Dear Vinicius,
I see. Yes, your explanation clarifies the nature of those last digits.
Thank you very much for your support and help in addressing this issue.
Best regards,
Elvira
пт, 11 авг. 2023 г. в 18:07, Vinicius Wilian Cruzeiro <vwcruz.stanford.edu>:
> Ohhh... Ok, ok. I finally understand your problem.
>
> When you restart your simulation, the initial time is read from your
> restart file. You can check what that time is with the following
> command: ncdump -v time [path to your rst7 file]
>
> Notice that the "TIME(PS)" section in your MDOUT file also follows from
> that value.
>
> Now... You must be thinking why the last 3 digits are then changing after
> a long simulation time, right? My guess is that it is happening due to a
> numerical noise. In memory, your timestep is actually 2 fs +- a tiny noise,
> so you need to simulate very long to see an effect on the third digit of
> the simulation time. Makes sense?
>
> I hope this clarifies all your doubts now, but please let me know if you
> have any further questions.
>
> All the best,
>
> *Vinícius Wilian D. Cruzeiro, PhD*
> Postdoctoral Researcher
> Department of Chemistry
> Stanford University
> Twitter: .vwcruzeiro
> ------------------------------
> *From:* Elvira Tarasova <elviratarasovachem.gmail.com>
> *Sent:* Friday, August 11, 2023 2:50 PM
> *To:* Vinicius Wilian Cruzeiro <vwcruz.stanford.edu>
> *Cc:* amber.ambermd.org <amber.ambermd.org>
> *Subject:* Re: [AMBER] Problem with time records in cpout- files in REMD
> CpHMD simulation
>
> Dear Vinicius,
>
> You're right, the left part (before the decimal point) increments by 2 ps
> (1000 steps * 0.002 ps). I can reconstruct it by incrementing it with a
> value of 2. Let me explain with some examples where my problem is.
> Below are 112 consequent records of time and time step in my cpout file:
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 100
> Time: 8400.198
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 1000
> Time: 8401.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 2000
> Time: 8403.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 3000
> Time: 8405.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 4000
> Time: 8407.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 5000
> Time: 8409.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 6000
> Time: 8411.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 7000
> Time: 8413.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 8000
> Time: 8415.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 9000
> Time: 8417.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 10000
> Time: 8419.998
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 11000
> Time: 8421.998
>
> I do not understand where .998 is coming from. Later in the file it
> changes, and I do not understand the pattern of the changes and how to
> calculate it.
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 2126000
> Time: 522651.*996*
>
> Solvent pH: 7.50000
> Monte Carlo step size: 100
> Time step: 2127000
> Time: 522653*.995*
>
> Please, let me know how I can calculate those digits after the decimal
> point.
> If you need more details or context, please let me know.
>
> Best regards,
> Elvira
>
> пт, 11 авг. 2023 г. в 14:42, Vinicius Wilian Cruzeiro <vwcruz.stanford.edu
> >:
>
> Hello Elvira,
>
> The challenge lies in reconstructing the three digits after the decimal
> point.
>
> Sorry, I don’t understand why this is a problem. On each block you also
> have the "Time step", right? From this information and the dt variable, you
> can directly compute the "Time".
>
> I hope this helps,
>
> *Vinícius Wilian D. Cruzeiro, PhD*
> Postdoctoral Researcher
> Department of Chemistry
> Stanford University
> Twitter: .vwcruzeiro
> ------------------------------
> *From:* Elvira Tarasova <elviratarasovachem.gmail.com>
> *Sent:* Friday, August 11, 2023 11:13 AM
> *To:* Vinicius Wilian Cruzeiro <vwcruz.stanford.edu>
> *Cc:* amber.ambermd.org <amber.ambermd.org>
> *Subject:* Re: [AMBER] Problem with time records in cpout- files in REMD
> CpHMD simulation
>
> Dear Vinicius,
>
> Thank you for your reply.
> Unfortunately, rerun of this simulation is not a possibility as this is a
> very computationally expensive simulation.
>
> As far as I understand, timestep changes (0.002 ps) primarily impact the
> numerical portion to the left of the decimal point, it is not a problem to
> reconstruct them. The challenge lies in reconstructing the three digits
> after the decimal point.
> Where those three digits connected to/responsible for? If I change all
> records "Time: **********" with accurately reconstructed time part to the
> left of the decimal point and set three digits after the decimal point to
> random number (let's say to ".000"), will it have a negative impact on the
> results?
>
> Best regards,
> Elvira
>
> чт, 10 авг. 2023 г. в 13:41, Vinicius Wilian Cruzeiro <vwcruz.stanford.edu
> >:
>
> Hello Elvira,
>
> These 3 digits are defined in the function that prints the CPOUT file
> inside the Amber code. It was likely chosen this way to capture the
> timestep changes, which are on the order of 0.002 ps.
>
> If you want to rerun your similation, I can show you what you need to
> change in Amber to avoid the ****. Just let me know.
>
> All the best,
>
> *Vinícius Wilian D. Cruzeiro, PhD*
> Postdoctoral Researcher
> Department of Chemistry
> Stanford University
> Twitter: .vwcruzeiro
> ------------------------------
> *From:* Elvira Tarasova <elviratarasovachem.gmail.com>
> *Sent:* Thursday, August 10, 2023 9:41 AM
> *To:* Vinicius Wilian Cruzeiro <vwcruz.stanford.edu>
> *Cc:* amber.ambermd.org <amber.ambermd.org>
> *Subject:* Re: [AMBER] Problem with time records in cpout- files in REMD
> CpHMD simulation
>
> Dear Vinicius,
>
> As I mentioned in my previous email, fixremdcouts.py worked wonderfully
> for sorting the cpout files. However, I have an issue when attempting to
> use the sorted cpout files for computing protonation state statistics. The
> cphstats is unable to process them, presumably because they still contain
> the "Time: **********" record.
>
> It seems that I'll need to fix the time records to resolve this problem.
> However, I'm a bit confused about the format of the time record. Could you
> please explain the significance of the three digits after the decimal point
> in the time records in the cpout files? Here's an example of what I'm
> referring to:
>
> Time: 578417.*995*
> Time: 588400.*195*
> Time: 588401.*995*
> ...
> Could you point me to where I can find these three digits after the
> decimal point?
>
> Thank you for your help.
> Best regards,
> Elvira
>
>
> вт, 8 авг. 2023 г. в 14:45, Elvira Tarasova <elviratarasovachem.gmail.com
> >:
>
> Dear Vinicius,
>
> Thank you for your reply. I used fixremdcouts.py as you suggested and it
> worked!
> Thank you very much for your help.
>
> Best regards,
> Elvira
>
> пт, 4 авг. 2023 г. в 17:22, Vinicius Wilian Cruzeiro <vwcruz.stanford.edu
> >:
>
> Hello Elvira,
>
> I am sorry I missed your e-mail before.
>
> It is very likely that cphstats is having a problem while trying to parse
> **** as a float. If you want to use cphstats, you will need to manually
> correct the time stamps (or write a script to do that).
>
> However, you can parse your cpout files using an alternative AmberTool.
> There is a chance it will not break because of this issue. All you need to
> run is: fixremdcouts.py --couts *.cpout . To get the full usage details,
> type: fixremdcouts.py --help
>
> I hope this helps. Please let me know if you have any further questions.
>
> All the best,
>
> *Vinícius Wilian D. Cruzeiro, PhD*
> Postdoctoral Researcher
> Department of Chemistry
> Stanford University
> Twitter: .vwcruzeiro
> ------------------------------
> *From:* Elvira Tarasova via AMBER <amber.ambermd.org>
> *Sent:* Friday, August 4, 2023 1:50 PM
> *To:* amber.ambermd.org <amber.ambermd.org>
> *Subject:* Re: [AMBER] Problem with time records in cpout- files in REMD
> CpHMD simulation
>
> Dear Amber community,
>
> I'm reaching out to remind everyone about my previous question, which
> unfortunately hasn't received any responses yet. I would really appreciate
> any help or insights you can provide.
>
> The question is below:
>
>
>
>
> Dear Amber-community,
>
> I have a question regarding REMD CpHMD simulation.
>
> During my REMD CpHMD simulation (I used Amber 18), which spanned 1500 ns
> with a pH range from 3.0 to 7.5 and a 0.5 increment, I faced an error
> related to the cphstats program (cphstats --fix-remd) while preprocessing
> the cpout-files after the first 1000 ns. The specific error message was as
> follows:
>
>
> *Did not recognize the format of cpout remd-*_pH3_5_00101.cpout.Error:
> Cpout file remd-*_pH3_5_00101.cpout is invalid! Skipping.*
>
> Upon inspecting the cpout files, I noticed that time records were missing
> after
> 1000ns, replaced instead by *"Time: **********"*. The last available time
> record was "*Time: 999999.990*".
> I assume that the absence of time records is the root cause of the error.
> Please, correct me if I am mistaken.
>
> Therefore, I would like to ask you if there is a solution to this problem
> that doesn't involve manually fixing the time records. If there are
> alternative methods to resolve this issue, I would greatly appreciate your
> guidance.
>
> Furthermore, I would like to understand the significance of the three
> digits after the decimal point in the time records in cpout-files, as shown
> in example below:
> Time: 578417*.995*
> Time: 588400*.195*
> Time: 588401*.995*
> ...
> And where can I find these three digits after the decimal point in case I
> need to fix this problem manually?
>
> Thank you for your time and support.
> Best Regards,
> Elvira
>
>
>
> ср, 19 июл. 2023 г. в 13:31, Elvira Tarasova <elviratarasovachem.gmail.com
> >:
>
> > Dear Amber-community,
> >
> > I have a question regarding REMD CpHMD simulation.
> >
> > During my REMD CpHMD simulation (I used Amber 18), which spanned 1500 ns
> > with a pH range from 3.0 to 7.5 and a 0.5 increment, I faced an error
> > related to the cphstats program (cphstats --fix-remd) while preprocessing
> > the cpout-files after the first 1000 ns. The specific error message was
> as
> > follows:
> >
> >
> > *Did not recognize the format of cpout remd-*_pH3_5_00101.cpout.Error:
> > Cpout file remd-*_pH3_5_00101.cpout is invalid! Skipping.*
> >
> > Upon inspecting the cpout files, I noticed that time records were missing
> > after 1000ns, replaced instead by *"Time: **********"*. The last
> > available time record was "*Time: 999999.990*".
> > I assume that the absence of time records is the root cause of the error.
> > Please, correct me if I am mistaken.
> >
> > Therefore, I would like to ask you if there is a solution to this problem
> > that doesn't involve manually fixing the time records. If there are
> > alternative methods to resolve this issue, I would greatly appreciate
> your
> > guidance.
> >
> > Furthermore, I would like to understand the significance of the three
> > digits after the decimal point in the time records in cpout-files, as
> > shown in example below:
> > Time: 578417*.995*
> > Time: 588400*.195*
> > Time: 588401*.995*
> > ...
> > And where can I find these three digits after the decimal point in case I
> > need to fix this problem manually?
> >
> > Thank you for your time and support.
> > Best Regards,
> > Elvira
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Aug 11 2023 - 15:30:02 PDT
