Hi,
Sorry for the delay in replying.
On Thu, May 16, 2019 at 1:02 AM Reber, David <David.Reber.empa.ch> wrote:
>
> Ok, so I can't finish writing the hbseries.gnu file before I run out of memory. In order to assess lifetimes, is it completely wrong to take hbavg.dat and then take the average of the frames column? In my case 1 frame per ps --> average hbond lifetime in ps?
Why not just skip writing the hbseries.gnu file at all? Just do the
analysis on the sets in-memory. If you don't have enough memory, maybe
try switching to the plain .dat format or try caching the series data
sets in memory (requires a recent version of cpptraj with the
'usediskcache' command - use it right before the 'hbond' command).
Also, if what you're looking for are the average behavior of water
around the ions, I recommend using the 'solventdonor' and
'solventacceptor' keywords for water. In other words, since you don't
care which particular water is around sodium, you don't need to track
each individual one. This will greatly cut down on the number of data
sets generated, and is probably more what you actually want. So
something like:
hbond sodium :Na+ solventdonor :WAT solventacceptor :WAT.O out
hbna.dat avgout hbnaavg.dat dist 4 angle -1 nointramol series
lifetime sodium name nalife out nalife.dat
Hope this helps,
-Dan
>
> I'm also trying to look at residence times of WAT around ions. I use
>
> trajin MD333.mdcrd
> hbond sodium donormask :Na+ acceptormask :WAT.O out hbna.dat avgout hbnaavg.dat dist 4 angle -1 nointramol
> go
>
> in hbnaavg.dat I get this:
> #Acceptor DonorH Donor Frames Frac AvgDist AvgAng
> Na+_688.Na+ WAT_955.H1 WAT_955.O 16873 0.3375 2.6007 51.0838
> Na+_688.Na+ WAT_955.H2 WAT_955.O 16873 0.3375 2.6007 52.0804
> Na+_600.Na+ WAT_1043.H2 WAT_1043.O 14103 0.2821 2.6007 50.9209
> Na+_600.Na+ WAT_1043.H1 WAT_1043.O 14103 0.2821 2.6007 51.1632
> ...
> Where I can delete every other line. Again, is it completely wrong to take the average of the frames column in order to compare the average residence time of WAT around Na+ in different systems?
>
> Thank you very much in advance,
> David
>
>
>
>
>
> > -----Original Message-----
> > From: Reber, David [mailto:David.Reber.empa.ch]
> > Sent: Donnerstag, 16. Mai 2019 03:52
> > To: AMBER Mailing List
> > Subject: Re: [AMBER] hbond lifetime filesize issues
> >
> > Hi Dan
> >
> > Thanks for your reply. You are right, it's the hbseries.gnu file that gets that
> > big. I assume I can try to let it finish and once hbwater-lifetime.dat is written I
> > can delete hbseries.gnu? Or can you recommend anything else to tackle
> > large files?
> >
> > Thanks and best regards
> > David
> >
> >
> > > -----Original Message-----
> > > From: Daniel Roe [mailto:daniel.r.roe.gmail.com]
> > > Sent: Mittwoch, 15. Mai 2019 15:36
> > > To: AMBER Mailing List
> > > Subject: Re: [AMBER] hbond lifetime filesize issues
> > >
> > > Hi,
> > >
> > > On Wed, May 15, 2019 at 5:53 AM Reber, David <David.Reber.empa.ch>
> > > wrote:
> > > > I do get the hbwater-lifetime.dat file but it is enormous (after 10GB I
> > > stopped).
> > > > Did I do something wrong or is this normal for a system containing many
> > > WAT? If yes, what's the best way to handle such large files in order to
> > extract
> > > relevant information? Basically I want to compare two electrolytes and see
> > if
> > > there's differences in hbond dynamics.
> > >
> > > It's not impossible, especially considering you're looking at WAT-WAT
> > > hydrogen bonds. There will be one line in the lifetime entry for each
> > > hydrogen bond, and even with only 1000 waters that's like close to 2
> > > million possible donor-acceptor pairs. However, what is unusual is
> > > that you're able to write out hbavg.dat but not hbwater-lifetime.dat -
> > > they should have the same number of lines. Are you sure the file
> > > that's gigantic isn't hbseries.gnu? I could easily see the time series
> > > file getting out of control.
> > >
> > > -Dan
> > >
> > > >
> > > > Thank you in advance for any hints!
> > > > David
> > > >
> > > > _______________________________________________
> > > > 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
> >
> > _______________________________________________
> > 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
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue May 21 2019 - 10:00:02 PDT
