Hi, a simulation lenght depends from what information you want to get from your MD.

Generally a MD simulation should be carried out untill the property you are looking for reaches convergence. For example, if you are interested in generating a set of snapshots for MMPBSA analysis, the property you might look at can be the protein RMSD, so you can run your simulation untill the RMSD vs time plot reaches a plateau. If you start from a well defined crystal structure, generally you'll reach convergence in a relatively short time. Otherwise, if you have to test different ligands docked in the same protein structure, simulation time will depend on how the protein need to "adapt" to the different ligands.

Differently, if you need or want to observe large protein motion following a particular "perturbation" (i.e. activation following the binding of an agonists to an inactive state of the protein), simulation time might be decidedly longer.

You might want to give a look to http://dx.doi.org/10.1016/j.jmgm.2013.02.002

The MD simulation time depends upon protein size (i.e. no. of amino acids). Big protein takes more time to obtain a stable confirmation. You can consider RMDS as a papramenter to consider MDS time. Secondly, MD time depends upon your aim (why you want to go for MD, what are your expected results from MD), if you are performing it on a ligand-receptor docked complex then you can go for a short MD time i.e. 1-2 ns. But for the homology models, you really require a long MDS duration (may be several ns).

@Girdhar Deora; Im afraid that 1-2ns would not be sufficient for predicting motions of complexes especially when ligands are involved. A general notion is that 10-30ns is minimally required to predict overall conformational changes for average length proteins. Ofcourse, 1-2ns would be sufficient for smaller peptides.

hi girdhar..

thanx for your reply...as I have already mentioned I am very much new to MD..can you please tell what kind of observations you expect from long simulation run in comparison to short simulations???

how one can assure that a given simulation is optimum (in terms of time period)??

Hi, a simulation lenght depends from what information you want to get from your MD.

Generally a MD simulation should be carried out untill the property you are looking for reaches convergence. For example, if you are interested in generating a set of snapshots for MMPBSA analysis, the property you might look at can be the protein RMSD, so you can run your simulation untill the RMSD vs time plot reaches a plateau. If you start from a well defined crystal structure, generally you'll reach convergence in a relatively short time. Otherwise, if you have to test different ligands docked in the same protein structure, simulation time will depend on how the protein need to "adapt" to the different ligands.

Differently, if you need or want to observe large protein motion following a particular "perturbation" (i.e. activation following the binding of an agonists to an inactive state of the protein), simulation time might be decidedly longer.

You might want to give a look to http://dx.doi.org/10.1016/j.jmgm.2013.02.002

@ alessandro:

thankx alot for your suggestion...can u please mail me abovementioned link as it is not possible for me to acess this paper from my institute.

My mail id is : [email protected]

@ Lasitha, my apologies for any confusion it created. but as per my experience, to check initial docked protein-ligand complex stability a short MD time up 5 ns is enough, you can refer several research papers on it. (no doubt in case of small protein, as in case of large protein surely it require long time), again, if we looking for preferable confirmation of a flexible loop or flexibility in various protein structure/loops then no doubt we have to go for long and multiple MD simulation cycles.

@ Giridhar , you mentioned some research papers with short MD time of 5 ns . for protein ligand complex stability. I would be grateful if you could post some links for those papers ........i am an undergraduate student in similar research and am having doubts about simulation length

@vishal: do update me also ...if you happened to get some good material regarding simulation length

In general, now days, its usual to have atleast 200 ns simulation. But MD simulation time depends upon the achievement of stability of the trajectory. Once you see that the trajectory is stable for 50-60 ns continuous, u can use it for analysis, so sometimes even 100 ns works excellent.

It fully depends upon your problem only. If you want to see protein folding, then nanosecond scale never help you. You must do microseconds. If you want to see drug protein interactions, then depending upon your computational infrastructure do as much you can. Increase in time scale give you more reliable results. Never judge your time scale based on what reviewer wants. It must depend upon the question you want to address. You can also do enhance sampling simulation, where you can decrease the timescale to get your desired results. By the way if you can mention which type of problem you are addressing then we can give you more accurate information.

Happy Simulation!!!