Hello,
I am new to Molecular Dynamics and eager to know how do I begin analysing. I will divide my query into points as follows:
1)I used namd2 command to get the output into an equilibration.out file. This is the line I wrote in my script that I submitted using sbatch (my job is currently in queue) I wrote four more similar lines for steps 6.2-6.6:
namd2 +p20 +idlepoll step6.1_equilibration.inp > step6.1_equilibration.out
Question: can I begin analysing the equilibration? if so, how? do we wait till equilibration.out file is produced and write a script that uses data from equilibration.out file to generate graphs ? Any guidance on how to specifically write this script is much appreciated.
2) My second question is : I read posts saying that you can analyse the script after ~60 ps to check certain parameters - see whether they stabilized. what are the parameters exactly? Energy of lipids and proteins and RMSD?I appreciate if you give me an information source to guide me through this.
3) How do we know if 60 ps of time has passed if I can only see the time in hours for running my job on server. is it according to the CHARMM GUI image I attached below such that one equilibration step produced ~50-100 ps have passed?
Equilibration is hard to determine, especially for large, complex and heterogeneous systems. Many observables fluctuate during the simulations and it is difficult to determine a criterion for the magnitude of fluctuations.
A simple way is to use the block averaging method. Ignore the equilibration assumption and just perform MD simulations. After gaining some statistics (e.g. 500 ns), you divide the whole 500 ns dataset into several smaller segments. Then you analyze the average of some observables of interest. When identical or similar estimates are obtained from different time blocks, you can say they are equilibrated and the sampling is converged. For those that deviate from the equilibrated ones in a statistically significant way, they could be regarded as non-equilibrated regions. Then, from the first equilibrated time block, you have an equilibrated system.
The equilibrium properties to check could be the density of the system and its decomposition into each region, the radial distribution function, the distance between some atoms, the dihedral of some side-chain atoms and so on.
Dear Asma,
Your questions are a bit general, so maybe I should do the same. I work more on applying MD, so my answer will be in such context.
1) The equilibration step is applied only to stabilize the system for the 'real' MD. Basically, you'll add temperature and pressure in a way to do not blow your system (that's why you use position restraint, for example).
Usually, we don't take a look at this step, but if you want this, maybe you would like to check the temperature, potential energy, pressure...it really depends on what you are looking for. In NAMD, the outputs are printed in the line 'ENERGY'. Take a look at these variables (and at the header too). You can output such columns and plot them. And, of course, wait the simulation finish before that.
2) After the equilibration, the MD starts properly. The variables that you will select are directly related to your questions. For example, if you want to check the stability of your protein (or complex), use RMSD; if you want to check how each residue vibrated, use RMSF; if you want to check the global form of you protein, use the radius of gyration analysis...Got it?! It's an open question.
3) In the output file, you will see the number of steps performed. So you should multiply the value to the time integration (in classical MD we use 0.002 ps). For 60 ps, you should do: 60 / 0.002 = 30000. So at 30000 steps you will have 60 ps.
I'll leave here some tutorials really useful. I hope they will help you in some way:
1. http://www.mdtutorials.com/gmx/
2. https://www.bonvinlab.org/education/molmod/simulation/
3. http://nmr.chem.uu.nl/~adrien/course/molmod/index.html#introduction
Best,
Antoniel.
principally equilibration phase (NVT o NPT ensemble), is used for "equilibrate", thermodynamics parameter and kinectis, for example, in nvt (heating in namd) you can view if no there is much variation in temperature, by other hand in npt you can view density of system if is corect to 1 bar o 1 atm of presion. Other good parameter is kinetic energy, this information are relationated with temperature control by maxwell-boltzman distribution. other common parameters are RMSD , RMSF, Rg are structural parameter for watch statility of your system. Them you analysis is varible and are associate to your investigation questions.
- Badges
- Science topic
- Similar topics
- Filing