Dear all,
I have docked a receptor protein to another receptor (Hetero) using ZDOCK and top 10 predictions are obtained. Of the 10, 9 predictions look little weird for me, where both the structures are absolutely perpendicular. For the remaining one, the two proteins are parallel but the protein heights are not even [ one protein is half above the other protein]. Now how do I evaluate these structures?
Also, I want to simulate the complex structure using GROMACS package. How can I place this unevenly distributed complex in a lipid bilayer? Moreover, I wish to perform steered MD simulations. Is it possible to perform sMD in lipid bilayer configurations?
I highly recommend you to use Roseta or at least Robetta server, it has a protocol for protein protein docking, and it is very accurate..you can generate 5000 models, order it by interface score and then use the best conformation for MD simulation
If you don't want to use the program for docking you can use the score function of Roseta to evaluate a lot of parameters of the model.
Hi Ms.Claudia.
Do robetta server really do protein-protein docking? I didn't find any unfortunately. Rosetta server using [http://rosie.rosettacommons.org/docking2] basically use GRAMM or ZDOCK for global positioning, which always leads my file to error, as my terminals are methyamidated and acetylated.
Confused.
Protein-protein docking methods are not reliable enough to be used for studying of how two large proteins link to each other unless you have access to experimental data about the interaction interfaces. These methods are feasible to refine the system and interface domains, not for predicting of the complex from ab initio.
You are right, Robetta don't perform protein protein docking, but using Rosetta software you should have no problems. This program is a little complicated because it doesn't have a graphic environment, but it is really powerfull
Dear Ramin,
I understand. But here we need to build something out of predictions for better results. So looking for best tools. As I have said in my question, the docking results were so weird. If there are any protocol that can be carried out before docking, then I can add it before I proceed further.
Thank you
You must search the literature and experimental reports and find at least one or two amino acids (the more amino acids, the more luck you have) contributing to the binding domains, then you can start with these methods.
Dear Ramin
Thank you. I will make a detailed survey.
And about the patched terminals, I have removed the patch using charmm-gui. might be useful for someone.
Any answers about the sMD part will be useful.
It depends on the aims of your projects what kind of strategy is used for your systems, sMD or others.
But for now your problem is not what type of MD simulation you implement for the systems; the main challenge is that how to construct the initial conformer. A good beginning makes a good ending.
- Badges
- Science topic