ensemble of models, the distribution of which gives a measure of the precision of the NMR structure.
In principle, I would suggest visualizing all the conformations with vmd and aligning their backbones, to see the majority clusters of the general structure. You can do this by opening the .pdb file in vmd, displaying it in the NewCartoon representation and coloring it by secondary structure. In the same window of graphic representations, go to the Trajectory tab and in "draw multiple frames", you should delete "now" and write 0: (number of frames that have the pdb) Example 0:15.
Once you can see all the images, you align them. For that, go to Extensions> Analysis> RMSD Trajectory tool. Select the protein, the backbone and the 0 frame as the reference one. Then, ALIGN.
Once aligned, you will surely notice if there is much variability in certain areas of the protein.
You can calculate the RMSD of each conformation from Extensions> Analysis> RMSD Visualizer. And try to choose a conformation within which they remain stable.
However, if what you are going to do is a simulation of molecular dynamics, when you stabilize the protein and also in the executed simulation, the protein accesses all degrees of freedom. If the result of the MNR does not show drastic changes in the structure, what frame is choosing would not be decisive.
Regards,
Flor
Thank you very much for your kind and detailed answer. I do agree with you that if the models from the NMR structure are similar, starting a MD simulation with any of those models will be fine: during the simulation the protein will be able to explore some (or all) of those conformations. But what happens when the NMR models show two extreme clusters? Why an average structure would be more suitable for the MD simulations? In my opinion, in that case, several MD simulations should be performed with different representative starting conformers rather than with an average calculated structure...Hope someone can enlighten me!
Dear Xavier,
Kindly, I will suggest that you be a little more specific when you ask your questions, to avoid getting answers that do not indicate what you really want to solve.
In relation to your "new question", I recommend the gmx cluster of Gromacs tools, to solve your problem. You can save using vmd a frame of your protein in .gro and all the frames together in .trr format (surely you already knew it, I put it in case this is useful for another person), what saves you steps, for your choice of the conformation most representative of the most populated clusters. You can choose, of course, one conformation of each cluster.
Regards
- Badges
- Science topic