[Chimera-users] MD simulation in Chimera : 10 ns simulation is equivalent to 1 million of production steps 10 fs each?

Fri Jan 29 09:33:58 PST 2021

Dear Giovanni,
Sorry, it is beyond the scope of this software e-mail help address to give general research advice or teach about molecular modeling.  I can only recommend reading papers with similar studies, or textbooks on molecular modeling, or talking to any colleagues with more experience in modeling and molecular dynamics.

The MD tool in Chimera is slow to calculate, does not provide access to very many simulation parameters, or allow much control over parametrization of nonstandard residues like ligands, or include as much information in the output as dedicated dynamics packages (Amber, Gromacs, etc.).  Thus you will not be able to get quantities such as free energy differences (because they are not in the output), and it will not be possible (or very difficult) to run the longer calculations required for precision anyway because it will take too long, and you may not be able to use settings that might be needed for accurate results in your specific situation, because the software does not have controls for those settings.

The MD simulation tool was developed in collaboration with another group, who did much of the work. Chimera is no longer funded for any active development, we are only maintaining it currently.

Again, in my first reply mail I gave links that included advice on other software you could use (Amber etc.).  Yes they will be harder to learn and use, but that is because molecular dynamics simulations are complicated.  In order to do a good simulation, you will need to understand many different parameters in order to set them correctly for your specific project.

I hope this helps,
Elaine

> On Jan 29, 2021, at 9:18 AM, Giovanni Roviello <giroviel at unina.it> wrote:
> 
> Dear Elaine,
> 
> thank you very much. I will study the links you sent me. I am almost new in the field of MD (I work mainly in the field of synthetic chemistry and spectroscopy), sorry for so many questions. What one can adfirm from the specific plot of RMSD vs trajectory frame number I sent you in my previous email (attached)? Was the system stabilized enough to adfirm that the structure present at the end of the simulation is 'realistic'?
> 
> Another question, you already explained me that for long simulations and for quantitative scopes Chimera MD is not the ideal. Does it mean that it will not work or that it could work but with not realistic results? I wonder if your team will consider to improve in the future the MD tool extending it to larger molecular systems and also rendering suitable for longer simulation times. In fact, I find Chimera easier and more user-friendly than other programs.
> 
> 
> Thank you and kind regards,
> 
> Giovanni
> 
> _______________________________________________________
> Dr. Giovanni N. Roviello, Ph.D.
> Researcher at Istituto di Biostrutture e Bioimmagini
> IBB - CNR; Via Mezzocannone 16; I-80134
> Naples; Italy
> giroviel at unina.it
> giovanni.roviello at cnr.it
> 
> http://www.ibb.cnr.it/?command=viewu&id=387&lang=en
> https://orcid.org/0000-0001-6065-2367
> 
> 
> 
> Elaine Meng <meng at cgl.ucsf.edu> ha scritto:
> 
>> Dear Giovanni,
>> The X-axis of the plot shows the trajectory frame number, i.e. 10 means the 10th set of coordinates in whatever trajectory you are viewing.
>> 
>> From the MD Movie help:  "Clicking Plot plots the measurement value versus trajectory frame number."
>> 
>> <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/movie/framemovie.html>
>> <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/movie/movie.html#plotting>
>> 
>> How many time steps per trajectory frame depends on what value you chose in the Molecular Dynamics tool, Run Parameters, Other Runtime Options section, "Save once every [N] steps" (default N=10).   I don't know if you saved equilibration and production all in one trajectory file or in two separate trajectory files, and if the latter, I don't know which trajectory you are viewing in the MD Movie tool.  However, if the save option N= 10 and your time step is 1 fs, then each 1 on the x axis of the plot = 10 fs of time.
>> 
>> <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/md/md.html>
>> <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/md/md.html#run>
>> 
>> There is no way in Chimera to combine different kinds of measurements into a single plot, sorry.  You would need to use the "Dump Values" button at the bottom of each plot to save the values to a text file, and then use all these text files of values in some other plotting software. This Dump Values button is mentioned at the end of the MD Movie help section on plotting.
>> <https://www.rbvi.ucsf.edu/chimera/docs/ContributedSoftware/movie/movie.html#plotting>
>> 
>> I hope this helps,
>> Elaine
>> -----
>> Elaine C. Meng, Ph.D.
>> UCSF Chimera(X) team
>> Department of Pharmaceutical Chemistry
>> University of California, San Francisco
>> 
>>> On Jan 29, 2021, at 3:30 AM, Giovanni Roviello <giroviel at unina.it> wrote:
>>> 
>>> Dear Elaine,
>>> 
>>> I performed a MD simulation with 500000 (5.10^5) equilibration stes (time step 1 fs) and 5000 production steps (time step 1 fs). I obtained the attached RMSD Plot. I see RMSD values (Angstrom?) plotted against an unspecified 'x' ranging from 0 to 50000 (5.10^4). What do I have: time as ps?
>>> 
>>> I also wonder how I can save the MD movie window (with Analysis function) that appears after MD completion in view of reopening PLOTS in future and how can open the saved PLOTS (as .eps, .ps, .raw, .svg files), overlay each other,  in other words which software I can use to elaborate, overlay etc the Plots obtained by CHIMERA after different MD runs.
>>> 
>>> Thank you a lot for your assistance.
>>> 
>>> Kind regards,
>>> 
>>> Giovanni
>>> _______________________________________________________
>>> Dr. Giovanni N. Roviello, Ph.D.
>>> Researcher at Istituto di Biostrutture e Bioimmagini
>>> IBB - CNR; Via Mezzocannone 16; I-80134
>>> Naples; Italy
>>> giroviel at unina.it
>>> giovanni.roviello at cnr.it
>>> 
>>> http://www.ibb.cnr.it/?command=viewu&id=387&lang=en
>>> https://orcid.org/0000-0001-6065-2367
>>> 
>>> Elaine Meng <meng at cgl.ucsf.edu> ha scritto:
>>> 
>>>> Dear Giovanni,
>>>> Time step multiplied by number of steps is the total time.  However, we do not provide detailed advice over what settings you should use, because it always depends on your specific project.
>>>> 
>>>> However, Chimera is not recommended for long simulations or calculating precise quantities: In general, the Molecular Dynamics Simulation tool is a simple tool mainly used for teaching and small-molecule dynamics, or very limited structural relaxation of macromolecules.  Because the tool in Chimera is quite slow relative to other programs, it is not recommended for large (many-atom) systems, or long simulations, or for getting any quantitative result.  See these previous posts for more details and possibilities of other programs you could use instead.
>>>> 
>>>> <https://plato.cgl.ucsf.edu/pipermail/chimera-users/2020-September/017228.html>
>>>> <https://plato.cgl.ucsf.edu/pipermail/chimera-users/2020-September/017234.html>
>>>> 
>>>> I hope this clarifies the situation,
>>>> Elaine
>>>> -----
>>>> Elaine C. Meng, Ph.D.
>>>> UCSF Chimera(X) team
>>>> Department of Pharmaceutical Chemistry
>>>> University of California, San Francisco
>>>> 
>>>>> On Jan 24, 2021, at 6:37 AM, Giovanni Roviello <giroviel at unina.it> wrote:
>>>>> 
>>>>> Dear Sirs,
>>>>> 
>>>>> I would like to perform a MD simulation by UCSF Chimera.
>>>>> 
>>>>> Should I wish to perform a 10 ns simulation (for example) how could I set the parameters related to time? For example in production a number of phase steps of 1000000 and a time step of 10 fs is OK?
>>>>> 
>>>>> Does the relaxation time Nosé thermostat has effect on the overall simulation time?
>>>>> 
>>>>> In simpler words if I set 1000000 steps, 10 fs time step, and I leave selected Nosè thermostat with 0.2 ps (but leave unselected the barostat option) will it correspond to a 10 ns MD simulation?
>>>>> 
>>>>> Of course if I will publish using Chimera I will cite the program and the reference as indicated in
>>>>> 
>>>>> https://www.cgl.ucsf.edu/chimera/docs/credits.html
>>>>> 
>>>>> Thank you in advance for your kind help.
>>>>> 
>>>>> Kind regards,
>>>>> 
>>>>> Giovanni
>>>>> _______________________________________________________
>>>>> Dr. Giovanni N. Roviello, Ph.D.
>>>>> Researcher at Istituto di Biostrutture e Bioimmagini
>>>>> IBB - CNR; Via Mezzocannone 16; I-80134
>>>>> Naples; Italy
>>>>> giroviel at unina.it
>>>>> giovanni.roviello at cnr.it
>>>>> 
>>>>> http://www.ibb.cnr.it/?command=viewu&id=387&lang=en
>>>>> https://orcid.org/0000-0001-6065-2367
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> _______________________________________________
>>>>> Chimera-users mailing list: Chimera-users at cgl.ucsf.edu
>>>>> Manage subscription: https://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>>>>> 
>>>> 
>>>> 
>>> 
>>> <goodMDRMSD.png>_______________________________________________
>>> Chimera-users mailing list: Chimera-users at cgl.ucsf.edu
>>> Manage subscription: https://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users
>> 
>> 
> 
> <goodMDRMSD.png>