[gmx-users] Fw: wrong number of control atomes in hdb

Mehreen Jan mehreen_jan1990 at yahoo.com
Tue Feb 9 10:47:25 CET 2016



--- On Tue, 9/2/16, Mehreen Jan <mehreen_jan1990 at yahoo.com> wrote:

> From: Mehreen Jan <mehreen_jan1990 at yahoo.com>
> Subject: wrong number of control atomes  in  hdb
> To: "Mehreen Jan" <mehreen_jan1990 at yahoo.com>
> Received: Tuesday, 9 February, 2016, 2:44 PM
> 
> --------------------------------------------
> On Tue, 9/2/16, Mehreen Jan <mehreen_jan1990 at yahoo.com>
> wrote:
> 
>  Subject: Re: gromacs.org_gmx-users Digest, Vol 142, Issue
> 42
>  To: gromacs.org_gmx-users at maillist.sys.kth.se
>  Received: Tuesday, 9 February, 2016, 2:43 PM
>  
>   Subject: Fw: wrong number of control
>  atomes in
>  > hdb
>  > >
>  
>  
>  > > > > respected sir!
>  > > > > thank you for advise
>  > > > >
>  > > > > thank you sir for advise i just copy
>  folder
>  > 43A1p
>  > > and
>  > > > past
>  > > > > in top file and in GMXLIB and it work
>  error
>  > is
>  > > solved
>  > > > know
>  > > > > after running the command:
>  > > > >  pdb2gmx -f protein.top -o
>  protein.gro
>  > > > > select the forcefield 43A1p the
>  following
>  > error
>  > > is
>  > > > occurs
>  > > > > i google it but i can't get it exactly
>  and i
>  > read
>  > > > manual but
>  > > > > difficult to understand because of 
>  no
>  > experience
>  > > .
>  > > > > respected sir kiendle find the
>  attachment .
>  > > > >
>  > > > >    
>  > > > > error in hdb file
>  > ./gromacs43A1p/aminoacid.hdb:
>  > > > >
>  > > > > wrong number of control atoms (2 iso3)
>  on
>  > line:
>  > > > > 2 3 N -C -CA
>  --------------------------------------------
>  On Tue, 9/2/16, gromacs.org_gmx-users-request at maillist.sys.kth.se
>  <gromacs.org_gmx-users-request at maillist.sys.kth.se>
>  wrote:
>  
>   Subject: gromacs.org_gmx-users Digest, Vol 142, Issue
> 42
>   To: gromacs.org_gmx-users at maillist.sys.kth.se
>   Received: Tuesday, 9 February, 2016, 1:20 PM
>   
>   Send gromacs.org_gmx-users mailing
>   list submissions to
>       gromacs.org_gmx-users at maillist.sys.kth.se
>   
>   To subscribe or unsubscribe via the World Wide Web,
> visit
>       https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or, via email, send a message with subject or body
> 'help'
>   to
>       gromacs.org_gmx-users-request at maillist.sys.kth.se
>   
>   You can reach the person managing the list at
>       gromacs.org_gmx-users-owner at maillist.sys.kth.se
>   
>   When replying, please edit your Subject line so it is
> more
>   specific
>   than "Re: Contents of gromacs.org_gmx-users
> digest..."
>   
>   
>   Today's Topics:
>   
>      1. Different results for same protein's
>   simulation in    4.6.5 and
>         5.0.4 (SAPNA BORAH)
>      2. Re: Different results for same
>   protein's simulation in 4.6.5
>         and 5.0.4 (Mark Abraham)
>      3. Re: creating representative structures
>   (Mark Abraham)
>      4. Re: pp/pme ratio differences between
>   gromacs 5.0.x and    5.1.x
>         (Mark Abraham)
>      5. Re: Fwd: FW: PME settings in free
>   energy calculations
>         (Mark Abraham)
>   
>   
>  
> ----------------------------------------------------------------------
>   
>   Message: 1
>   Date: Tue, 9 Feb 2016 12:52:41 +0530
>   From: SAPNA BORAH <sapnauser365 at gmail.com>
>   To: gmx-users at gromacs.org
>   Subject: [gmx-users] Different results for same
> protein's
>   simulation
>       in    4.6.5 and 5.0.4
>   Message-ID:
>       <CAEL2Jf7tNT94MnMbW3h6YnsKL63tC0+xeZQhMFnre7WVpngVCQ at mail.gmail.com>
>   Content-Type: text/plain; charset=UTF-8
>   
>   Dear all,
>   
>   I have tried running the same protein in the two
> versions
>  of
>   gromacs 4.6.5
>   and 5.0.4. Now, both the runs are producing entirely
>   different results.
>   
>   The results till equilibration are same.
>   
>   After the run however, there is a total change in
> the
>   production results. I
>   have used the same mdp files.
>   
>   I have rerun the systems and they still produce
> similar
>   results in the same
>   version.
>   
>   The protein structure is a modelled one.
>   
>   
>   Thanks in advance.
>   Sapna.
>   
>   Sapna Mayuri Borah
>   c/o Dr. A. N. Jha
>   Research student
>   Tezpur University,
>   India
>   
>   
>   ------------------------------
>   
>   Message: 2
>   Date: Tue, 09 Feb 2016 07:37:54 +0000
>   From: Mark Abraham <mark.j.abraham at gmail.com>
>   To: gmx-users at gromacs.org
>   Subject: Re: [gmx-users] Different results for same
>   protein's
>       simulation in 4.6.5 and 5.0.4
>   Message-ID:
>      
>   <CAMNuMARf1tJD+thVQvoSuN9SPkYhW06Oo2w6=bbFh6UdvFsyYw at mail.gmail.com>
>   Content-Type: text/plain; charset=UTF-8
>   
>   Hi,
>   
>   This is quite normal.
>  
> http://www.gromacs.org/Documentation/Terminology/Reproducibility.
>   You
>   should see a similar range of variation in the
>  trajectories
>   produced in the
>   two versions that are incompletely sampling the same
>   ensemble, but one can
>   say very little about two single observed
> trajectories
>   sampled from them.
>   
>   Mark
>   
>   On Tue, Feb 9, 2016 at 8:23 AM SAPNA BORAH <sapnauser365 at gmail.com>
>   wrote:
>   
>   > Dear all,
>   >
>   > I have tried running the same protein in the
> two
>   versions of gromacs 4.6.5
>   > and 5.0.4. Now, both the runs are producing
> entirely
>   different results.
>   >
>   > The results till equilibration are same.
>   >
>   > After the run however, there is a total change
> in the
>   production results. I
>   > have used the same mdp files.
>   >
>   > I have rerun the systems and they still produce
>  similar
>   results in the same
>   > version.
>   >
>   > The protein structure is a modelled one.
>   >
>   >
>   > Thanks in advance.
>   > Sapna.
>   >
>   > Sapna Mayuri Borah
>   > c/o Dr. A. N. Jha
>   > Research student
>   > Tezpur University,
>   > India
>   > --
>   > Gromacs Users mailing list
>   >
>   > * Please search the archive at
>   > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before
>   > posting!
>   >
>   > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   >
>   > * For (un)subscribe requests visit
>   > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or
>   > send a mail to gmx-users-request at gromacs.org.
>   >
>   
>   
>   ------------------------------
>   
>   Message: 3
>   Date: Tue, 09 Feb 2016 07:46:31 +0000
>   From: Mark Abraham <mark.j.abraham at gmail.com>
>   To: gmx-users at gromacs.org
>   Subject: Re: [gmx-users] creating representative
>  structures
>   Message-ID:
>       <CAMNuMAQgBnhazSinyeDzpLdwEZDuNSbDmZ77bLJhO5qkt7w06g at mail.gmail.com>
>   Content-Type: text/plain; charset=UTF-8
>   
>   Hi,
>   
>   On Mon, Feb 8, 2016 at 6:24 PM Shyno Mathew <sm3334 at columbia.edu>
>   wrote:
>   
>   > Dear all,
>   >
>   >
>   > I have few questions regarding creating
>  representative
>   structures. For
>   > simplicity, let?s say I have a trajectory of 5
>  frames:
>   >
>   > 1.       g_rmsf: After reading
>   previous posts, here is what I understood.
>   > The average structures calculated using g_rmsf
> (by
>   specifying ?ox) is
>   > literally the average of x, y, z co-ordinates of
> each
>   atom over all the 5
>   > frames in my case. Energy minimizing this
> averaged
>   structure might give a
>   > meaningful structure?
>   >
>   
>   It'll give a structure that conforms to chemical
>   expectations, but the
>   "meaning" of any plausible candidate for an
> "average"
>   structure is doubtful
>   at best. If there is a group of microstates that
> share
>   structural
>   similarity, it does not follow that RMSD will
> identify it,
>   nor that
>   averaging of coordinates will produce a meaningful
>  average.
>   What is the RMS
>   average position of a set of balls being juggled?
>   
>   
>   > 2.       g_cluster: Here I am
>   using the gromos method. I read the reference
>   > paper (Daura et al.). Using gromos method and by
> just
>   specifying ?cl (not
>   > ?av) I get the middle structures for each
> cluster.
>   Let?s say I get 2
>   > clusters using rmsd 0.16. Now the
> representative
>   structures of these two
>   > clusters (obtained in the out.pdb) should
> exactly
>   correspond to two frames
>   > in my original trajectory (the one with 5
> frames)?
>   >
>   
>   I don't know, but you should try it and see. Make a
>   trajectory of a water
>   molecule as PDB with a few frames and play with
> editing
>  the
>   coordinates by
>   hand and passing them to g_cluster until you are
> happy you
>   understand how
>   it works.
>   
>   
>   > 3.       My final question is
>   regarding how exactly g_cluster works, here
>   > is what I understand from Daura et al.
>   >
>   > If I use g_cluster with gromos method, the code
> will
>   look for neighbors of
>   > each frame within specified cutoff.
>   >
>   > Assume the first time the code finds the
> following:
>   >
>   > frame 0 has two neighbors: frames 2,3  within
>   cutoff
>   >
>   > frame 1 has three neighbors: frames 2, 3, 4
> within
>   cutoff
>   >
>   > frame 2 has four neighbors: frames 0, 1, 3, 4
> within
>   cutoff
>   >
>   > frame 3 has three neighbors: frames 0,1, 2
> within
>   cutoff
>   >
>   > frame 4 has two neighbors: frames 1,2 within
> cutoff
>   >
>   >
>   >
>   > Since frame 2 has the highest number of
> neighbors,
>  it's
>   considered the
>   > center cluster and this frame along with
> neighbors
>  are
>   removed. The same
>   > calculation is performed on the remaining frames
> if I
>   had more frames.
>   >
>   
>   No idea, but again try it and see.
>   
>   Mark
>   
>   
>   > Thanks in advance for your help,
>   >
>   > Sincerely,
>   >
>   > Shyno
>   >
>   > --
>   > Shyno Mathew
>   > PhD Candidate
>   > Department of Chemical Engineering
>   > Graduate Assistant
>   > Office of Graduate Student Affairs
>   > The Fu Foundation School Of Engineering and
> Applied
>   Science
>   > Columbia University
>   > --
>   > Gromacs Users mailing list
>   >
>   > * Please search the archive at
>   > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before
>   > posting!
>   >
>   > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   >
>   > * For (un)subscribe requests visit
>   > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or
>   > send a mail to gmx-users-request at gromacs.org.
>   
>   
>   ------------------------------
>   
>   Message: 4
>   Date: Tue, 09 Feb 2016 07:59:18 +0000
>   From: Mark Abraham <mark.j.abraham at gmail.com>
>   To: gmx-users at gromacs.org
>   Subject: Re: [gmx-users] pp/pme ratio differences
> between
>   gromacs
>       5.0.x and    5.1.x
>   Message-ID:
>       <CAMNuMAReBaEYCG6fok-3MG5w0p4NUWats93xQvY6ge3ngPpNhw at mail.gmail.com>
>   Content-Type: text/plain; charset=UTF-8
>   
>   Hi,
>   
>   On Sun, Feb 7, 2016 at 8:00 PM Johannes Wagner <johannes.wagner at h-its.org>
>   wrote:
>   
>   > hey guys,
>   > came across an issue with pp/pme ratio
> difference
>  from
>   5.0 to 5.1. For a
>   > system running on 2 nodes with a pp/pme ratio
> of
>  80:16
>   on 5.0, the same
>   > thing starts with 72:24 on gromacs 5.1. The 5.1
> run
>  has
>   a much higher load
>   > imbalance and is hence slower. Are there known
>  changes
>   from 5.0 to 5.1 in
>   > the pp/pme ratio that cause this?
>   >
>   
>   Yes and no. Even on only two nodes, your observations
> can
>  be
>   affected by
>   network congestion, so I would investigate whether
> you can
>   reproduce the
>   observations. The changes that I identified that had
> the
>   capacity to change
>   performance are noted at
>  
> http://manual.gromacs.org/documentation/5.1/ReleaseNotes/performance.html.
>   For example, if you're using LINCS then the SIMD
>   improvements there affect
>   only the PP ranks so it might seem to the auto-tuner
> that
>   it's more
>   efficient to use fewer PP ranks. Then later some kind
> of
>   load imbalance has
>   a stronger effect. The interactions of DD DLB and
> PME
>  tuning
>   are complex,
>   and we can't rule out that there's a problem that we
>  should
>   fix. But we
>   can't say much without being able to compare some
>   (reproducible) log files.
>   
>   Mark
>   
>   cheers, johannes
>   >
>   > --
>   > Gromacs Users mailing list
>   >
>   > * Please search the archive at
>   > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before
>   > posting!
>   >
>   > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   >
>   > * For (un)subscribe requests visit
>   > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or
>   > send a mail to gmx-users-request at gromacs.org.
>   >
>   
>   
>   ------------------------------
>   
>   Message: 5
>   Date: Tue, 09 Feb 2016 08:20:17 +0000
>   From: Mark Abraham <mark.j.abraham at gmail.com>
>   To: gmx-users at gromacs.org
>   Subject: Re: [gmx-users] Fwd: FW: PME settings in
> free
>   energy
>       calculations
>   Message-ID:
>       <CAMNuMAQ9oJbDrAuW9danqaXXmhTNw=zsJXZ9S-tDw9WaRsrsgg at mail.gmail.com>
>   Content-Type: text/plain; charset=UTF-8
>   
>   Hi,
>   
>   On Fri, Feb 5, 2016 at 1:32 PM Dries Van Rompaey
> <dries.vanrompaey at gmail.com>
>   wrote:
>   
>   > Hi,
>   >
>   > The corresponding mdp output section is:
>   >
>   > ; OPTIONS FOR ELECTROSTATICS AND VDW
>   > ; Method for doing electrostatics
>   > coulombtype           
>     = PME
>   > coulomb-modifier     
>      = Potential-shift-Verlet
>   > rcoulomb-switch          = 0
>   > rcoulomb           
>        = 1.0
>   > ; Relative dielectric constant for the medium
> and the
>   reaction field
>   > epsilon-r           
>       = 1
>   > epsilon-rf           
>      = 0
>   > ; Method for doing Van der Waals
>   > vdwtype             
>       = cutoff
>   > vdw-modifier         
>      = Potential-switch
>   > ; cut-off lengths
>   > rvdw-switch           
>     = 0.9
>   > rvdw             
>          = 1.0
>   > ; Apply long range dispersion corrections for
> Energy
>   and Pressure
>   > DispCorr           
>        = EnerPres
>   > ; Extension of the potential lookup tables
> beyond the
>   cut-off
>   > table-extension          = 1
>   > ; Separate tables between energy group pairs
>   > energygrp-table          =
>   > ; Spacing for the PME/PPPM FFT grid
>   > fourierspacing       
>      = 0.08
>   > ; FFT grid size, when a value is 0
> fourierspacing
>  will
>   be used
>   > fourier-nx           
>      = 0
>   > fourier-ny           
>      = 0
>   > fourier-nz           
>      = 0
>   > ; EWALD/PME/PPPM parameters
>   > pme_order           
>       = 6
>   > ewald_rtol           
>      = 1e-06
>   > ewald-rtol-lj           
>   = 0.001
>   > lj-pme-comb-rule     
>      = Geometric
>   > ewald-geometry       
>      = 3d
>   > epsilon_surface          = 0
>   
>   
>   That all seems reasonable.
>   
>   
>   > I evaluated the energy with:
>   >
>   > Verlet - settings above (with standard
>   verlet-buffer-tolerance)
>   > @ s0 legend "Coulomb-14"
>   > @ s1 legend "Coulomb (SR)"
>   > @ s2 legend "Coul. recip."
>   >     0.000000 
>   -2221.295898  -546205.250000  9271.518555
>   > Total -     539155.0273
>   >
>   > Verlet - settings above - with
>  verlet-buffer-tolerance
>   -1
>   >     0.000000 
>   -2221.295898  -546205.250000  9271.516602
>   > Total   -539155.029296
>   >
>   
>   That kind of agreement is expected, but perhaps
> might
>  agree
>   exactly if you
>   were following
>   http://www.gromacs.org/Documentation/How-tos/Single-Point_Energy
>   
>   Group - settings above
>   > @ s0 legend "Coulomb-14"
>   > @ s1 legend "Coulomb (SR)"
>   > @ s2 legend "Coul. recip."
>   >     0.000000 
>   -2221.296387  -487993.593750  -48931.437500
>   > Total -539146.3276
>   >
>   
>   The difference in those components is expected - in
> PME
>  you
>   have to avoid
>   counting contributions in the long-range part from
> pairs
>   that are excluded
>   e.g. from the bonded topology. The two schemes handle
> this
>   in different
>   places. Their sum should match (within the limits of
>   floating-point
>   arithmetic), and a relative error under 1e-5 for
> summing a
>   few million
>   numbers is acceptable for judging the equivalence of
> the
>   implementations.
>   
>   Verlet - 3 nm cutoffs, 2.9 vdwswitch,
> coulomb-modifier
>  none
>   > @ s0 legend "Coulomb-14"
>   > @ s1 legend "Coulomb (SR)"
>   >     0.000000 
>   -2221.295898  -774817.875000
>   > Total -777039.1709
>   >
>   >
>   > Group - 3 nm cutoffs, 2.9 vdwswitch,
> coulomb-modifier
>   none
>   > @ s0 legend "Coulomb-14"
>   > @ s1 legend "Coulomb (SR)"
>   >     0.000000 
>   -2221.296387  -553398.125000
>   > Total -555619.4214
>   >
>   
>   That ought to agree much better, but It's hard to
>  speculate
>   on where the
>   problem lies without seeing log files. Please upload
> some
>  to
>   a file sharing
>   service and post the links here (list doesn't take
>   attachments)
>   
>   Mark
>   
>   
>   > Thanks,
>   >
>   > Dries
>   >
>   > On 4 February 2016 at 14:48, Michael Shirts
> <mrshirts at gmail.com>
>   wrote:
>   >
>   > > Also, can you be more precise about the
>   inconsistency in the outputs
>   > > you are seeing (with numerical output from
>  GROMACS
>   and analysis code).
>   > >
>   > > On Thu, Feb 4, 2016 at 6:17 AM, Michael
> Shirts
>   <mrshirts at gmail.com>
>   > wrote:
>   > > > Hi, Dries-
>   > > >
>   > > > Can you print out what the
> corresponding
>   section of mdout.mdp look
>   > like?
>   > > >
>   > > > Have you  tried using the cutoffs
> above
>   with group cutoffs, and seen
>   > > > what the difference is?
>   > > >
>   > > > On Thu, Feb 4, 2016 at 4:59 AM, Dries
> Van
>   Rompaey
>   > > > <dries.vanrompaey at gmail.com>
>   wrote:
>   > > >> Hi,
>   > > >>
>   > > >> I'm using PME with the default
>   Potential-Shift-Verlet modifier. If I'm
>   > > >> interpreting the manual and
> comments on
>   the mailinglist correctly, the
>   > > >> settings for rcoulomb-switch
> shouldn't
>   affect this and PME-switching
>   > > >> shouldn't be necessary (although
> there
>   seems to have been some
>   > > discussion
>   > > >> about this, https://gerrit.gromacs.org/#/c/2220/). Has a consensus
>   > been
>   > > >> reached elsewhere?
>   > > >>
>   > > >> Current nonbonded settings in my
> .mdp
>   files are:
>   > > >> ; Electrostatics
>   > > >> coulombtype       
>         = PME
>   > > >> rcoulomb       
>               = 1.0
>   > > >>
>   > > >> ; van der Waals
>   > > >> vdwtype         
>             = cutoff
>   > > >> vdw-modifier       
>       = Potential-switch
>   > > >> rvdw-switch       
>         = 0.9
>   > > >> rvdw         
>              
>      = 1.0
>   > > >>
>   > > >> ; Apply long range dispersion
>  corrections
>   for Energy and Pressure
>   > > >> DispCorr       
>             = EnerPres
>   > > >>
>   > > >> ; Spacing for the PME/PPPM FFT
> grid
>   > > >> fourierspacing       
>      = 0.08
>   > > >>
>   > > >> ; EWALD/PME/PPPM parameters
>   > > >> pme_order       
>           = 6
>   > > >> ewald_rtol       
>            = 1e-06
>   > > >> epsilon_surface     
>     = 0
>   > > >>
>   > > >> Thanks in advance,
>   > > >>
>   > > >> Dries
>   > > >>
>   > > >> On 4 February 2016 at 00:56,
> Michael
>   Shirts <mrshirts at gmail.com>
>   > wrote:
>   > > >>
>   > > >>> Hi, Dries-
>   > > >>>
>   > > >>> Questions like this are
> probably
>  best
>   answered on the gmx-users list.
>   > > >>> I can't say too much for the
> Verlet
>   scheme -- I know that it was
>   > > >>> relatively recently adapted
> for
>  free
>   energies, and there may be some
>   > > >>> combinations of settings that
> could
>   give unanticipated results.
>   > > >>>
>   > > >>> Pretty much all of our
> experience
>   about nonbonded calculations and
>   > > >>> free energies is collected in
> the
>   following paper:
>   > > >>> http://pubs.acs.org/doi/abs/10.1021/ct4005068. We used
>   the group
>   > > >>> cutoff scheme since that was
> the
>  only
>   one that was supported in
>   > > >>> GROMACS at the time.
>   > > >>>
>   > > >>> Since you haven't sent any
> files,
>   it's hard to tell what is actually
>   > > >>> going on. The one thing that
> has a
>   tendency to happen with the more
>   > > >>> recent update schemes is if
> you set
>  a
>   potential modifier that is a
>   > > >>> switch, but don't set the
> distance
>   the switch starts at, then it is
>   > > >>> automatically set to zero. 
>   Check the mdout.mdp to see if this is
>   > > >>> happening.
>   > > >>>
>   > > >>> > From: Van Rompaey Dries
> <Dries.VanRompaey at uantwerpen.be>
>   > > >>> Date: Wednesday, February 3,
> 2016
>  at
>   12:34 PM
>   > > >>> To: Michael Shirts <michael.shirts at colorado.edu>
>   > > >>> Subject: PME settings in free
>  energy
>   calculations
>   > > >>>
>   > > >>> Dear prof. Shirts,
>   > > >>>
>   > > >>> I'm currently trying to figure
> out
>   the PME settings for a relative
>   > > >>> free binding energy simulation
> I'm
>   working on. I took the parameters
>   > > >>> from Matteo Aldeghi's
>   > > >>> paper(
>   > > >>>
>   > >
>   > http://pubs.rsc.org/en/content/articlelanding/2015/sc/c5sc02678d#!divAbstract
>   > > >>> )
>   > > >>> as a starting point (adapted
> for
>  the
>   Verlet scheme by scaling the
>   > > >>> fourierspacing to 0.08 and
> setting
>   coulomb = rvdw = 1.0). I then
>   > tried
>   > > >>> to verify these settings by
>  comparing
>   a single point energy
>   > > >>> calculation with these
> settings and
>   one with very long coulomb
>   > cutoffs
>   > > >>> as recommended on
> alchemistry.org.
>   > > >>>
>   > > >>> Unfortunately, I can't seem to
> get
>   this quite right. I'm getting
>   > > >>> differences in the hundreds
> of
>   kj/mol, leading me to suspect I'm
>   > doing
>   > > >>> something wrong. I'm
> calculating
>  the
>   energy values by extracting
>   > > >>> CoulombSR and Coul-recip from
> the
>   energy.xvg files. I've tried
>   > > >>> calculating the energy with
> coulomb
>   cutoffs at 3 nm and 10 nm, but
>   > > >>> agreement with the PME
> results
>   remains rather poor. David Mobley
>   > > >>> mentioned you performed
> extensive
>   research on this topic, and I'm
>   > > >>> hoping you could point me in
> the
>   right direction.
>   > > >>>
>   > > >>> Thanks in advance,
>   > > >>>
>   > > >>> Dries
>   > > >>>
>   > > >>>
>   > > >>>
>   > > >>> ~~~~~~~~~~~~~~~~
>   > > >>> Michael Shirts
>   > > >>> Associate Professor
>   > > >>> michael.shirts at colorado.edu
>   > > >>> Phone: (303) 735-7860
>   > > >>> Office: JSCBB D317
>   > > >>> Department of Chemical and
>  Biological
>   Engineering
>   > > >>> University of Colorado
> Boulder
>   > > >>> --
>   > > >>> Gromacs Users mailing list
>   > > >>>
>   > > >>> * Please search the archive
> at
>   > > >>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before
>   > > >>> posting!
>   > > >>>
>   > > >>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   > > >>>
>   > > >>> * For (un)subscribe requests
> visit
>   > > >>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   > or
>   > > >>> send a mail to gmx-users-request at gromacs.org.
>   > > >>>
>   > > >> --
>   > > >> Gromacs Users mailing list
>   > > >>
>   > > >> * Please search the archive at
>   > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before
>   > > posting!
>   > > >>
>   > > >> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   > > >>
>   > > >> * For (un)subscribe requests
> visit
>   > > >> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or
>   > > send a mail to gmx-users-request at gromacs.org.
>   > > --
>   > > Gromacs Users mailing list
>   > >
>   > > * Please search the archive at
>   > > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before
>   > > posting!
>   > >
>   > > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   > >
>   > > * For (un)subscribe requests visit
>   > > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or
>   > > send a mail to gmx-users-request at gromacs.org.
>   > >
>   > --
>   > Gromacs Users mailing list
>   >
>   > * Please search the archive at
>   > http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before
>   > posting!
>   >
>   > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   >
>   > * For (un)subscribe requests visit
>   > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or
>   > send a mail to gmx-users-request at gromacs.org.
>   >
>   
>   
>   ------------------------------
>   
>   -- 
>   Gromacs Users mailing list
>   
>   * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List
>   before posting!
>   
>   * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>   
>   * For (un)subscribe requests visit
>   https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users
>   or send a mail to gmx-users-request at gromacs.org.
>   
>   End of gromacs.org_gmx-users Digest, Vol 142, Issue
> 42
>  
> ******************************************************


More information about the gromacs.org_gmx-users mailing list